17.11.2012 Views

Menp. ECO. MED. 2 - Ecologia Mediterranea

Menp. ECO. MED. 2 - Ecologia Mediterranea

Menp. ECO. MED. 2 - Ecologia Mediterranea

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Tome 28<br />

Fascicule 2, 2002<br />

ISSN 0153-8756<br />

Revue internationale<br />

d’écologie méditerranéenne<br />

International Journal<br />

of <strong>Mediterranea</strong>n Ecology


ecologia<br />

mediterranea<br />

Revue internationale<br />

d’écologie méditerranéenne<br />

International Journal<br />

of <strong>Mediterranea</strong>n Ecology<br />

Tome 28 • Fascicule 2 • 2002


Rédacteur en chef • Managing editor Secrétariat • Secretariat<br />

FRÉDÉRIC MÉDAIL MICHELLE DOUGNY<br />

ARONSON J., CEFE-CNRS, Montpellier<br />

BARBERO M., IMEP, Université Aix-Marseille III<br />

BEAULIEU J.-L. DE, IMEP, Université Aix-Marseille III<br />

BROCK M., University of New England, Armidale, Australie<br />

CHEYLAN M., EPHE, Montpellier<br />

DEBUSSCHE M., CEFE-CNRS, Montpellier<br />

FADY B., INRA, Avignon<br />

GRILLAS P., Station biologique Tour du Valat, Arles<br />

GUIOT J., CEREGE-CNRS, Aix-en-Provence<br />

HOBBS R. J., CSIRO, Midland, Australie<br />

KREITER S., ENSA-M-INRA, Montpellier<br />

LE FLOC’H E., CEFE-CNRS, Montpellier<br />

Rédacteurs • Editors<br />

LAURENCE AFFRE PHILIP ROCHE<br />

THIERRY DUTOIT THIERRY TATONI<br />

JÉRÔME ORGEAS ERIC VIDAL<br />

Fondateur • Founder<br />

PROFESSEUR PIERRE QUÉZEL<br />

Comité de lecture • Advisory board<br />

<strong>Ecologia</strong> mediterranea<br />

MARGARIS N. S., University of the Aegean, Mytilène, Grèce<br />

OVALLE C., CSI-Quilamapu, INIA, Chili<br />

PEDROTTI F.,Universita degli Studi, Camerino, Italie<br />

PLEGUEZUELOS J. M., Université de Grenade, Espagne<br />

PONEL P., IMEP, CNRS, Marseille<br />

PRODON R., Lab. Arago, Université P. & M. Curie, Paris VI<br />

RIDCHARSON D. M., University Cape Town, Afrique du Sud<br />

SANS F. X., Université de Barcelone, Espagne<br />

SHMIDA A., Hebrew University of Jérusalem, Israël<br />

Abonnements • Subscription (contacter Edisud)<br />

Un an : 2 numéros • One year : 2 issues<br />

— France : 61 € + 9,12 € de frais de port<br />

— Europe : 61 € + 12,2 € de frais de port<br />

— Amérique, Afrique, Asie : 61 € + 18,3 € de frais de port<br />

TROUMBIS A., University of the Aegean Mytilene, Grèce<br />

URBINATI C., Agripolis, Legnaro, Italie<br />

Faculté des sciences et techniques de Saint-Jérôme<br />

Institut méditerranéen d’écologie et de paléoécologie, case 461<br />

F-13997 Marseille, cédex 20, France<br />

Tél. : + 33 04 91 28 85 35 – Fax : + 33 04 91 28 80 51<br />

email : f.medail@univ.u-3mrs.fr – URL : http://www.ecologia.fst.u-3mrs.fr<br />

Éditions Édisud<br />

La Calade, 3120 route d’Avignon, 13090 Aix-en-Provence<br />

Tél. : 00 33 04 42 21 61 44 – Fax : 00 33 04 42 21 56 20<br />

email : http://www.edisud.com – Internet : info@edisud.com<br />

© Édisud, 2002, tous droits réservés.<br />

ISSN 0153-8756


ecologia<br />

mediterranea<br />

Revue internationale<br />

d’écologie méditerranéenne<br />

International Journal<br />

of <strong>Mediterranea</strong>n Ecology<br />

Tome 28 • Fascicule 2 • 2002


Belowground traits of mediterranean woody plants<br />

in a Portuguese shrubland<br />

Caractéristiques du système racinaire de ligneux méditerranéens<br />

dans un maquis du Portugal<br />

Joaquim S. Silva 1,Francisco C. Rego 2, Maria A. Martins-Loução 3<br />

1. Escola Superior Agrária de Coimbra, 3040-316 Coimbra, Portugal<br />

2. Instituto Superior de Agronomia, Tapada da Ajuda, 1349-017 Lisboa, Portugal<br />

3. Centro de <strong>Ecologia</strong> e Biologia Vegetal, Faculdade de Ciências de Lisboa, 1749-016 Lisboa, Portugal.<br />

Fax :+ 351 239802979 ; tél. : + 351 802284. email: jss@mail.esac.pt<br />

Abstract<br />

Belowground traits vary widely. Apart from the influence of the<br />

environment both genetic and ontogenic factors are responsible for<br />

this variation. For mediterranean woody plants there is also evidence<br />

of a relationship between regenerative strategies and root system<br />

characteristics. With the general aim of studying these different<br />

aspects, the root systems of seventeen obligate seeders and sixteen<br />

resprouters from ten different species and different developmental<br />

stages were excavated at Tapada Nacional de Mafra in<br />

Central West Portugal. Root systems were photographed, weighted<br />

and measured. Root length and the average root diameter were<br />

determined using digital image software. Root-to-shoot ratio (R/S)<br />

and the specific root length (SRL) were computed for all plants.<br />

Basal section was used as an indicator of plant development. A<br />

principal component analysis (PCA) was performed in order to<br />

study the relationships between variables and between plants. The<br />

analysis showed a clear distinction of plants according to the respective<br />

developmental stage but also according to the regenerative<br />

characteristics of the different species. Allometric relationships were<br />

found between root biomass, shoot biomass and basal section.<br />

Statistical tests showed that resprouters had higher maximum rooting<br />

depth, average root diameter and R/S and lower SRL, than<br />

obligate seeders. A decrease of R/S and SRL with basal section was<br />

verified for a sub-sample of four species.<br />

Key-words<br />

Root development, mediterranean shrublands, regenerative strategies,<br />

root systems, allometric relationships.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 5-13<br />

Résumé<br />

Il existe une grande variabilité associée aux caractéristiques souterraines<br />

des plantes. Outre l’influence de l’environnement, les responsables<br />

en sont les facteurs génétiques et les facteurs ontogéniques.<br />

Dans le cas des plantes ligneuses méditerranéennes, il existe aussi<br />

un rapport entre les stratégies régénératives et les caractéristiques des<br />

systèmes racinaires. En envisageant l’étude de ces différents aspects,<br />

les systèmes racinaires de dix-sept plantes se régénérant par semis et<br />

seize plantes à régénération végétative appartenant à dix espèces<br />

différentes à divers stades de développement, ont été déterrés à<br />

Tapada Nacional de Mafra, dans le Centre-Ouest du Portugal. Les<br />

racines ont été photographiées, pesées et mesurées. La longueur et le<br />

diamètre moyen ont été déterminés en utilisant un programme de<br />

mesure à partir d’images numériques. Le rapport entre la biomasse<br />

des racines et celle de la partie aérienne ainsi que la longueur spécifique<br />

des racines ont été calculés pour toutes les plantes. La section<br />

basale a été utilisée comme un indicateur du développement<br />

de la plante. Une analyse par composantes principales a été utilisée<br />

pour étudier les rapports entre variables et entre plantes. L’analyse<br />

a montré une nette différence des plantes selon leur stade de développement,<br />

mais aussi en fonction des caractéristiques régénératives<br />

des différentes espèces. Des rapports allométriques ont été trouvés<br />

entre la biomasse des racines, la biomasse de la partie aérienne et<br />

la section basale. Des tests statistiques ont montré que les plantes<br />

de régénération végétative présentent un enracinement plus profond,<br />

un diamètre moyen plus important, un plus grand rapport<br />

biomasse racinaire/biomasse aérienne et une longueur racinaire spécifique<br />

plus faible que les plantes de régénération par semis. Une<br />

diminution du rapport biomasse racinaire/biomasse aérienne et de<br />

la longueur racinaire spécifique avec la section basale a été vérifiée<br />

chez un sous-échantillon de quatre espèces.<br />

Mots-clés<br />

Développement racinaire, maquis méditerranéen, stratégies régénératives,<br />

systèmes racinaires, rapports allométriques.<br />

5


6<br />

◆ JOAQUIM S. SILVA, FRANCISCO C. REGO, MARIA A. MARTINS-LOUÇÃO<br />

INTRODUCTION<br />

The origins of the diversity of root systems can be seen<br />

as an optimisation of two primary functions: acquisition<br />

of soil-based resources (water and nutrients) and anchorage<br />

(Fitter, 1996). In the specific case of the <strong>Mediterranea</strong>n<br />

regions of the world, root systems have probably<br />

evolved to deal with the strong spatial and temporal limitations<br />

on availability of water, which are typical of these<br />

regions (Canadell & Zedler, 1995). This evolution has<br />

resulted in various root type morphologies, characteristic<br />

of mediterranean woody species. Several attempts were<br />

made to classify root systems in terms of basic structural<br />

characteristics (Cannon, 1949; Hellmers et al., 1955;<br />

Specht & Rayson, 1957). However the well known plasticity<br />

of root systems as a result of environmental conditions,<br />

has limited the use of these classifications (Fitter,<br />

1996). Bengough et al. (2000) suggested the study of<br />

variability as an alternative to seeking for homogeneous<br />

features. A simple classification of plants into deep rooted<br />

and shallow rooted is generally accepted, although no<br />

strict limits are normally established to define the two<br />

types. These two basic types of root systems correspond<br />

to different adaptations to the highly seasonal water availability<br />

typical of <strong>Mediterranea</strong>n-type climates and have<br />

been associated with the strategy to regenerate after natural<br />

disturbances (fire and grazing). Typical resprouters<br />

normally present deep root systems whereas seeders are<br />

shallow rooted (Keeley, 1986; Correia & Catarino, 1994;<br />

Bell, 2001). In fact, resprouters need to have deep root<br />

systems to supply the growth of new shoots since they<br />

can not rely on a seed bank to regenerate after fire or other<br />

kind of disturbance (Keeley & Zedler, 1978; Clemente et<br />

al., 1996). Although sharing the same climatic conditions<br />

as resprouters, obligate seeders present specific water<br />

saving adaptations such as a higher stomatal control or<br />

leaf hairs, allowing these species to survive in drier conditions<br />

(Keeley, 1986; Correia, 1988; Correia & Catarino;<br />

1994; Silva & Rego, 1998). These water saving mechanisms<br />

partly explain why obligate seeders withstand summer<br />

drought without the use of deep root systems.<br />

The study of root systems have also showed the existence<br />

of different patterns of root development. In the<br />

case of many mediterranean plants it is known the preferential<br />

allocation of resources to roots at early stages<br />

(Canadell & Zedler, 1995) and there is evidence that<br />

woody plants in general present a decreasing trend of rootto-shoot<br />

ratio with age (Kozlowski et al., 1991).The fact<br />

that many species, especially those under dry conditions,<br />

may develop deep tap roots at early stages preceding the<br />

development of lateral roots (Spurr & Barnes, 1980) is<br />

an evidence of the specificity of rooting patterns at different<br />

developmental stages. Ontogeny seems then to be at<br />

least as important as phylogeny to the overall variation of<br />

root systems and the respective traits.<br />

It is an obvious fact that root systems are difficult to<br />

study.These difficulties of root studies increase in the case<br />

of plants from mediterranean ecosystems because soils are<br />

frequently shallow and heterogeneous and also because<br />

many species are growing deep root systems (Kummerow,<br />

1981). Additional problems are found with plants from<br />

dense shrub communities because different individuals<br />

form an intricate root net which makes it extremely difficult<br />

to trace individual root systems. Consequently there<br />

is a considerable lack of knowledge of basic root system<br />

characteristics of mediterranean woody plants, and no<br />

information at all could be found concerning the belowground<br />

traits of the species studied in the present work.<br />

The relative scarcity of root studies in mediterranean ecosystems<br />

is well reflected in the planetary compilation of<br />

root distribution data by Jackson et al. (1996) where the<br />

<strong>Mediterranea</strong>n Region is represented by only 4 studies out<br />

of 250. However it is generally recognised the importance<br />

of obtaining information on root systems for modelling<br />

the functions of ecosystems both at the plant as at the community<br />

level (Caldwell & Richards, 1986; Pagés, 2000) or<br />

even in global scale simulation models (Zeng, 2001).<br />

The general objective of this paper is to assess the existence<br />

of relationships concerning the belowground traits<br />

of mediterranean woody plants. In particular the present<br />

paper is focused on: a) the relationships between different<br />

belowground traits; b) the relationships between plant<br />

development and belowground traits; c) the relationships<br />

between regenerative strategies and belowground traits.<br />

MATERIALS AND METHODS<br />

Plants were collected at Tapada Nacional de Mafra,<br />

a public estate located in the Central West region of<br />

Portugal.Tapada Nacional de Mafra, is a protected area<br />

with 827 ha, about 30 km Northwest of Lisbon and 12<br />

km East from the coast (38º 56’ 37’’ to 38º 58’ 30’’ N<br />

and 9º 15’ 52’’ to 9º 18’ 43’’ W). The lowest altitude is<br />

90 m and the highest is 358 m. Soil is a sandy loam classified<br />

as a humic cambisol (FAO classification) derived<br />

from sandstone. Bedrock is normally located below 2 m.<br />

Mean annual precipitation is 798 mm and mean annual<br />

temperature is 14.6 ºC. Summer precipitation (June,<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


July, August) accounts for only 3.1 % of the total annual<br />

rainfall. Most of the area is constituted by shrublands<br />

dominated by Erica scoparia L. and Erica lusitanica<br />

Rudolphi.<br />

Between September 2000 and April 2001 the complete<br />

root systems of 33 plants from ten woody species<br />

were hydraulically excavated (Böhm, 1979). Although we<br />

have tried to include a wide variety of woody species, our<br />

choice was limited by the floristic composition of the study<br />

area.Within these limits plants were chosen according to<br />

their developmental stage and regenerative characteristics.<br />

From the ensemble of plants, seventeen were obligate<br />

seeders (Lavandula luisieri (Rozeira) Rivas Martinez<br />

(4 plants), Cistus crispus L. (9 plants) and Cistus salviifolius<br />

L. (4 plants) and sixteen were resprouters (Erica scoparia<br />

L. (2 plants), Erica lusitanica Rudolphi (1 plant),<br />

Crataegus monogyna Jacq. (4 plants), Ulex jussiaei Webb<br />

(3 plants), Daphne gnidium L. (4 plants), Pistacia lentiscus<br />

L. (1 plant) and Myrtus communis L.(1 plant)). In<br />

order to obtain four individuals representative of four<br />

classes of basal section, 2 obligate seeders (L. luisieri and<br />

C. crispus) and 2 resprouters (D. gnidium and C. monogyna)<br />

were sampled more intensively, constituting a pooled<br />

sub-sample of sixteen plants.These four species apparently<br />

represented different root system types, within the<br />

obligate seeder and resprouter strategies. Basal section<br />

was used as an indicator of the plants developmental stage.<br />

The four classes of basal section were defined as: class 1<br />

for plants up to 5 mm 2, class 2 from 5 to 25 mm 2, class<br />

3 from 25 to 125 mm 2 and class 4 for plants showing a<br />

basal section higher than 125 mm 2. Despite having sampled<br />

nine C. crispus plants, no plants were collected at<br />

class 4, given the small size of the individuals present at<br />

the excavation site. No replications could be obtained for<br />

each combination species/basal section class due to the<br />

enormous amount of work required for the excavation<br />

and processing of complete root systems. All plants were<br />

excavated in spots where soil depth did not seem to limit<br />

the vertical development of roots, thus allowing the full<br />

expression of potential growth of deep roots. Besides<br />

having observed the plants used in this study, we were<br />

able to observe in the field other partially excavated individuals<br />

from the same species. This allowed to recognise<br />

some basic morphological characteristics, enabling a more<br />

complete description of root systems.<br />

After excavation, the maximum rooting depth, the<br />

maximum average root width and the basal section of<br />

each plant were measured. Maximum rooting depth represents<br />

the depth achieved by the deepest root, maximum<br />

average root width was computed as the average between<br />

ecologia mediterranea, tome 28, fascicule 1, 2002<br />

BELOWGROUND TRAITS OF <strong>MED</strong>ITERRANEAN WOODY PLANTS IN A PORTUGUESE SHRUBLAND ◆<br />

the maximum horizontal width and the respective orthogonal<br />

width and basal section was obtained as the cross<br />

sectional area at the stem base. In the case of plants with<br />

several stems, basal section was obtained by summing all<br />

the stem sections.<br />

In order to obtain the total root length and the average<br />

root diameter, root systems were photographed using a<br />

Fujifilm MX 2900 Zoom (Fuji Photo Film Co., Ltd.,<br />

Tokyo) digital camera, featuring a maximum resolution<br />

of 2.3 millions of pixels. Digital images were analysed<br />

using software WinRhizo 4.1b (Regent Instruments,<br />

Quebec). Some remarks have to be made on the determination<br />

of root length. Root length is difficult to measure<br />

when dealing with extensive root systems. Since most<br />

of the root length is associated with fine roots, it is extremely<br />

difficult to have an accurate estimate of total root<br />

length from shrubs or trees collected in the field.The main<br />

source of errors is the excavation process (Böhm, 1979;<br />

Caldwell & Virginia, 1989), where considerable amounts<br />

of fine roots are inevitably lost.Therefore root length was<br />

determined for comparison purposes only and it should<br />

be strictly interpreted on a relative basis and not as absolute<br />

values.<br />

For biomass determination plants were oven-dried at<br />

85 ºC for 48 hours and weighted, separately for root and<br />

shoot fractions. In addition to these variables the specific<br />

root length (SRL; cm of total root length/g of root dry<br />

weight) and the root-to-shoot ratio (R/S; root dry<br />

weight/shoot dry weight) were computed for each plant.<br />

All root variables (including R/S ratio) were log transformed<br />

and used to perform a principal components analysis<br />

(PCA) for all plants. The principal components<br />

(PC’s) extracted by the analysis were related to the different<br />

individual plants collected, in order to evaluate which<br />

were the main variables responsible for the variation<br />

found and their relationship with species strategies and<br />

plant development. Allometric relationships between<br />

variables were established considering the results obtained<br />

from the PCA.<br />

Given that most variables did not follow a normal distribution<br />

(Shapiro Wilk W test), the Mann-Whitney U test<br />

was used to determine the significance of differences between<br />

obligate seeders and resprouters, for all variables.<br />

Tests were performed both with pooled samples of obligate<br />

seeders and resprouters and within developmental<br />

stages 2 and 3 (not enough plants within stages 1 and 4).<br />

The sub-sample of 16 plants referred above was used<br />

to assess the effect of the developmental stage on R/S and<br />

SRL for the different species.<br />

Species nomenclature followed Castroviejo (1999).<br />

7


8<br />

◆ JOAQUIM S. SILVA, FRANCISCO C. REGO, MARIA A. MARTINS-LOUÇÃO<br />

RESULTS<br />

All C.monogyna plants presented thick, structural roots,<br />

with a smooth pale brownish bark. L. luisieri was characterised<br />

by a fibrous root system composed of an intricate<br />

horizontal network of very thin pale roots. Despite having<br />

only collected young U. jussiaei plants, our observation of<br />

partially excavated roots from adult individuals revealed<br />

that this species presented deep, thick, poorly lignified tap<br />

roots, showing a wrinkled white bark. C. salviifolius and<br />

C.crispus plants showed very similar characteristics. Roots<br />

of both species were relatively simple in terms of branching,<br />

with few dark brown structural roots exploring surface<br />

soil layers.The two Erica species were also very similar<br />

in terms of root system characteristics. Both presented<br />

a lignotuber to which a few deep, black, strongly lignified,<br />

tap roots were connected. Both species were also<br />

showing an intricate network of horizontal, not very<br />

widelly spreading fine roots exploring surface soil layers.<br />

D. gnidium plants showed a rooting pattern similar to the<br />

Erica plants with reddish, deep, tap roots and with only<br />

a few horizontal roots. Root tissues were poorly lignified<br />

which made coarse roots soft and rubber-like, similar to<br />

some succulent shrubs. All D. gnidium plants showed a<br />

swelling region, similar to typical lignotubers, just below<br />

the stem base. M. communis was characterised by the presence<br />

of coarse, pale brown laterally spreading roots,<br />

connected to a main deep reaching root. Given the impossibility<br />

of observing more than one root system in the field<br />

we did not retain the morphological characteristics of<br />

P. lentiscus.<br />

A quantitative description of all plants grouped by species<br />

is presented in table 1. Considering all excavated<br />

plants, root biomass ranged from 111.8 g (E. scoparia) to<br />

0.2 g (C. crispus) and shoot biomass ranged from 141.8 g<br />

(E.scoparia) to 0.3 g (C. crispus). Maximum rooting depth<br />

ranged from 185 cm (D. gnidium) to 12 cm (C. crispus).<br />

Other species, E. lusitanica and E. scoparia, also showed<br />

deep roots (160 cm and 140 cm, respectively). Root system<br />

width ranged from 95 cm (L. luisieri) to 5 cm<br />

(C. monogyna). Total measured root length ranged from<br />

2791.0 cm (L. luisieri) to 65.7 cm (C. monogyna). Another<br />

C. monogyna plant had the second highest length of roots<br />

(1956.8 cm). The average root diameter ranged from<br />

3.6 mm (C. monogyna) to 0.7 mm (C. crispus). R/S ratio<br />

ranged from 3.2 (D. gnidium) to 0.2 (L. luisieri).The three<br />

highest values of R/S ratio were from D. gnidium and the<br />

three lowest were from L. luisieri. Only seven plants pre-<br />

Species n Reg. Basal sect. S. biom. Maxim. root. Root system Root length Aver. root R. biom R/S SRL<br />

strat. (mm2) (g) depth (cm) width (cm) (cm) diam. (mm) (g) (g.g-1) (cm.g-1) C. crispus 9 s 16.7 ± 3.6 2.6 ± 1.0 30.2 ± 5.0 19.1 ± 3.6 302.5 ± 66.1 1.0 ± 0.1 1.4 ± 0.4 0.7 ± 0.1 322.8 ± 88.7<br />

C. salvifolius 4 s 17.0 ± 10.7 3.1 ± 0.9 26.5.0 ± 5.0 15.6 ± 2.2 165.2 ± 29.8 1.4 ± 0.2 1.9 ± 0.9 0.5 ± 0.2 162.6 ± 56.2<br />

C. monogyna 4 r 86.4 ± 51.8 25.3 ± 16.9 45.3 ± 11.9 34.3 ± 15.2 610.7 ± 452.8 2.3 ± 0.3 19.7 ± 12.8 0.9 ± 0.1 51.3 ± 24.4<br />

D. gnidium 4 r 101.8 ± 82.3 31.2 ± 28.5 120.0 ± 36.6 31.8 ± 6.4 652.9 ± 225.7 2.1 ± 0.3 18.3 ± 13.8 1.8 ± 0.6 109.6 ± 41.0<br />

E. lusitanica 1 r 152.1 ± nd 43.1 ± nd 160.0± nd 48.5 ± nd 1738.6 ± nd 1.5 ± nd 36.7 ± nd 0.9 ± nd 47.4 ± nd<br />

E. scoparia 2 r 94.7 ± 78.4 72.9 ± 68.9 90.0 ± 50.0 25.8 ± 11.8 1253.6 ± 494.4 1.5 ± 0.6 57.0 ± 54.8 0.7 ± 0.1 179.6 ± 163.9<br />

L. luisieri 4 s 66.4 ± 51.0 27.6 ± 23.9 28.3 ± 2.3 50.1 ± 16.9 1077.7 ± 585.9 1.2 ± 0.2 5.1 ± 4.1 0.3 ± 0.1 475.1 ± 120.2<br />

M. communis 1 r 41.4 ± nd 8.3 ± nd 120.0 ± nd 73.5 ± nd 676.0 ± nd 2.0 ± nd 11.0 ± nd 1.3 ± nd 61.4 ± nd<br />

P. lentiscus 1 r 19.4 ± nd 2.7 ± nd 50.0 ± nd 20.0 ± nd 393.5 ± nd 1.5 ± nd 2.2 ± nd 0.8 ± nd 176.4 ± nd<br />

U. jussiaei 3 r 11.5 ± 2.9 3.3 ± 1.7 67.7 ± 28.7 14.0 ± 4.5 562.9 ± 212.4 1.1 ± 0.2 2.6 ± 1.5 0.8 ± 0.1 284.3 ± 57.2<br />

Table 1. Descriptive parameters (mean ± SE) of the root systems of 33 plants excavated at Tapada Nacional de Mafra, distributed by species. Legend for abbreviations:<br />

n – number of plants; Reg. strat. – regenerative strategy; Basal sect. – basal section; S. biom. – shoot biomass; Maxim. root. depth – maximum rooting depth; Aver. root<br />

diam. – average root diameter; R. biom. – root biomass; R/S – root-to-shoot ratio; SRL – specific root length; s – obligate seeder; r – resprouter.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


BELOWGROUND TRAITS OF <strong>MED</strong>ITERRANEAN WOODY PLANTS IN A PORTUGUESE SHRUBLAND ◆<br />

sented values of R/S higher than 1. SRL ranged from<br />

848.4 cm.g -1 (C. crispus) to 15.6 cm.g -1 (E. scoparia).The<br />

six highest values of SRL were from three C. crispus plants<br />

and three L. luisieri plants.<br />

The first two principal components extracted by the<br />

PCA (figure 1A) explained 80 % of the total variance.<br />

Root length and root width were best correlated with<br />

Figure 1. PCA diagrams.A represents the components loadings for each variable<br />

and B represents the components scores for each plant individual. Legend for<br />

variables: DIAMETER – Average root diameter; BIOMASS – Root biomass;<br />

DEPTH – Maximum rooting depth;WIDTH – Root system width; LENGTH<br />

– Root length; R/S – Root-to-shoot ratio; SRL – Specific root length. Legend for<br />

species: Cc – Cistus crispus; Cs – Cistus salviifolius; Cm – Crataegus monogyna;<br />

Dg – Daphne gnidium; El – Erica lusitanica; Es – Erica scoparia; Ll –<br />

Lavandula luisieri; Mc – Myrtus communis; Pl – Pistacia lentiscus; Ru –<br />

Rubus ulmifolius; Uj – Ulex jussiaei. Symbols in bold correspond to obligate<br />

seeders. The developmental stage, as obtained by the respective basal section, is<br />

indicated by the number following the species symbol. Stage 1: 0 to 5 mm 2; stage<br />

2: 5 to 25 mm 2; stage 3: 25 to 125 mm 2; stage 4: > 125 mm 2.<br />

ecologia mediterranea, tome 28, fascicule 1, 2002<br />

PC1 (loadings 0.94 and 0.89, respectively). Root diameter<br />

and SRL were best correlated with PC2 (loadings<br />

0.83 and –0.84, respectively). Root biomass was best<br />

correlated with PC1 (loading 0.82) and R/S was best<br />

correlated with PC2 (loading 0.74). Maximum rooting<br />

depth was poorly correlated with both principal components<br />

(loadings 0.69 and 0.44 for PC1 and PC2, respectively).<br />

The plot of component scores (figure 1B)<br />

presented a clear arrangement of plants according to<br />

their developmental stage, as defined by the basal section<br />

classes, with the only exceptions of a C. crispus and<br />

a C. salviifolius plant. Plants were also arranged according<br />

to the respective species and the two regenerative<br />

strategy groups. Most obligate seeders presented lower<br />

scores for PC2 (specially L. luisieri and C. crispus) while<br />

most resprouters presented higher scores for PC2 (specially<br />

D. gnidium and C. monogyna).This pattern was less<br />

evident at developmental stage 2.<br />

Given the results obtained with the PCA, allometric<br />

relationships were established through linear regression<br />

Figure 2.Relationships between Basal Section and two root system indices:Specific<br />

Root Length and Root-Shoot (root-to-shoot ratio),for two obligate seeders (Cistus<br />

crispus and Lavandula luisieri;represented in bold) and two resprouters (Daphne<br />

gnidium and Crataegus monogyna; normal lettering).<br />

9


10<br />

◆ JOAQUIM S. SILVA, FRANCISCO C. REGO, MARIA A. MARTINS-LOUÇÃO<br />

Variables a b r 2 p<br />

Root biomass (0.1 g) 0.335 1.003 0,79 < 0.001<br />

Shoot biomass (0.1 g) 0.615 1.046 0.83 < 0.001<br />

Root system length (cm) 4.383 0.528 0.61 < 0.001<br />

Root system width (cm) 2.059 0.357 0.55 < 0.001<br />

Table 2.Allometric relationships obtained by linear regression between basal section<br />

(mm 2) and four different root variables: root biomass, shoot biomass, root<br />

system length and root system width.All variables were log-transformed.Biomass<br />

data is indicated in decigrams in order to obtain only positive values. For each<br />

linear regression it is indicated the intercept (a), the slope (b), the coefficient of<br />

determination (r 2) and the associated probability (p).<br />

using the log-transformed values of root biomass, shoot<br />

biomass, root system length and root system width as<br />

independent variables and the log-transformed values of<br />

basal section as the dependent variable (table 2). The<br />

linear regressions for root and shoot biomass presented<br />

coefficients of determination of 0.79 and 0.83 (p < 0.001<br />

for both regressions), respectively. The linear regression<br />

slopes were very similar (1.003 and 1.046, respectively).<br />

In the cases of root system length and root system width<br />

the relationships were still highly significant (p < 0.001)<br />

showing coefficients of determination of 0.61 and 0.55,<br />

respectively.<br />

Figure 3. Images of 16 plants representing<br />

different developmental stages (as defined in figure 1)<br />

of two resprouters:<br />

A – Daphne gnidium (stages 1,2,3 and 4),<br />

B – Crataegus monogyna (stages 1,2,3 and 4);<br />

and two obligate seeders:<br />

C – Lavandula luisieri (stages 1,2,3 and 4),<br />

D – Cistus crispus (stages 1,2,2 and 3).<br />

The vertical bars represent 0.5 m.<br />

Arrows indicate the ground surface.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


ecologia mediterranea, tome 28, fascicule 1, 2002<br />

BELOWGROUND TRAITS OF <strong>MED</strong>ITERRANEAN WOODY PLANTS IN A PORTUGUESE SHRUBLAND ◆<br />

According to the Mann-Whitney U test the ensemble<br />

of obligate seeders presented significantly lower values<br />

than the ensemble of resprouters for maximum rooting<br />

depth (p < 0.001), average root diameter (p < 0.01), root<br />

biomass (p < 0.01) and R/S (p < 0.01). Resprouters presented<br />

significantly lower values than obligate seeders for<br />

SRL (p< 0.01). Basal section, shoot biomass, root width<br />

and root length were not significantly different. Due to<br />

the smaller number of individuals, these differences could<br />

be only partially confirmed within developmental stages<br />

2 (maximum rooting depth, p < 0.05; root biomass,<br />

p


12<br />

◆ JOAQUIM S. SILVA, FRANCISCO C. REGO, MARIA A. MARTINS-LOUÇÃO<br />

roots may be critical for surviving the first dry season after<br />

germination. The expected decreasing trend observed for<br />

SRL (Fitter, 1985) was much more marked for the two<br />

obligate seeder species, which may hint at different strategies<br />

of root growth. Apparently, differences are due to the<br />

preferential investment on structural roots and storage tissues<br />

of the two resprouters at early developmental stages,<br />

contrasting with the preferential investment on fine roots<br />

of the two obligate seeders at similar developmental stages.<br />

Despite the existence of a broad range of R/S ratios<br />

corresponding to different species and different developmental<br />

stages, root biomass and shoot biomass were similarly<br />

and consistently correlated with basal section. This<br />

was confirmed by both the linear regressions and the PCA.<br />

The existence of close relationships between stem variables<br />

(diameter and/or basal section) and root biomass, has been<br />

found in different studies on tree species ( e.g. Santantonio<br />

et al., 1977; Drexhage & Colin, 2001; Hoffmann &<br />

Usoltsev, 2001; Ranger & Gelhaye, 2001). The establishment<br />

of a single consistent allometric relationship for different<br />

species between root biomass and stem variables<br />

(Santantonio et al., 1977), is in accordance with the widespread<br />

pipe model theory, (Shinozaki et al., 1964). Stem<br />

section seems to limit root and shoot development in a<br />

similar way for different species and different developmental<br />

stages. Given the different sample sizes of each<br />

developmental stages and each species, it must be pointed<br />

out that the allometric relationships found are only<br />

valid within the scope of the present work. Sampling limitations<br />

have also to be kept in mind when interpreting the<br />

differences obtained between regenerative strategies. The<br />

similarity of belowground traits among obligate seeders<br />

certainly reflected the phylogenetic relationships between<br />

plants since only two genera could be sampled.Thus, the<br />

existence of common traits can be associated to a common<br />

regenerative strategy, or just simply to the fact that<br />

part of the plants were closely related in terms of phylogeny.<br />

Special sampling designs have been established to<br />

overcome this difficulty (Nicotra et al., 2002) but they are<br />

not always applicable within a single plant community as<br />

in our case. Therefore, the significant differences found<br />

between the two regenerative groups have to be interpreted<br />

also in view of the role of phylogeny. However our<br />

results are supported by similar findings (Hellmers et al.,<br />

1955; Higgins et al., 1987; Bell, 2001) in other <strong>Mediterranea</strong>n-type<br />

regions, which lead us to assume that the same<br />

adaptive mechanisms existing elsewhere are also valid in<br />

the plant community where this study was undertaken.<br />

One of the aspects lacking in our study was the assessment<br />

of the influence of site conditions on belowground<br />

traits. Most authors agree on the fundamental influence<br />

of environmental factors on root systems characteristics<br />

(e.g. Spurr & Barnes, 1980; Kummerow, 1981; Fitter,<br />

1996; Atkinson, 2000). In our case the absence of evident<br />

soil constraints for the development of roots and the proximity<br />

of the excavation sites, lead us to assume the existence<br />

of relatively homogeneous environmental conditions<br />

allowing the comparison of root systems on a genetic and<br />

an ontogenic basis. Nevertheless we have to admit that<br />

the existence of differences in plant density or micro-relief<br />

may have definitely contributed to the overall variance of<br />

belowground traits.<br />

AKNOWLEDGMENTS<br />

We thank the Director of Tapada Nacional de Mafra,<br />

Ricardo Paiva, for making the plant excavation possible,<br />

and two referees for comments on this paper.This study<br />

was supported by EU funds from the research project<br />

Mod<strong>MED</strong> III-Modelling <strong>Mediterranea</strong>n Ecosystem<br />

Dynamics (contract ENV4-CT97-0680).<br />

References<br />

ATKINSON,D., 2000. Root characteristics:Why and what to measure.<br />

In: Smit A.L., Bengough A.G., Engels C., van Noordwijk<br />

M., Pellerin S. & van de Geijn S.C. (eds.), Root Methods, a<br />

Handbook. Springer Verlag, Berlin:1-32.<br />

BAZZAZ F.A., 1997. Allocation of resources in plants: state of the<br />

science and critical questions. In: Bazzaz F.A. & Grace J. (eds.),<br />

Plant Resource Allocation. Academic Press, San Diego: 1-37.<br />

BELL D.T., 2001. Ecological response syndromes in the flora of<br />

southwestern Western Australia: Fire resprouters versus reseeders.<br />

Bot. Rev. 67: 417-440.<br />

BENGOUGH A.G., CASTRIGNANO A., PAGÈS L. & VAN NOORDWIJK<br />

M., 2000. Sampling strategies, scaling and statistics. In: Smit<br />

A.L., Bengough A.G., Engels C., van Noordwijk M., Pellerin<br />

S. & van de Geijn S.C. (eds.), Root Methods, a Handbook.<br />

Springer Verlag, Berlin: 147-173.<br />

BÖHM W., 1979. Methods of studying root systems.Springer Verlag,<br />

Berlin. 188 p.<br />

CALDWELL M.M. & RICHARDS J.H., 1986. Competing root systems:<br />

morphology and models of absorption. In: Givnish T.J.<br />

(ed.), On the Economy of Plant Life and Function. Cambridge<br />

University Press, New York: 251-273.<br />

CALDWELL M.M. & VIRGINIA R.A., 1989. Root systems. In:<br />

Pearcy R.W., Ehleringer J.R., Hooney H.A. & Rundel P. (eds.),<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


ecologia mediterranea, tome 28, fascicule 1, 2002<br />

BELOWGROUND TRAITS OF <strong>MED</strong>ITERRANEAN WOODY PLANTS IN A PORTUGUESE SHRUBLAND ◆<br />

Physiological plant ecology: field methods and instrumentation.<br />

Chapman and Hall, London: 367-398.<br />

CANADELL J., JACKSON R.B., EHLERINGER J.R., MOONEY H.A.,<br />

SALA O.E. & SCHULZE E.-D., 1996. Maximum root depth of<br />

vegetation types at the global scale. Œcologia 108: 583-595.<br />

CANADELL J. & ZEDLER P.H., 1995. Underground structures of<br />

woody plants in <strong>Mediterranea</strong>n Ecosystems of Australia,<br />

California and Chile. In: Arroyo M.T., Zedler P.H. & Fox<br />

M.D. (eds.), Ecology and biogeography of <strong>Mediterranea</strong>n ecosystems<br />

in Chile, California and Australia. Springer Verlag,<br />

Berlin: 177-210.<br />

CANNON W.A., 1949. A tentative classification of root systems.<br />

Ecology 30: 452-458.<br />

CASTROVIEJO S. (ED.), 1999. Flora Iberica. Real Jardin Botánico,<br />

CSIC, Madrid.<br />

CLEMENTE A., REGO F. & CORREIA O., 1996. Demographic patterns<br />

and productivity of post-fire regeneration in Portuguese<br />

mediterranean maquis. Int. J. Wildl. Fire 6: 5-12.<br />

CORREIA O., 1988. Contribuição da fenologia e ecofisiologia em estudos<br />

da sucessão e dinâmica da vegetação mediterrânica. PhD<br />

thesis, University of Lisbon.<br />

CORREIA O. & CATARINO F., 1994. Seasonal changes in soil-toleaf<br />

resistance in Cistus sp and Pistacia lentiscus. Acta Œcol.<br />

15: 289-300.<br />

DREXHAGE M. & COLIN F., 2001. Estimating root system biomass<br />

from breast-height-diameters. Forestry 74: 491-497.<br />

FITTER A.H., 1996. Characteristics and functions of root systems.<br />

In:Waisel Y., Eshel A. & Kafkafi U. Plant Roots, the hidden<br />

half. Marcel Dekker, New York: 1-20.<br />

FITTER A.H., 1985. Functional significance of root morphology<br />

and root system architecture. In: Fitter A.H., Atkinson D.,<br />

Read D.J. & Usher M.B. (eds.), Ecological interactions in soil,<br />

Special publication of the British Ecological Society Nº4.<br />

Blackwell Scientific, Oxford: 87-106.<br />

GRACE J., 1997.Toward models of resource allocation by plants.<br />

In: Bazzaz F.A. & Grace J. (eds.), Plant Resource Allocation.<br />

Academic Press, San Diego: 279-291.<br />

HELLMERS H., HORTON J.S, JUHREN G. & O’KEEFE J., 1955. Root<br />

systems of some chaparral plants in southern California.<br />

Ecology 36: 667-678.<br />

HIGGINS K.B., LAMB A.J. & VAN WILGEN,B.W., 1987. Root systems<br />

of selected plant species in mesic mountain fynbos in<br />

the Jonkershoek Valley, South-Western Cape Province. S. Afr.<br />

J. Bot., 53: 249-257.<br />

HOFFMANN C. & USOLTSEV V.A., 2001. Modelling root biomass<br />

distribution in Pinus sylvestris forests of the Turgai depression<br />

of Kazakhstan. For. Ecol. Manage. 149: 103-114.<br />

JACKSON R.B., CANADELL J., EHLERINGER J.R., MOONEY H.A.,<br />

SALA O.E. & SHULZE E.D. 1996. A global analysis of root<br />

distributions for terrestrial biomes. Œcologia 108: 389-411.<br />

KEELEY J.E., 1986. Resilience of <strong>Mediterranea</strong>n shrub communities<br />

to fires. In: Dell A., Hopkins J.M., & Lamont B.B. (eds.),<br />

Resilience in <strong>Mediterranea</strong>n-type Ecosystems. Dr. Junk<br />

Publishers, Dordrecht: 95-112.<br />

KEELEY J.E. & ZEDLER P.H., 1978. Reproduction of Chaparral<br />

shrubs after fire: a comparison of sprouting and seeding strategies.<br />

Am. Midl. Nat. 99: 142-161.<br />

KOZLOWSKI T.T., KRAMER P.J. & PALLARDY S.G., 1991. The<br />

Physiological Ecology of woody plants. Academic Press Inc.,<br />

San Diego. 657 p.<br />

KUMMEROW J., 1981. Structure of roots and root systems. In: di<br />

Castri F., Goodall D.W. & Specht R.L. (eds.), <strong>Mediterranea</strong>n-<br />

Type Shrublands. Elsevier, New York: 269-288.<br />

NEPSTAD D.C., CARVALHO C.R., DAVIDSON E.A., JIPP P.H.,<br />

LEFEBVRE P.A., NEGREIROS G.H., SILVA E.D. DA,STONE T.A.,<br />

TRUMBORE S.E. & VIEIRA S., 1994.The role of deep roots in<br />

the hydrological and carbon cycles of Amazonian forests and<br />

pastures, Nature 372: 666-669.<br />

NICOTRA A., BABICKA N. & WESTOBY M., 2001. Seedling root<br />

anatomy and morphology: an examination of ecological differentiation<br />

with rainfall using phylogenetically independent<br />

contrasts. Œcologia 130: 136-145.<br />

NOBEL P.S., 1996. Ecophysiology of roots of desert plants, with<br />

special emphasis on Agaves and Cacti. In:Waisel Y., Eshel A.<br />

& Kafkafi U. Plant Roots,the hidden half. Marcel Dekker, New<br />

York: 823-844.<br />

PAGÉS L., ASSENG S., PELLERIN S. & DIGGLE A., 2000. Modelling<br />

root system growth and architecture. In: Smit A.L., Bengough<br />

A.G., Engels C., van Noordwijk M., Pellerin S. & van de<br />

Geijn S.C. (eds.), Root Methods,a Handbook. Springer Verlag,<br />

Berlin: 113-146.<br />

RANGER J. & GELHAYE D., 2001. Belowground biomass and<br />

nutrient content in a 47-year-old Douglas-fir plantation. Ann.<br />

For. Sci. 58: 423-430<br />

SANTANTONIO D., HERMANN R.K. & OVERTON W.S., 1977. Root<br />

biomass studies in forest ecosystems. Pedobiologia 17: 1-31.<br />

SHINOZAKI K.,YODA K., HOZUMI K. & KIRA T., 1964. A quantitative<br />

analysis of plant form- the pipe model theory II.<br />

Further evidence of the theory and its application in forest<br />

ecology. Jap. J. Ecol. 14: 133-139.<br />

SILVA J.S. & REGO F.C., 1998. Factors affecting the establishment<br />

of woody species after fire in Central Portugal. In: L Trabaud<br />

(ed.), Fire management and landscape ecology. International<br />

Association of Wildland Fire, Fairfield: 103-114.<br />

SPECHT R.L. & RAYSON P., 1957. Dark Island heath (ninety mile<br />

plain, South Australia). Aust. J. Bot. 5: 103-114.<br />

SPURR S.H. & BARNES B.V., 1980. Forest Ecolgy. Krieger<br />

Publishing Company, Malabar. 687 p.<br />

ZENG X., 2001. Global vegetation root distribution for land<br />

modelling. J. Hydromet. 2:525-530.<br />

13


Above-ground phytomass and below-ground biomass<br />

production of Salvia verbenaca Linné<br />

Production de phytomasse et biomasse racinaire<br />

chez Salvia verbenaca Linné<br />

M. Conceição Castro 1, Carlos Roquete 2, Luís Gazarini 3<br />

1. Dep. Planeamento Biofísico e Paisagístico, Univ. Évora, Edif. António Verney, Apartado 94, 7002-554 Évora, Portugal;<br />

Fax: + 351 266 744968; email: mccastro@uevora.pt (correspondence)<br />

2. Dep. Zootecnia, Univ. Évora, Núcleo da Mitra, 7002-554 Évora, Portugal<br />

3. Dep. Biologia, Univ. Évora, Núcleo da Mitra, 7002-554 Évora, Portugal<br />

Abstract<br />

Salvia verbenaca L. is a ruderal herbaceous plant, very common<br />

in the <strong>Mediterranea</strong>n region. Its flowers are prized for their appearance,<br />

and it displays a high degree of adaptability in relation to<br />

habitat sites with a paucity of resources.<br />

The objective of this study was to obtain data about the production<br />

of phytomass and root biomass, with a view to the probable<br />

use of this species in urban green spaces. A two-year field experiment<br />

was undertaken, comprising four treatments of eight randomly<br />

selected plots, four being used in each year. This species was<br />

manual seed broadcast at the following densities: 20 000, 15 000,<br />

10 000 and 5 000 viable seeds m 2.<br />

The production of phytomass and root biomass was assessed through<br />

periodic collections, during the seasonal cycle of the species under<br />

study. In the first year there appears to have been a tendency for an<br />

increase in seed density to cause an increase in phytomass production.<br />

In the second year, by which time the plants had settled, when<br />

seed density rose from 5 000 to 10 000, a slight increase in phytomass<br />

production occurred. However, when the seed density rose from<br />

10 000 to 20 000 seeds m -2, a decrease in phytomass production<br />

was observed. Root biomass, in the first year, had minimum value<br />

in January and a maximum value in June-July. In the second year,<br />

the highest values were observed in May. The relationship between<br />

above and below ground biomass, for the first-year experiment,<br />

decreased over the cutting dates. However, an inverse behaviour<br />

occurred in the second year. Although the values of the phytomass<br />

obtained did not reveal statistically significant differences between<br />

treatments, when the plants were found already established, they<br />

appeared to indicate that densities of seeding between 5 000 and<br />

10 000 m 2 could be more advantageous.<br />

Key-words<br />

Phytomass, seed density, root biomass, shoot/root ratio.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 15-22<br />

Résumé<br />

Salvia verbenaca L. est une herbacée rudérale très commune dans<br />

la région méditerranéenne. Ses fleurs sont très prisées pour des raisons<br />

esthétiques. Cette espèce présente un degré d’adaptation important<br />

pour des habitats pauvres en ressources. L’objectif de cette étude<br />

a consisté à analyser la production de la phytomasse et de la biomasse<br />

racinaire dans l’optique d’utiliser cette espèce dans les espaces<br />

verts des villes. Durant deux ans, une expérience a été réalisée, comprenant<br />

quatre traitements avec huit plots aléatoires ; quatre plots<br />

étant utilisés chaque année. Les graines ont été semées selon les densités<br />

suivantes : 20 000, 15 000, 10 000 et 5 000 graines/m2. La production de la phytomasse et de la biomasse racinaires a été<br />

mesurée suite à des récoltes périodiques pendant la durée du cycle<br />

saisonnier. Dans la première année, il semble que l’augmentation de<br />

la densité des semis a conduit à un accroissement de la production de<br />

phytomasse.<br />

Dans la deuxième année, lorsque la plante était déjà installée, l’augmentation<br />

de la densité des graines de 5 000 à 10 000 a entraîné une<br />

légère croissance de la production de phytomasse. Cependant, lorsque<br />

la densité des graines a augmenté de 10 000 à 20 000 graines/m2, une réduction de la production de phytomasse a été observée. Dans la<br />

première année, la biomasse racinaire a présenté une valeur minimum<br />

en janvier et une valeur maximum en juin-juillet. Dans la<br />

deuxième année, les valeurs les plus hautes ont été observées en mai.<br />

La relation entre la phytomasse et la biomasse racinaire, pour la première<br />

année, diminue au fil des dates de coupe. Cependant, un phénomène<br />

inverse s’est produit au cours de la deuxième année. Bien que<br />

les valeurs de la phytomasse n’aient pas révélé de différences significatives<br />

entre les traitements, lorsque les plantes sont déjà établies, elles<br />

indiquent que les densités de graines comprises entre 5 000 et<br />

10 000/m2 pourraient être plus avantageuses.<br />

Mots-clés<br />

Phytomasse, densité des graines, biomasse racinaire, ratio tige/racine.<br />

15


16<br />

◆ M. CONCEIÇÃO CASTRO, CARLOS ROQUETE, LUÍS GAZARINI<br />

INTRODUCTION<br />

The importance of studying spontaneous herbaceous<br />

species from the point of view of their use in urban green<br />

spaces has been approached by various authors, due to<br />

the difficulty that certain exotic species have in adapting<br />

to places with a paucity of resources, and thus making<br />

them unable to survive in adverse conditions. (Cole &<br />

Keen, 1976;Wells, 1987; Brown, 1989; Clément, 1990).<br />

The spatial and temporal distribution of photosynthesis-active<br />

organs plays an important role in determining<br />

the quantity of sunlight which is captured by the<br />

plant, thus contributing towards an increase in the quantity<br />

of elements assimilated (Norman & Campbell, 1992).<br />

Root biomass, expressed in dry weight, reflects the<br />

effort of the plant to distribute the products of photosynthesis.<br />

The thickest roots have a greater capacity for<br />

penetrating more compact soils, are more durable and<br />

contribute over the long term to the development of the<br />

root system (Oliveira et al., 2000).<br />

The partial or total removal of the aerial part of the<br />

plant is invariably followed by compensatory growth, that<br />

is, distributed assimilated elements or reserves of carbohydrates<br />

in the root system are channelled towards the<br />

formation of leaf tissue, and away from root formation,<br />

in such a way that the equilibrium between the two systems<br />

is restored (Troughton, 1977).<br />

There is an interactive relationship between aerial and<br />

root biomass, which is reflected in the level of distribution<br />

of photosynthesised products between the above and below<br />

ground systems, which have simultaneously different and<br />

complementary functions.While the aerial part of the plant<br />

depends on the absorption capacity of the root system for<br />

its growth, the root system needs the carbohydrates pro-<br />

duced by the organs of photosynthesis. Thus, there is an<br />

interdependent relationship between them (Troughton,<br />

1977; Bohm, 1979 cit. in Oliveira, 1988; Klepper, 1991).<br />

In zones with a <strong>Mediterranea</strong>n climate, the amount<br />

and timing of rainfall is often the limiting factor in the<br />

choice of species, the choice being those preferring dry<br />

land to meadowland, having originating in the different<br />

climatic conditions of the <strong>Mediterranea</strong>n.Thus, the objective<br />

of this study was to evaluate the production of phytomass<br />

and root biomass, using four densities of seeding,<br />

with a view to the probable use of the species under study<br />

in urban green spaces.<br />

MATERIALS AND METHODS<br />

Salvia verbenaca (Labiatae) is a herbaceous plant, growing<br />

on uncultivated land, which is very commonly found<br />

in the <strong>Mediterranea</strong>n region, Southern and Western<br />

Europe and Northern Scotland (Tutin et al., 1972). It<br />

produces lilac flowers, with spring blooming (March to<br />

April). The aerial part of the plant ages and dies in early<br />

summer. After the fall of the first autumn rains, the vegetative<br />

stage begins, with the emergence of the first leaves.<br />

Study site<br />

The study was carried out at the University of Évora<br />

in the Alentejo region of Portugal (38º 32’ N; 8º 01’ W;<br />

200 m altitude).The soils were classified as alfisoils (SSS,<br />

1992) or luvisoils according to the FAO/UNESCO (1988).<br />

The climate is an oceanic <strong>Mediterranea</strong>n pluvial-seasonal<br />

bio-climate (MPO), according to Rivas Martínez<br />

(1995). Figure 1 shows climate data for the study period<br />

Figure 1.Thermo-pluviometric chart<br />

referring to Évora -Malagueira Meteorological<br />

Station (1993-1995).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


ABOVE-GROUND PHYTOMASS AND BELOW-GROUND BIOMASS PRODUCTION OF SALVIA VERBENACA LINNÉ ◆<br />

(1993 to 1995). A two-year field experiment was performed<br />

comprising four treatments of eight randomly selected<br />

plots, four being used in each year. Each plot, sited<br />

30 cm apart, had an area of 1 m 2. Seeding was carried<br />

out by manual seed broadcast, using the following densities:<br />

20 000, 15 000, 10 000 and 5 000 viable seeds<br />

(Germination capacity: 65 %; Degree of purity: 99 %;<br />

Field losses: 78.7 %;Weight of 1,000 seeds: 2.643 g). Seed<br />

densities were selected in accordance with the work carried<br />

out by Harper & Gajic, 1961; Ross & Harper, 1972;<br />

Sheldon, 1974;Weaver & Cavers, 1979; Gross & Werner,<br />

1982; Shaw & Antonovics, 1986. The study was performed<br />

in field conditions. In the first year, only four plots<br />

were used. The other four were used in the second year,<br />

when plants were already settled. At the end of the first<br />

year, when the plants reached the stage of senescence, the<br />

above-ground material was removed.<br />

Determination of phytomass<br />

Sampling sites with an area of 625 cm 2 (25 cm x<br />

25 cm) were randomly selected in each plot. A border of<br />

15 cm width was set between each site. Phytomass was<br />

divided into vegetative, support and reproductive structures.<br />

The dry matter weight was recorded from the<br />

samples which were oven-dried at 80 ºC until constant<br />

weight was achieved, after approximately 48 hours (Milner<br />

& Hughes, 1968).The dry weight of the structures, which<br />

included some ash, was related to soil surface unit. Total<br />

phytomass was obtained from the sum of the weight of<br />

all the structures, and calculated per soil surface unit.<br />

Determination of root biomass<br />

At the site where the above-ground material of plants<br />

was removed, four soil samples per layer were extracted<br />

using a hand auger, comprising a cylindrical tube 10 cm<br />

long, with an inside diameter of 7 cm (Oliveira et al., 2000).<br />

Roots were extracted from the soil sample by a hydropneumatic<br />

elutriation system (Smucker et al., 1982;<br />

Oliveira et al., 2000).<br />

Soil samples, containing roots for each depth and each<br />

sampling event, were placed in individual plastic bags,<br />

which were labelled and frozen awaiting laboratory treatment<br />

(Oliveira et al., 2000).<br />

The root dry weight was recorded for root samples,<br />

which were oven-dried at 70 ºC until there was no further<br />

change in weight, after approximately 48 hours<br />

(Böhm, 1979; Schuurman & Goedewaagen, 1971 cit. in<br />

Oliveira et al., 2000).<br />

ecologia mediterranea, tome 28, fascicule 1, 2002, p. 15-22<br />

In accordance with Gregory (1988) and Oliveira et al.<br />

(2000) among others, the root dry weight was related to<br />

soil volume unit.<br />

The relationship between above<br />

and below ground biomass<br />

In order to evaluate the relationship between above<br />

and below ground biomass, the shoot-root ratio was estimated<br />

(Böhm, 1979; Pagès et al., 2000; Atkinson, 2000).<br />

Data Analysis<br />

The data recorded were analysed by descriptive statistics<br />

and analysis of variance (ANOVA), followed by<br />

comparison of means among treatments, using Newman/<br />

Keul’s test.<br />

RESULTS AND DISCUSSION<br />

Phytomass production<br />

The mean values of phytomass dry weight, in the first<br />

year of the study, varied from 141.9±79.9 g.m -2 (seed<br />

density 5 000) to 189.6±89.6 g.m -2 (seed density 20 000).<br />

Table 1 shows that there was a tendency for an<br />

increase in phytomass up until June (223 days after seeding),<br />

when production peaked. In the second year<br />

(table 2), there was a tendency for production to peak<br />

in May, 212 days after the emergence of the first leaves.<br />

Average annual values varied from 502.3±230.7 g.m -2<br />

(seed density 20 000) to 554.3±290.1 g.m -2 (seed density<br />

10 000).Thus, in the first year, there seems to have<br />

been a tendency for the increase in seed density to cause<br />

an increase in phytomass production. In the second year,<br />

a time by which the plants had settled, when seed density<br />

rose from 5 000 to 10 000, an increase in phytomass<br />

production occurred. However, when the seed density<br />

rose from 15 000 to 20 000, a decrease in phytomass<br />

production was observed. According to Harper<br />

(1961), the above ground weight of the plant may not<br />

initially be influenced by seed density but, at high seed<br />

densities, the competition between individual plants leads<br />

to a reduction in individual plant weight.The behaviour<br />

of the species observed in this study may be explained<br />

by competition being less when there are fewer plants<br />

per soil surface unit, with these plants taking the opportunity<br />

to increase phytomass production.<br />

17


18<br />

◆ M. CONCEIÇÃO CASTRO, CARLOS ROQUETE, LUÍS GAZARINI<br />

Treatments<br />

Cutting date Sv 20 000 Sv 15 000 Sv 10 000 Sv 5 000<br />

17/01/94 51.0 ± 7.6 54.7 ± 7.4 37.3 ± 11.0 18.6 ± 6.8<br />

c c b a<br />

07/03/94 199.7 ± 99.6 134.9 ± 31.8 143.6 ± 34.8 80.9 ± 5.3<br />

b ab ab a<br />

11/04/94 176.3 ± 58.9 154.7 ± 24.8 192.2 ± 35.8 156.8 ± 38.6<br />

a a a a<br />

09/05/94 224.4 ± 61.3 155.6 ± 25.4 158.9 ± 23.1 169.6 ± 25.3<br />

a a a a<br />

06/06/94 271.4 ± 30.4 222.9 ± 29.7 240.9 ± 51.3 270.6 ± 14.7<br />

a a a a<br />

11/07/94 223.0 ± 81.5 165.4 ± 19.5 185.9 ± 46.9 170.7 ± 18.0<br />

a a a a<br />

Annual average 189.6 ± 89.6 148.0 ± 55.3 156.9 ± 70.8 141.9 ± 79.9<br />

b a a a<br />

Different letters note significant differences (P


ABOVE-GROUND PHYTOMASS AND BELOW-GROUND BIOMASS PRODUCTION OF SALVIA VERBENACA LINNÉ ◆<br />

Table 3. Salvia verbenaca L. (Sv)<br />

average dry weight<br />

of root bomass (mg.m -3),<br />

for periodic cuttings and treatments.<br />

Table 4. Salvia verbenaca L. (Sv)<br />

average dry weight<br />

of root bomass (mg.m -3),<br />

for periodic cuttings and treatments.<br />

Table 5. Shoot and root ratio (s/r)<br />

of Salvia verbenaca L. (Sv),<br />

for periodic cuttings and treatments.<br />

ecologia mediterranea, tome 28, fascicule 1, 2002, p. 15-22<br />

Treatments<br />

Cutting date Sv 20 000 Sv 15 000 Sv 10 000 Sv 5 000<br />

17/01/94 0.14 ± 0.04 0.12 ± 0.03 0.12 ± 0.02 0.05 ± 0.01<br />

a a a b<br />

07/03/94 0.24 ± 0.19 0.24 ± 0.19 0.28 ± 0.19 0.15 ± 0.12<br />

a a a a<br />

11/04/94 0.42 ± 0.39 0.43 ± 0.35 0.48 ± 0.42 0.36 ± 0.26<br />

a a a a<br />

09/05/94 0.58 ± 0.50 0.52 ± 0.43 0.52 ± 0.41 0.50 ± 0.45<br />

a a a a<br />

06/06/94 0.58 ± 0.58 0.55 ± 0.58 0.57 ± 0.60 0.65 ± 0.49<br />

a a a a<br />

11/07/94 0.75 ± 0.80 0.60 ± 0.65 0.70 ± 0.63 0.47 ± 0.53<br />

a a a a<br />

Annual average 0.54 ± 0.58 0.49 ± 0.50 0.53 ± 0.51 0.45 ± 0.45<br />

a a a a<br />

Different letters note significant differences (P


20<br />

◆ M. CONCEIÇÃO CASTRO, CARLOS ROQUETE, LUÍS GAZARINI<br />

Treatments<br />

Cutting date Sv 20 000 Sv 15 000 Sv 10 000 Sv 5 000<br />

19/12/94 1.32 1.08 1.02 1.31<br />

31/03/95 2.92 2.44 2.10 3.78<br />

23/05/95 2.64 2.72 2.97 2.72<br />

10/07/95 3.12 1.99 2.01 3.60<br />

Annual average 2.50 ± 0.81 2.06 ± 0.72 2.02 ± 0.80 2.85 ± 1.13<br />

Different letters note significant differences (P


According to Davidson (1969), a plant constitutes a<br />

“whole”, although the absorption of water and nutrients<br />

is carried out by the roots in distinct edaphic conditions<br />

from those in which photosynthesis occurs. As the difference<br />

of temperature between leaves and roots changes,<br />

there seems to be a tendency for the plant to compensate<br />

for this by means of changes in root activity, increasing<br />

or decreasing the relative size of the root system. In field<br />

conditions, according to Davidson, there is great degree<br />

of disparity between root activity and photosynthesis activity<br />

as days decrease in length and summer progresses<br />

towards winter, in which the drop in soil temperature is<br />

higher than the drop in air temperature, and thus root<br />

growth is limited. This may provide a possible explanation<br />

for the seasonal variation of the patterns of growth<br />

observed in this study.<br />

When either the root or the aerial part of the plant is<br />

removed, a compensatory process is activated in the plant<br />

which restores the s/r ratio which is characteristic of the<br />

species. Brouwer and De Wit (1969 cit. in Klepper, 1991)<br />

state that when they removed half of the aerial part of<br />

bean plants they found that, later, the growth of the aerial<br />

part was faster than that of the roots until the initial r/s<br />

ratio was attained. This could contribute to an understanding<br />

of the evolution of the s/r relationship in the<br />

second year of this study, by virtue of the fact that, at the<br />

end of the first year, the aerial part of the plants on the<br />

plots which were the object of study in the second year<br />

was removed in its entirety.<br />

CONCLUSION<br />

ABOVE-GROUND PHYTOMASS AND BELOW-GROUND BIOMASS PRODUCTION OF SALVIA VERBENACA LINNÉ ◆<br />

The levels of production for each of the two years of<br />

the study were not compared statistically, because the first<br />

year was for seeding. However it was an evidently lower<br />

level of production in the first year. In the second year,<br />

by which time the plants had settled, there seems to have<br />

been a tendency for the increase in seed density to cause<br />

a decrease in phytomass production.This behaviour may<br />

be explained by possible competition between individuals,<br />

at high seed densities.<br />

Assuming the above and below ground system interdependency,<br />

shoot/root ratio provided some indications<br />

as to the changes occurring in growth in these systems.<br />

In the first year of the study, s/r values progressively<br />

decreased throughout the cutting dates, which could be<br />

interpreted as a greater increase in the below ground than<br />

in the above ground system. In the second year, the oppo-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 15-22<br />

site occurred. Probably, due to the low levels of precipitation,<br />

which occurred in the first year of the study, root<br />

system development predominated, allowing deeper soil<br />

layers to be explored. In the second year, the plants produced<br />

more aboveground biomass, probably using the<br />

reserves stored in the root system. Nonetheless, the removal<br />

of the entirety of the aboveground biomass of the existing<br />

plants, at the end of the first year, after senescence,<br />

could also have contributed to the plants attempting to<br />

adjust in the following year to the s/r relationship, which<br />

is characteristic of the species.<br />

The results seem to indicate that biomass production<br />

was more sensitive to environmental or seasonal changes<br />

than to seed densities.<br />

Nowadays, in Portugal, the seeds from wild plants are<br />

collected by hand in its habitat.Timing seed collection is<br />

one of the most crucial and difficult steps in the process<br />

as flowers and fruits are produced in succession. Thus,<br />

the seed collector has either to select only mature seeds<br />

or to wait until the whole inflorescence will be ripen, losing<br />

seeds from the earliest ripened fruits.Therefore, this is an<br />

expensive method of obtaining large quantities of seeds.<br />

Although the values of the phytomass obtained did not<br />

reveal statistically significant differences between treatments,<br />

when the plants were found already established,<br />

they appeared to indicate that densities of seeding between<br />

5 000 and 10 000 m -2 could be more advantageous,<br />

reducing seed collection.<br />

References<br />

ANTEN N.P.R. & HIROE T., 1999. Interspecific differences in<br />

above-ground growth patterns resulting in spatial and temporal<br />

partitioning of light among species in a tall meadow<br />

grass. J. Ecol. 87: 583-597.<br />

ATKINSON D., 2000. Root characteristics: why and what to measure.<br />

In: Smit et al. (eds), Root methods:a handbook. Springer-<br />

Verlag, Berlin, 1-32.<br />

BÖHM W., 1979. Methods of studying root systems. Springer-Verlag,<br />

New York.<br />

BROWN R. J., 1989. Seeds of success. Lands. Design 181: 34-36.<br />

CLÉMENT G., 1990. Le Jardin en mouvement. Pandora Éditions,<br />

Paris.<br />

COLE L. & KEEEN C., 1976. Dutch techniques for the establishment<br />

of natural plant communities in urban areas. Lands.<br />

Design 116: 31-34.<br />

DAVIDSON R. L., 1969. The effect of root/leaf temperature differentials<br />

on root/shoot ratios in some pasture grasses and<br />

clover. Ann. Bot. 33: 561-569.<br />

21


22<br />

◆ M. CONCEIÇÃO CASTRO, CARLOS ROQUETE, LUÍS GAZARINI<br />

FAO/UNESCO, 1988. Soil map of the world. Food and<br />

Agriculture Organization of the United Nations, Rome.<br />

GREGORY P. J., 1988. Growth and functioning of plant root. In:<br />

Wild A. (ed.), Russel’s soil conditions and plant growth. Bath.<br />

Press, Avon, 113-167.<br />

GROSS K. J. & WERNER P., 1982.The colonizing abilities of biennial<br />

plant species in relation to ground cover: implications<br />

for their distributions in a successional series. Ecology 63:<br />

921-931.<br />

HARPER J. L. & GAJIC D., 1961. Experimental studies of the mortality<br />

and plasticity of a weed. Weed Res. 1: 91-104.<br />

HARPER J. L., 1961. Approaches to the study of plant competition.<br />

Symposia of the society for Exp. Biol. 15: 1-39.<br />

KLEPPER B., 1991. Root-shoot relationships. In:Waisel,Y., Eshel,<br />

A. & Kafkafi, U. (eds.), Plant Roots.The hidden half. Marcel<br />

Dekker Inc., New York, 265-286.<br />

KUIPER D., 1992. Phenotypic plasticity in Plantago species. In:<br />

Kuiper, P. J. C. & Bos, M. (eds.), Plantago:A Multidisciplinary<br />

study. Springer-Verlag, New York, 169-183.<br />

MILNER C. & HUGHES R. E., 1968. Methods for measurement of<br />

the primary production of grassland. IBP Handbook. Blackwell<br />

Scientific Publications, Oxford.<br />

NORMAN J. M. & CAMPBEL G. S, 1992. Canopy structure. In:<br />

Pearcy R. W., Ehleringer J., Mooney H. A. & Rundel P. W.<br />

(eds). Plant physiological ecology. Field methods and instrumentation.<br />

Chapman & Hall, London, 301-325.<br />

OLIVEIRA M. R. G., 1988. Comportamento do sistema radical de<br />

algumas espécies forrageiras e pratenses com interesse para a<br />

modernização dos sistemas de agricultura no Alentejo. PhD<br />

Universidade de Évora.<br />

OLIVEIRA M. R., NOORDWIJK M.VAN,GAZE S. R., BROUWER G.,<br />

BONA S., MOSCA G. & HAIRIAH K, 2000. Auger sampling, in<br />

growth cores and pinboard methods. In: Smit et al. (eds.),<br />

Root methods: a handbook. Springer-Verlag, Berlin, 175-210.<br />

PAGÈS L., ASSENG S., PELLERIN S. & DIGGLE A., 2000. Modelling<br />

root system growth and architecture. In: Smit et al. (eds.),<br />

Root methods: a handbook. Springer-Verlag, Berlin, 113-146.<br />

RIVAS-MARTÍNEZ S., 1995. Classificación bioclimática de la tierra.<br />

Folia Bot. Madritensis 16: 1-32.<br />

ROSS M. A. & HARPER J., 1972. Occupation of biological space<br />

during seedling establishment. J. Ecol. 60: 77-88.<br />

SHAW R. G. & ANTONOVICS J., 1986. Density in Salvia lyrata, a<br />

herbaceous perennial: the effects of the experimental alteration<br />

of seed densities. J. Ecol. 74: 797-813.<br />

SHELDON J. C., 1974. The behaviour of seeds in soil. I - The<br />

influence of seed morphology and the behaviour of seedlings<br />

on the establishment of plants from surface-lying seeds.<br />

J. Ecol. 62: 47-66.<br />

SMUKER A. J. M., MC BURNEY S. L. & SRIVASTAVA A. K., 1982.<br />

Quantitative separation of root from compacted soil profiles<br />

by hidropneumatic elutriation system. Agron. J. 74:<br />

500-503.<br />

SOIL SURVEY STAFF (SSS), 1992. Key to soil taxonomy. 5th ed.,<br />

Pocohontas Press, Virginia.<br />

TAPPEINER U. & CERNUSCA A., 1996, Microclimate and fluxes<br />

of water vapour, sensible heat and carbon dioxide in structurally<br />

differing sub-alpine plant communities in the Central<br />

Caucasus. Plant Cell Envir. 19: 403-417.<br />

TAPPEINER U. & CERNUSCA A., 1998. Model simulation of spatial<br />

distribution of photosynthesis in structurally differing<br />

plant communities in the central Caucasus. Ecol. Model. 113:<br />

201-223.<br />

TROUGHTON 1977. Relationship between the root and shoot systems<br />

of grasses. In: Marshall (ed.), The below ground ecosystem:<br />

a synthesis of plant-associated processes. Colorado State<br />

University, 26: 39-51.<br />

TUTIN T. G., HEYWOOD V. H., BURGES N. H., MOORE D. M.,<br />

VALENTINE D. H.,WALTERS S. M. & WEB D. A. (EDS.), 1972.<br />

Flora Europaea. Cambridge University Press, Cambridge,<br />

vol. 3.<br />

WEAVER S. E. & CAVERS P. B., 1979.The effects of date of emergence<br />

and emergence order on seedling survival rates in<br />

Rumex crispus and Rumex obtusifolius.Can.J.Bot.57: 730-738.<br />

WELLS T. C. E., 1987. The establishment of floral grasslands.<br />

Acta Horticulturae 195: 59-67.<br />

ZELLER V., BAHN M., AICHNER M. & TAPPEINER U., 2000. Impact<br />

of land-use change on nitrogen mineralization in sub-alpine<br />

grasslands in the southern Alps. Biol.Fertile Soils 31: 441-448.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


Diversité génétique des populations naturelles<br />

de l’Argyrolobium uniflorum Jaub. et Spach. (Fabaceae) en Tunisie<br />

Genetic diversity in Tunisian populations<br />

of Argyrolobium uniflorum Jaub. et Spach. (Fabaceae)<br />

Zaouali Yosr, Ben Fadhel Najeh, Boussaid Mohamed<br />

Laboratoire de biotechnologie végétale, Institut national des sciences appliquées et de technologie, BP 676, 1080 Tunis Cedex, Tunisie<br />

email : bossdmohd@insat.rnu.tn – Tél. : (216) 71 703 829 (929) – Fax : (216) 1 71704 329<br />

Résumé<br />

Argyrolobium uniflorum Jaub. et Spach., est une Fabacée pastorale,<br />

diploïde (2n=2x=16) et cléistogame, qui pousse spontanément<br />

dans les parcours steppiques de la Tunisie centrale et méridionale.<br />

Les populations de cette espèce disparaissent progressivement sous l’influence<br />

continue du surpâturage et du défrichement des parcours.<br />

Les mesures de sauvegarde de l’espèce doivent passer au préalable par<br />

l’analyse de la diversité génétique des populations naturelles.<br />

Douze populations d’Argyrolobium uniflorum, collectées dans différentes<br />

régions de la Tunisie, ont fait l’objet d’une analyse du polymorphisme<br />

de six systèmes enzymatiques.<br />

Vingt-quatre loci, dont douze se sont avérés polymorphes, ont été<br />

détectés. La diversité génétique intrapopulation, estimée par le<br />

nombre moyen d’allèles par locus polymorphe, le pourcentage de<br />

loci polymorphes et les taux moyens d’hétérozygotie, est importante.<br />

Elle résulterait d’une forte hétérogénéité des populations ancestrales<br />

et de migration de graines à partir de populations adjacentes.<br />

La différenciation des populations, estimée par les indices de fixation<br />

de Wright et par une analyse en composantes principales effectuée sur<br />

les fréquences alléliques, est assez forte. La perturbation des sites et la<br />

cléistogamie de l’espèce joueraient un rôle primordial dans cette structuration.<br />

L’indice d’identité génétique de Nei (1978) entre les populations<br />

est élevé. La faible divergence génétique entre les populations<br />

et leur forte différenciation témoigneraient de leur fragmentation<br />

récente, essentiellement sous l’influence des pressions anthropiques. Le<br />

dendrogramme établi à partir des indices d’identité génétique de Nei<br />

(UPGMA) fait ressortir deux groupes de populations. Au sein de<br />

chaque ensemble, l’agrégation des populations selon leur proximité<br />

géographique ou bioclimatique ne s’est opérée que pour un nombre<br />

limité d’entre elles. La conservation de l’espèce doit intégrer dans un<br />

premier temps la collecte d’un maximum d’individus à l’intérieur<br />

des populations et préserver celles ayant présenté un niveau élevé de<br />

polymorphisme, particulièrement celles à allèles rares.<br />

Mots-clés<br />

Argyrolobium uniflorum, cléistogamie, conservation, diversité<br />

génétique, isozymes, Tunisie.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 23-37<br />

Abstract<br />

Argyrolobium uniflorum Jaub. et Spach., is a pasture legume,<br />

diploid (2n=2x=16) and cleistogamous which grows in the Central<br />

and the Southern rangelands of Tunisia. Natural populations are<br />

endangered by overgrazing and clearing. Conservation actions of<br />

the species require knowledge of the genetic diversity of populations.<br />

Twelve Tunisian Argyrolobium uniflorum populations collected<br />

in different parts of Tunisia, were assessed by the analysis of six<br />

isozymes.<br />

The enzymes analyzed were encoded by 24 loci and 12 out of them<br />

were polymorphic.<br />

The mean number of alleles per polymorphic locus, the percentage<br />

of loci polymorphic and the observed and the expected heterozygotities<br />

were very high. They indicate a high level of genetic diversity<br />

within populations as a result both of a high diversity of genets<br />

within the former populations and the seed recruitment from adjacent<br />

populations. A high genetic differentiation of populations, estimated<br />

by Wright’s indices and a principal component analysis performed<br />

on allelic frequencies, was detected. The species mating system<br />

and site perturbances may play great roles in population differentiation.<br />

The genetic divergence between populations estimated<br />

by Nei’s (1978) genetic identity was globally low. The high level<br />

of similarities between populations and their substantial differentiation<br />

might indicate recent population isolation dictated essentially<br />

by anthropic pressures.<br />

The UPGMA, established through genetic Nei’s genetic identities,<br />

revealed two groups of populations. Within each group, populations<br />

from the same bioclimatic stage or geographically near were not<br />

clustered obligatory together.<br />

Preservation of populations with high levels of polymorphism and<br />

exhibiting unique alleles, and the collect of a high number of individuals<br />

within populations may constitute the starting point for<br />

the species conservation program.<br />

Key-words<br />

Argyrolobium uniflorum, cleistogamy, conservation, genetic diversity,<br />

isozymes, Tunisia.<br />

23


24<br />

◆ ZAOULI YOSR, BEN FADHEL NAJEH, BOUSSAID MOHA<strong>MED</strong><br />

ABREGED ENGLISH VERSION<br />

Argyrolobium uniflorum Jaub. et Spach. (Fabaceae) is<br />

a pasture legume growing spontaneously in Tunisian arid<br />

rangelands. The species is diploid (2n=2x=16), strongly<br />

and even strictly cleistogamous. It produces a high number<br />

of pods with numerous small seeds which are dispersed<br />

by wind and streaming.<br />

Overgrazing and human activities (farming, clearing<br />

of rangelands) in Tunisian arid areas had led to habitat<br />

destruction and to the reduction of the number and size<br />

of populations. At present, populations were restricted to<br />

limited perturbed lands (cultivated and abandoned cereal<br />

fields) and to some recent protected pasture areas.<br />

Rangeland management programs and conservation<br />

actions that will promote adapted Argyrolobium uniflorum<br />

genotypes require knowledge of the genetic variability of<br />

the species first.<br />

In this study, we investigated the genetic diversity of<br />

twelve populations representing the geographic area of the<br />

species using 13 % starch horizontal gel electrophoresis.<br />

Six isozymes (PGI, MDH, PGM, ADH, 6PGD and ICD)<br />

were analyzed to assess genetic variations within and among<br />

populations.Twenty individuals per population were sampled<br />

at random. Electrophoresis protocols (protein extraction,<br />

buffer preparation, gel staining…) were taken from<br />

Pasteur et al. (1987), and Wendel and Weeden, (1989).The<br />

computer program Biosys 1 (Swofford & Selander, 1981)<br />

was used to calculate standard parameters in order to assess<br />

the levels of genetic variation (number of alleles per polymorphic<br />

locus, percentage of polymorphic loci, observed<br />

and expected heterozygosities). Wright’s F statistics and<br />

deviation from Hardy-Weinberg expectations were calculated<br />

to estimate the level of heterozygosity and population<br />

differentiation. A principal component analysis, performed<br />

on allelic frequencies, was also used to estimate the differentiation<br />

of populations. Genetic divergence between<br />

populations was estimated using Nei’s (1978) unbiased<br />

genetic identity. A hierarchical cluster analysis of genetic<br />

identity values was performed using the UPGMA method.<br />

The species maintains high levels of genetic diversity<br />

within populations. Genetic measure parameter values<br />

were higher than those reported for cleistogamous or autogamous<br />

species (57.64 % of polymorphic loci, 1.62 alleles<br />

per polymorphic locus, and 0.103 and 0.207 respectively<br />

observed and expected heterozygosities).The high levels<br />

of variation within populations might result from the high<br />

diversity of genets at the starting population and the<br />

recruitment of new genets via seed migration from adjacent<br />

populations by wind and streaming.<br />

A deficiency of heterozygosity for the majority of loci<br />

for most of the populations and a high level of inbreeding<br />

(F IS = 0,488) were detected as a result of the species<br />

mating system.<br />

Estimated values of Wright’s F indexes indicated a substantial<br />

structuration among populations (F ST = 0,163).<br />

The number of migrants per generation is low (Nm =<br />

1.28). The high genetic differentiation and the low gene<br />

flow may be the result of both site perturbance by human<br />

activities and the species mating system. Plots derived<br />

from the ACP analysis showed also a large separation between<br />

populations.<br />

The average of Nei’s (1978) genetic identity values (I)<br />

was high (0.917 to 1.000), and reflected the paucity of<br />

genetic divergence between populations.The high similarities<br />

between populations and the high level of their genetic<br />

differentiation were indicators of their recent isolation.<br />

UPGMA analysis showed at I = 0.940 two aggregates<br />

of populations. In each subgroup, there is no evident relationship<br />

between population neighboring, bioclimatic<br />

stage population belonging and clustering.<br />

Conservation implications of our survey are still fragmentary.<br />

Other studies including molecular and demographic<br />

estimates are necessary to elaborate an appropriate<br />

population preservation program. However, our<br />

study showed that some populations exhibited high levels<br />

of polymorphism and some of them have particular alleles.<br />

The conservation of these populations and the collect of<br />

a high number of individuals within populations may<br />

constitute a starting step for conservation programs.<br />

INTRODUCTION<br />

Les parcours naturels de la Tunisie présentent une<br />

diversité floristique spécifique et infraspécifique importante<br />

(Gounot, 1995 ; Le Houérou, 1969 ; Le Floc’h,<br />

1995 ; Thiault, 1995 ; Waechter, 1995).<br />

Le défrichement des parcours (au profit de la céréaliculture<br />

et l’arboriculture) et le surpâturage, en Tunisie centrale<br />

et méridionale, ont conduit à une dégradation des<br />

habitats et à un appauvrissement de leur diversité floristique<br />

(Boussaid et al., 1998). Un nombre élevé d’espèces,<br />

à bonne valeur pastorale, se raréfient progressivement et<br />

sont remplacées par des espèces peu ou pas palatables<br />

(Boussaid et al., 1998 ; Ben Fadhel et al., 2000). Medicago<br />

sativa L. ssp. tunetana Murb. (Fabaceae), Cenchrus ciliaris<br />

L. (Poaceae),Tricholaena teneriffae Link. (Poaceae),<br />

Tetrapogon villosus L. (Poaceae) et Hedysarum carnosum<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


DIVERSITÉ GÉNÉTIQUE DES POPULATIONS NATURELLES DE L’ARGYROLOBIUM UNIFLORUM JAUB. ET SPACH… ◆<br />

Desf. (Fabaceae), ne sont actuellement représentés que<br />

par de petites populations dans des sites refuges (parcelles<br />

mises en défens, parcs nationaux…). La présence<br />

d’Hedysarum flexuosum Desf. et H. humile L. n’est attestée<br />

que par des listes d’herborisation fort anciennes (Quézel,<br />

1991 ; Schoenenberger, 1995 ; Boussaid et al., 1998).<br />

Argyrolobium uniflorum Jaub. et Spach., Fabaceae pastorale<br />

diploïde (2n=2x=16), à port prostré et appartenant<br />

à la tribu des Génistées (Raynaud, 1974) est bien<br />

représentée dans les parcours steppiques de la Tunisie<br />

centrale et méridionale. Elle est considérée, comme les<br />

autres espèces du même genre [A. abissynicum Jaub. et<br />

Spach., A. arabicum (Desf.) Jaub. et Spach., A. zanonii<br />

(Turra) Ball. et A. saharaea (Pomel)], comme une bonne<br />

espèce pastorale pouvant intervenir dans la réhabilitation<br />

des parcours en milieux arides de l’Afrique du Nord<br />

(Raynaud, 1974 ; Lwogo, 1984 ; Attia, 1986 ; Ben Fadhel<br />

et al., 2000). Elle est très appréciée par le bétail, présente<br />

une bonne valeur nutritive et a une production de matière<br />

verte étalée dans le temps (de septembre à juillet). Elle<br />

résiste au piétinement et au surpâturage (développement<br />

de bourgeons végétatifs latents à tous les niveaux de la<br />

plante, port prostré…) et produit un nombre élevé de<br />

gousses (plus de 50 par individu) (Le Houérou & Ionsco,<br />

1973 ; Floret & Pontanier, 1982 ; Zaouali, 1999 ; Ben<br />

Fadhel et al., 2000). La déhiscence des gousses est échelonnée<br />

dans le temps et une à cinq graines restent souvent<br />

conservées dans la partie proximale de la gousse,<br />

proche du pédoncule (Zaouali, 1999). Les graines (8 à<br />

15 par gousse) sont ovoïdes, de très petite taille (1,5-<br />

3 mm) et de couleur variable (jaune, brun et orange).<br />

Elles sont dispersées par le vent et les eaux de ruissellement<br />

(Zaouali, 1999 ; Ben Fadhel et al., 2000). L’espèce<br />

se reproduit sous un régime très fortement, sinon exclusivement,<br />

cléistogame (Abdelkefi et al., 1990). Les fleurs,<br />

axillaires, sont jaunâtres et de petite taille (4 à 6 mm de<br />

longueur). La déhiscence des anthères et la formation<br />

des graines s’opèrent avant l’épanouissement de la fleur<br />

(Dardour et al., 1989 ; Abdelkefi et al., 1990). Ce n’est<br />

que dans de très rares cas que nous avons pu observer<br />

une émergence du stigmate en dehors du bouton floral<br />

(3 fleurs sur 600 individus d’origine géographique<br />

diverse). Cette situation favoriserait des allofécondations<br />

à condition que le stigmate soit encore réceptif, que les<br />

ovules (ou une bonne proportion d’entre eux) ne soient<br />

pas fécondés par l’autopollen, et qu’il n’y ait pas des phénomènes<br />

d’allo-incompatibilité.<br />

L’espèce est décrite dans les flores comme étant une<br />

chaméphyte pérenne pouvant se reproduire aussi bien par<br />

graine que par drageon (Quézel & Santa, 1962 ; Ozenda,<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 23-37<br />

1977 ; Pottier Alapetite, 1979). Toutefois, en Tunisie et<br />

dans les conditions naturelles, l’espèce se comporte<br />

comme une annuelle, ne se multipliant le plus souvent<br />

que par graine. En effet, la production de drageons est<br />

faible et le développement de ces structures est très limité<br />

(1 à 2 cm) suite au dessèchement estival des plantes<br />

(observations personnelles).<br />

En Tunisie, les populations d’Argyrolobium uniflorum<br />

se raréfient progressivement sous l’influence du surpâturage<br />

et du défrichement des parcours. Des mesures de<br />

sauvegarde s’imposent pour assurer la diversité floristique<br />

des parcours steppiques, préserver la diversité génétique<br />

de l’espèce et promouvoir sa gestion durable. La mise en<br />

œuvre de stratégies efficaces de conservation nécessite au<br />

préalable une évaluation de la diversité génétique des<br />

populations par des marqueurs génétiques efficaces<br />

(isoenzymatiques et moléculaires particulièrement). Cette<br />

étude, chez des espèces menacées d’extinction telles que<br />

l’Argyrolobium uniflorum, devrait permettre d’estimer la<br />

taille effective des populations, d’évaluer leur dépression<br />

de consanguïnité, d’analyser leur structuration génétique<br />

et de mesurer les flux géniques entre elles (Batista et al.,<br />

2001 ; Hedrick, 2001). L’importance relative de ces paramètres,<br />

associée à la connaissance de la biologie de l’espèce<br />

(régime de reproduction notamment) et des facteurs<br />

du milieu environnant, doit contribuer à évaluer l’organisation<br />

de la variabilité à l’intérieur et entre les populations,<br />

et à orienter efficacement les programmes de restauration<br />

des populations naturelles.<br />

L’analyse du polymorphisme isoenzymatique, en dépit<br />

de ses limites (détection d’une faible proportion de variations<br />

pour existence d’enzymes multigéniques, redondance<br />

du code génétique, modifications post transcriptionnelles…),<br />

constitue actuellement un moyen utile et<br />

relativement peu coûteux pour évaluer la variabilité génétique<br />

et le degré de différenciation des populations, et<br />

mesurer les flux géniques entre elles afin de mieux étayer<br />

les stratégies de conservation (Pernès, 1984).<br />

MATÉRIELS ET MÉTHODES<br />

Populations analysées<br />

Douze populations d’Argyrolobium uniflorum, représentatives<br />

de l’aire géographique de l’espèce en Tunisie,<br />

ont été analysées. Les populations se développent dans les<br />

étages bioclimatiques aride inférieur et aride supérieur<br />

(Emberger, 1955) sous des pluviométries qui varient de<br />

25


26<br />

◆ ZAOULI YOSR, BEN FADHEL NAJEH, BOUSSAID MOHA<strong>MED</strong><br />

Figure 1. Localisation<br />

géographique et bioclimatique<br />

des douze populations<br />

d’Argyrolobium uniflorum<br />

analysées en Tunisie.<br />

Figure 1. Geographic<br />

and bioclimatic situation<br />

of the twelve<br />

Argyrolobium uniflorum<br />

populations analyzed in Tunisia.<br />

1, 2, 3… : populations analysées<br />

Hi : humide inférieur,<br />

Sh : subhumide,<br />

Sas : semi-aride supérieur,<br />

Sam : semi-aride moyen,<br />

Sai : semi-aride inférieur,<br />

As : aride supérieur,<br />

Ai : aride inférieur,<br />

S:saharien.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


DIVERSITÉ GÉNÉTIQUE DES POPULATIONS NATURELLES DE L’ARGYROLOBIUM UNIFLORUM JAUB. ET SPACH… ◆<br />

150 à 300 mm/an. Les populations se rencontrent dans<br />

les lits et sur les berges d’oueds temporaires et sur des terrains<br />

sableux dérivant de steppes primaires à chaméphytes<br />

pérennes (Rhanterium suaveolens (Desf.) Q. et S., Artemisia<br />

campestris L. et A.herba-alba Asso.,Stipa tenacissima L.,etc.)<br />

défrichées récemment au profit de la céréaliculture et la<br />

création de réserves fourragères (Opuntia ficus-indica L.,<br />

Acacia L. ssp.). La localisation géographique des populations<br />

est représentée dans la figure 1. Les caractéristiques<br />

écologiques essentielles des sites figurent dans le tableau 1.<br />

Vingt individus pourvus de gousses mûres ont été collectés<br />

au hasard, dans chaque population, dans une superficie<br />

qui varie de 2 à 7 ha. Les plantes ont été prélevées<br />

à des distances supérieures à 10 m pour éviter la collecte<br />

d’individus issus d’une même plante mère.<br />

Sur chaque plante, nous avons prélevé une gousse.<br />

Une graine par gousse, représentant un individu par population,<br />

a été mise en germination dans une boîte de Pétri<br />

sur papier filtre imbibé d’eau pendant une semaine.<br />

Electrophorèse et systèmes<br />

enzymatiques analysés<br />

Six systèmes enzymatiques : la malate déshydrogénase<br />

(MDH ; EC 1.1.1.37), l’alcool déshydrogénase (ADH ;<br />

EC 1.1.1.1), la 6 phosphogluconate déshydrogénase<br />

(6PGD ; EC 1.1.1.44), l’isocitrate déshydrogénase (ICD ;<br />

EC 1.1.1.42), la phosphoglucoisomérase (PGI ; EC<br />

5.3.1.9), et la phosphoglucomutase (PGM ; EC 2.7.5.1),<br />

ont été analysés par électrophorèse horizontale sur gel<br />

Population Code Étages Pluviométrie Nature des sites<br />

population bioclimatiques annuelle<br />

et variantes* (mm/an)<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 23-37<br />

d’amidon à 13 % (Pasteur et al., 1987). Ces enzymes interviennent<br />

dans les voies métaboliques primaires chez les<br />

végétaux (voie des pentoses, cycles de Krebs et de<br />

Calvin…).<br />

Les protéines enzymatiques sont extraites à partir de<br />

chaque germination par broyage à froid (4 °C) dans<br />

100 ml de tampon ascorbate de sodium (0,1 mol.L-1 ;<br />

pH 8,4) additionné de 5 mM de ß mercaptoéthanol<br />

(0,035 %) et 8,35 % de saccharose. L’homogénat est centrifugé<br />

à 4 °C à 13 500 tours/mn pendant 30 mn.<br />

Les extraits sont déposés dans les puits du gel à l’aide<br />

de papier Wateman (1 x 0,5 cm) imbibé d’extrait. Un bac<br />

de glace est déposé sur le gel pour éviter la dénaturation<br />

des protéines. La migration, anodique, se déroule dans<br />

une chambre froide (4 °C) à ampérage constant (20 mA)<br />

et dure 12 heures. Le tampon de migration est à base<br />

d’histidine (0,065 mol.L-1 L-histidine et 0,02 mol.L-1 d’acide citrique, pH 6,8). L’intensité du courant et la durée<br />

de migration ont été fixées suite à des essais préliminaires<br />

pour optimiser la résolution des bandes électrophoriques.<br />

Dans ces manipulations nous avons fait varier l’ampérage<br />

(de 20 à 40 mA) et la durée de migration (de 6 à 12 h).<br />

La meilleure résolution des zymogrammes, dans nos<br />

conditions de travail, a été obtenue pour 20 mA et 12 h<br />

de migration.<br />

La révélation des isozymes a été effectuée selon les<br />

procédures préconisées par Stuber et al. (1988) et Wendel<br />

et Weeden (1989).<br />

Mazzouna 1 AI (hd) 197 steppe défrichée et replantée en cactus inerme.<br />

Skira 1 2 AI (hd) 180 champ abandonné et réservé à des cultures céréalières (orge).<br />

Skira 2 3 AI (hd) 180 champ abandonné et réservé à des cultures céréalières (orge).<br />

Mareth 4 AI (hd) 190 steppe à Lygeum spartum dégradée par érosion éolienne et hydrique.<br />

Ben Guerdene 5 AI (hd) 181 terrain reboisé en Acacia ssp. et Prosposis ssp.<br />

Sidi Maklouf 6 AI (hd) 152 champ cultivé en céréales (orge) et Lathyrus ssp.<br />

Skira 3 7 AI (hd) 175 champ abandonné, réservé à des cultures céréalières (orge).<br />

El Houareb 1 8 AS (htd) 286 berges et lit d’oued.<br />

El Houareb 2 9 AS (hd) 286 terrain reboisé en Acacia ssp et Atriplex nummularia.<br />

Sbeitla 10 AS (hf) 282 ravins et lit d’oued.<br />

Sidi Bouzid 11 AS (hd) 247 steppe à Rantherium suavelens défrichée et replantée en cactus inerme.<br />

Baten 12 AS (hd) 254 terrain abandonné, réservé pour des cultures céréalières (orge).<br />

AI : Aride inférieur; AS : Aride supérieur; hd : Hiver doux ; htd : Hiver tempéré à doux ; hf : Hiver frais<br />

* Les étages et les variantes bioclimatiques ainsi que la pluviométrie sont issus de la carte bioclimatique de la Tunisie. Echelle 1/1 000 000 (Nabli, 1995).<br />

Tableau 1. Caractéristiques écologiques principales des 12 populations d’Argyrolobium uniflorum analysées en Tunisie.<br />

Table 1. Main ecological traits of the twelve Tunisian Argyrolobium uniflorum population analyzed.<br />

27


28<br />

◆ ZAOULI YOSR, BEN FADHEL NAJEH, BOUSSAID MOHA<strong>MED</strong><br />

Méthodes d’analyse des résultats<br />

Dans notre travail l’interprétation des zymogrammes<br />

n’a pas été confrontée à une analyse de génétique formelle.<br />

Pour chaque système isoenzymatique et par comparaison<br />

de tous les zymogrammes, nous avons émis des<br />

hypothèses sur le contrôle génétique de chaque isozyme<br />

selon les principes avancés par Pernès (1984), Pasteur et<br />

al. (1987),Wendel et Weeden (1989) et Hannan et Orick<br />

(2000). Nos interprétations génétiques ont également tenu<br />

compte de travaux antérieurs effectués sur des Fabaceaes<br />

fourragères et pastorales allogames telles que Hedysarum<br />

coronarium L. (Trifi et al., 1989), H. carnosum Desf.<br />

(Chatti, 1987) ou autogames telles que Medicago polymorpha<br />

L. (Salhi-Hanachi et al., 1998).<br />

Pour chaque système enzymatique, les locus identifiés<br />

sont désignés successivement par les chiffres 1,2,3..., selon<br />

leur vitesse de migration anodique. Les allèles, pour<br />

chaque locus, sont désignés par des lettres a, b, c…. selon<br />

leur vitesse de migration vers l’anode.<br />

Pour chaque population, les fréquences génotypiques<br />

ont été calculées. La matrice des données a été traitée par<br />

le programme informatique Biosys1 (Swofford &<br />

Selander, 1981). Pour chaque population nous avons<br />

estimé : les fréquences alléliques ; le nombre moyen d’allèles<br />

par locus polymorphe (Ap)<br />

1 L<br />

Ap = — ∑ ni<br />

L i=1<br />

où ni est le nombre d’allèles par locus polymorphe, et L<br />

le nombre total de loci polymorphes) ;<br />

le pourcentage de loci polymorphes (P)<br />

np<br />

P = — x 10<br />

Nt<br />

où np est le nombre de loci polymorphes et Nt le nombre<br />

total des loci) ;<br />

le taux moyen d’hétérozygotie observée (Ho) et attendue<br />

(He) sous l’équilibre de Hardy-Weinberg<br />

L<br />

H o = ∑ H ok<br />

i=1<br />

————<br />

L<br />

où H ok, le taux d’hétérozygotes au locus k, est égal à<br />

n<br />

∑P ij (i≠j),<br />

i, j=1<br />

L = le nombre des loci, n= nombre d’allèles et P ij = fréquence<br />

du génotype ij au locus k.<br />

La déviation par rapport à la panmixie pour chaque<br />

locus a été évaluée par un test χ 2 de Pearson (Dagnellie,<br />

1974). L’excès ou le déficit d’hétérozygotie ont été testés<br />

par les indices de fixation F de Wright (Wright, 1965).<br />

La structuration des populations est estimée par le calcul<br />

des indices de subdivision de Wright (1978). La formule<br />

de base utilisée pour le calcul des valeurs des différents<br />

indices, est 1-F IT = (1-F IS) (1-F ST) où F IT est le paramètre<br />

de différenciation globale de toutes les subdivisions,<br />

F IS (F IS=1-H o/H e) est le paramètre de différenciation dans<br />

les populations et F ST est le paramètre de différenciation<br />

entre les populations<br />

k 1 S 2<br />

F ST = ∑ — ∑ [qx(i) – ¯q (i)]<br />

i=1 S X=1<br />

où qx(i) est la fréquence relative de l’allèle i dans les subdivisions<br />

; ¯q (i) est la moyenne des fréquences alléliques<br />

de i ; k est le nombre d’allèles ; S est le nombre de subdivisions.<br />

Un test χ2 a été utilisé pour juger de la signification<br />

des différences des valeurs de FST.Pour chaque locus,<br />

la valeur du χ2 est calculée par la formule χ2 = 2N FST (K-<br />

1) à (K-1)(S-1) degrés de liberté ; N est le nombre total<br />

des individus, K est le nombre d’allèles par locus et S est<br />

le nombre des populations (Workman & Niswander,<br />

1970 ; Gao et al., 2000). Le nombre de migrants (Nm)<br />

par génération entre les populations est estimé par la formule<br />

Nm=[(1-FST)]/4 FST (Wright, 1951).<br />

Pour mieux visualiser la différenciation des populations,<br />

nous avons effectué, par le programme informatique<br />

SAS (SAS, 1990) une analyse en composantes principales<br />

(ACP) sur les fréquences alléliques au niveau des<br />

loci polymorphes chez toutes les populations.<br />

La divergence génétique entre les populations est estimée<br />

à travers les indices d’identité (I) non biaisées de Nei<br />

(1978)<br />

¯J xy<br />

I = ———<br />

√¯J x • ¯J y<br />

où ¯J x et ¯J y sont les moyennes arithmétiques de J x et J y probabilités<br />

d’identité de deux allèles tirés au hasard dans la<br />

même population et ¯J xy moyenne arithmétique de J xy probabilité<br />

d’identité de deux allèles pris au hasard dans la<br />

population x et l’autre dans la population y (Pernès, 1984).<br />

Un dendrogramme (UPGMA ; Unweighted Pair Group<br />

Method Using Arithmetic Averages), établi à partir de ces<br />

indices, a été construit pour l’ensemble des populations<br />

pour juger de leur similarité génétique.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


DIVERSITÉ GÉNÉTIQUE DES POPULATIONS NATURELLES DE L’ARGYROLOBIUM UNIFLORUM JAUB. ET SPACH… ◆<br />

RÉSULTATS<br />

Variabilité génétique intrapopulation<br />

Pour les six systèmes enzymatiques analysés, 24 loci<br />

ont été détectés. Douze d’entre-deux (Adh-1, Icd-2, Icd-3,<br />

Mdh-1, Mdh-2, Mdh-3, Pgm-1, Pgm-2, Pgi-1, Pgi-2, Pgi-<br />

3 et Pgi-8) se sont avérés polymorphes (tableau 2). Chaque<br />

locus, excepté le Pgi-3 (à 3 allèles) comporte deux allèles.<br />

Les fréquences alléliques sont variables selon les populations<br />

(tableau 2). Des allèles particuliers n’ont été détectés<br />

que dans un petit nombre de populations. C’est le cas des<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 23-37<br />

Populations<br />

allèles Adh1-b, Pgi 8-a et Pgm1-b. La présence de ces allèles<br />

(à des fréquences souvent supérieures à 15 %) n’est pas liée<br />

à des zones géographiques ou bioclimatiques données. En<br />

effet, des populations proches géographiquement et se développant<br />

dans la même zone bioclimatique peuvent présenter<br />

des allèles uniques différents (tableaux 1 et 2).<br />

Le nombre moyen d’allèles par locus polymorphe (Ap)<br />

varie de 1,4 (populations 2, 3 et 8) à 1,8 (populations 5,<br />

7, 11 et 12) (tableau 3). Les pourcentages de loci polymorphes<br />

sont relativement élevés. Ils varient de 41,7<br />

(populations 2, 3 et 8) à 75 % (populations 5, 7 et 12).<br />

Le pourcentage moyen de loci polymorphes est de<br />

Locus 1 2 3 4 5 6 7 8 9 10 11 12<br />

Adh-1<br />

a 0,700 1,000 1,000 1,000 1,000 0,500 1,000 0,500 1,000 1,000 1,000 0,850<br />

b 0,300 0,000 0,000 0,000 0,000 0,500 0,000 0,500 0,000 0,000 0,000 0,150<br />

Icd-2<br />

a 1,000 0,850 0,850 0,850 0,900 1,000 1,000 1,000 1,000 0,950 1,000 0,900<br />

b 0,000 0,150 0,150 0,150 0,100 0,000 0,000 0,000 0,000 0,050 0,000 0,100<br />

Icd-3<br />

a 0,000 0,250 0,150 0,150 0,150 0,000 0,250 0,200 0,000 0,050 0,050 0,100<br />

b 1,000 0,750 0,850 0,850 0,850 1,000 0,750 0,800 1,000 0,950 0,950 0,900<br />

Mdh-1<br />

a 0,200 0,000 0,000 0,000 0,000 0,150 0,100 0,000 0,050 0,000 0,000 0,250<br />

b 0,800 1,000 1,000 1,000 1,000 0,850 0,900 1,000 0,950 1,000 1,000 0,750<br />

Mdh-2<br />

a 0,500 0,100 0,350 0,850 0,300 0,550 0,850 0,450 0,350 0,650 0,600 0,550<br />

b 0,500 0,900 0,650 0,150 0,700 0,450 0,150 0,550 0,650 0,350 0,400 0,450<br />

Mdh-3<br />

a 0,300 0,100 0,350 0,850 0,300 0,400 0,750 0,450 0,300 0,650 0,600 0,300<br />

b 0,700 0,900 0,650 0,150 0,700 0,600 0,250 0,550 0,700 0,350 0,400 0,700<br />

Pgm-1<br />

a 1,000 1,000 1,000 0,700 0,900 1,000 0,900 1,000 0,450 1,000 0,550 1,000<br />

b 0,000 0,000 0,000 0,300 0,100 0,000 0,100 0,000 0,550 0,000 0,450 0,000<br />

Pgm-2<br />

a 0,000 0,000 0,000 0,300 0,100 0,000 0,100 0,000 0,550 0,000 0,450 0,000<br />

b 1,000 1,000 1,000 0,700 0,900 1,000 0,900 1,000 0,450 1,000 0,550 1,000<br />

Pgi-1<br />

a 0,150 0,000 0,000 0,000 0,250 0,000 0,050 0,000 0,000 0,150 0,050 0,150<br />

b 0,850 1,000 1,000 1,000 0,750 1,000 0,950 1,000 1,000 0,850 0,950 0,850<br />

Pgi-2<br />

a 0,850 1,000 1,000 1,000 0,750 1,000 0,950 1,000 1,000 0,850 0,950 0,850<br />

b 0,150 0,000 0,000 0,000 0,250 0,000 0,050 0,000 0,000 0,150 0,050 0,150<br />

Pgi-3<br />

a 0,150 0,000 0,000 0,000 0,125 0,000 0,025 0,000 0,000 0,075 0,025 0,075<br />

b 0,425 0,500 0,500 0,500 0,500 0,500 0,500 0,500 0,500 0,500 0,500 0,500<br />

c 0,425 0,500 0,500 0,500 0,375 0,500 0,475 0,500 0,500 0,425 0,475 0,425<br />

Pgi-8<br />

a 0,000 0,000 0,025 0,000 0,000 0,250 0,000 0,000 0,000 0,000 0,000 0,000<br />

b 1,000 1,000 0,975 1,000 1,000 0,750 1,000 1,000 1,000 1,000 1,000 1,000<br />

Tableau 2. Fréquences alléliques des 12 loci polymorphes détectés chez les 12 populations d’Argyrolobium uniflorum analysées.<br />

Table 2. Allelic frequencies of the 12 polymorphic loci detected in the twelve Argyrolobium uniflorum populations analyzed.<br />

29


30<br />

◆ ZAOULI YOSR, BEN FADHEL NAJEH, BOUSSAID MOHA<strong>MED</strong><br />

Moyenne d’hétérozygotie<br />

Populations Code Nomb. moyen Nomb. moyen % loci<br />

individus / allèles / locus polymorphes (P) Obs.* Att.**<br />

population polymorphe (Ap) (Ho) (He)<br />

Mazzouna 1 20 1,7 58,3 0,121 0,238<br />

(0,2) (0,083) (0,067)<br />

Skira 1 2 20 1,4 41,7 0,083 0,127<br />

(0,1) (0,083) (0,052)<br />

Skira 2 3 20 1,5 41,7 0,087 0,168<br />

(0,2) (0,083) (0,061)<br />

Mareth 4 20 1,6 58,3 0,083 0,202<br />

(0,1) (0,083) (0,056)<br />

Ben Guerdene 5 20 1,8 75,0 0,083 0,255<br />

(0,2) (0,083) (0,057)<br />

Sidi Maklouf 6 20 1,5 50 0,175 0,223<br />

(0,2) (0,112) (0,070)<br />

Skira 3 7 20 1,8 75,0 0,083 0,193<br />

(0,2) (0,083) (0,050)<br />

Houareb 1 8 20 1,4 41,7 0,167 0,197<br />

(0,1) (0,112) (0,072)<br />

Houareb 2 9 20 1,5 50,0 0,083 0,210<br />

(0,2) (0,083) (0,071)<br />

Sbitla 10 20 1,7 58,3 0,083 0,186<br />

(0,2) (0,083) (0,062)<br />

Sidi Bouzid 11 20 1,8 66,7 0,083 0,236<br />

(0,2) (0,083) (0,070)<br />

Baten 12 20 1,8 75,0 0,108 0,255<br />

(0,2) (0,085) (0,056)<br />

Moyenne 20 1,625 57,64 0,103 0,207<br />

* obs : moyenne d’hétérozygotie observée<br />

** att : moyenne d’hétérozygotie attendue sous l’équilibre de Hardy-Weinberg<br />

Tableau 3. Nombre moyen d’allèles par locus polymorphe (Ap), pourcentage de loci polymorphes (P) et taux d’hétérozygotie (Ho et He)<br />

chez les 12 populations d’Argyrolobium uniflorum analysées. Les écarts types figurent entre parenthèse.<br />

Table 3. Mean number of alleles by polymorphic locus (Ap), percentage of polymorphic loci (P) and observed (Ho) and expected (He) heterozygosities<br />

in Argyrolobium uniflorum populations. Standard deviations were indicated within parenthesis.<br />

57,64 % (tableau 3). Les moyennes d’hétérozygotie observée<br />

(Ho) et attendue (He) sous l’équilibre de Hardy-<br />

Weinberg sont respectivement 0,103 et 0,207. Les populations,<br />

considérées dans leur ensemble, ne sont pas à<br />

l’équilibre de Hardy-Weinberg (Ho 0,05) (tableau 4). La majorité<br />

des valeurs de F (indice de fixation de Wright pour<br />

chaque locus) sont supérieures à zéro, témoignant d’une<br />

proportion observée en hétérozygotes inférieure à l’attendue.<br />

L’excès d’hétérozygotie n’a été observé que pour<br />

les loci Pgi-3 (chez toutes les populations) et Adh-1<br />

(populations 6 et 8).<br />

La valeur moyenne de FIS (indice de fixation de Wright<br />

décrivant la subdivision à l’intérieur des populations) pour<br />

l’ensemble des populations et pour tous les loci est de<br />

0,488. Pour la majorité des loci (9/12), les valeurs de FIS sont positives et égales à 1 (absence totale d’hétérozygotes)<br />

(tableau 5).Trois valeurs fortement négatives dénotant<br />

d’un excès en héterozygotes ont été observées pour<br />

les loci Adh-1 (-0,731), Pgi-3 (-0,856) et Pgi-8 (-0,646).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


DIVERSITÉ GÉNÉTIQUE DES POPULATIONS NATURELLES DE L’ARGYROLOBIUM UNIFLORUM JAUB. ET SPACH… ◆<br />

Pop. 1 2 3 4 5 6 7 8 9 10 11 12<br />

Locus<br />

Adh-1 ns - - - - -1,00*** - -1,00*** - - - ns<br />

Icd-2 - 1,00*** 1,00*** 1,00*** 1,00*** - - - - 1,00*** - 1,00***<br />

Icd-3 - 1,00*** 1,00*** 1,00*** 1,00*** - 1,00*** 1,00*** - 1,00*** 1,00** 1,00***<br />

Mdh-1 1,00*** - - - - 1,00*** 1,00*** - 1,00*** - - 1,00***<br />

Mdh-2 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00***<br />

Mdh-3 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00*** 1,00***<br />

Pgm-1 - - - 1,00*** 1,00*** - 1,00*** - 1,00*** - 1,00*** -<br />

Pgm-2 - - - 1,00*** 1,00*** - 1,00*** - 1,00*** - 1,00*** -<br />

Pgi-1 1,00*** - - - 1,00*** - 1,00** - - 1,00*** 1,00** 1,00***<br />

Pgi-2 1,00*** - - - 1,00*** - 1,00** - - 1,00*** 1,00** 1,00***<br />

Pgi-3 -0,38*** -1,00*** -1,00*** -1,00*** -0,69*** -1,00*** -0,91*** -1,00*** -1,00*** -0,77*** -0,91*** -0,77***<br />

Pgi-8 - - ns - - 0,73*** - - - - - -<br />

***. Significatif à P


32<br />

◆ ZAOULI YOSR, BEN FADHEL NAJEH, BOUSSAID MOHA<strong>MED</strong><br />

Pour mieux expliquer l’origine de ce déficit en hétérozygotes,<br />

nous avons calculé des indices de fixation<br />

moyen par population et dans trois régions géographiques<br />

(R1 : région de Gabès, R2 : région de Kairouan et R3 :<br />

région de Médenine). Chaque région regroupe des populations<br />

peu éloignées (sur un rayon de 20 km). En outre,<br />

le F ST pour l’ensemble des populations subdivisées en six<br />

sous-groupes [(populations 2, 3 et 7), (populations 4, 5<br />

et 6), (populations 8, 9 et 12), (population 1), (population<br />

10), (population 11)] a été également calculé. Au sein<br />

de chaque sous-groupe, les populations ont été rassemblées<br />

arbitrairement selon leur proximité géographique<br />

(figure 1 et tableau 5).<br />

Le FIS moyen par population (FIS= 1 – Ho/He) est élevé<br />

pour la majorité des populations et varie de 0,416 à 0,674.<br />

Deux valeurs, 0,215 et 0,152, relativement faibles, ont été<br />

observées respectivement pour les populations 6 et 8. Les<br />

FST moyens, selon les régions géographiques, sont de<br />

0,162 pour R1 (populations 2, 3 et 7), 0,147 pour R2<br />

(populations 8, 9 et 12) et 0,161 pour R3 (populations<br />

Axes Valeur propre Proportion (%) cumulé (%) Variables les plus corrélées aux axes de l’ACP<br />

Axe 1 6,913 27,65 27,65 Pgi-1a, Pgi-1b, Pgi-2a, Pgi-2b,<br />

Axe 2 6,118 24,47 52,13 Adh-1a, Adh-1b, Pgm-1b, Pgm-2a, Pgm-2b, Pgi-8a, Pgi-8b<br />

Axe 3 5,052 20,20 72,33 Icd-2a, Icd-2b, Icd-3a, Icd-3b, Mdh-1a, Mdh-1b, Mdh-2a,<br />

Mdh-3a, Mdh-3b<br />

Tableau 6. Analyse en composantes principales (ACP) : absorption de la variabilité et définition des trois premiers axes de l’ACP.<br />

Table 6. Principal component analysis: eigenvalues, proportions and axis definition.<br />

Figure 2. Analyse en composantes principales effectuée sur les fréquences alléliques des 12 loci polymorphes chez toutes les populations.<br />

A. Analyse selon le plan défini par les axes 1-2 ; B. Analyse selon le plan défini par les axes 1-3 (* = populations).<br />

Figure 2. Principal component analysis performed on allelic frequencies for the twelve populations for the 12 polymorphic loci<br />

A. Plot according to axis 1-2 ; B. Plot according to axis 1-3 (* = populations).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


DIVERSITÉ GÉNÉTIQUE DES POPULATIONS NATURELLES DE L’ARGYROLOBIUM UNIFLORUM JAUB. ET SPACH… ◆<br />

Populations code 1 2 3 4 5 6 7 8 9 10 11 12<br />

Mazzouna 1 **** 0,959 0,980 0,917 0,981 0,988 0,950 0,985 0,921 0,977 0,935 1,000<br />

Skira 1 2 **** 0,991 0,875 0,983 0,935 0,902 0,954 0,927 0,935 0,908 0,967<br />

Skira 2 3 **** 0,936 0,990 0,964 0,957 0,976 0,939 0,980 0,948 0,989<br />

Mareth 4 **** 0,919 0,917 0,992 0,926 0,930 0,972 0,985 0,931<br />

Ben Guerdene 5 **** 0,942 0,941 0,958 0,945 0,976 0,949 0,987<br />

Sidi Maklouf 6 **** 0,942 0,992 0,903 0,958 0,922 0,978<br />

Skira 3 7 **** 0,950 0,909 0,991 0,967 0,964<br />

Houareb 1 8 **** 0,909 0,964 0,930 0,977<br />

Houareb 2 9 **** 0,915 0,988 0,923<br />

Sbitla 10 **** 0,961 0,984<br />

Sidi Bouzid 11 **** 0,940<br />

Baten 12 ****<br />

Tableau 7. Indices de similarité de Nei (1978) pour l'ensemble des populations d’Argyrolobium uniflorum analysées.<br />

Table 7. Nei’s (1978) unbiased genetic identity calculated among populations of Argyrolobium uniflorum.<br />

4, 5 et 6). Elles indiquent une structuration substantielle<br />

entre les populations dans chaque région (tableau 5). Le<br />

F ST moyen pour les 6 sous-groupes de populations est<br />

faible (F ST = 0,073) et témoigne d’une faible structuration<br />

régionale de la variabilité.<br />

La structuration des populations, évaluée par l’indice<br />

de subdivision de Wright F ST calculé toutes les populations<br />

(12 populations), est significative. Pour tous les locus<br />

les valeurs de F ST sont significativement différentes de<br />

zéro (tableau 5). La valeur moyenne de F ST est de 0,163,<br />

c’est-à-dire que 16,3 % de la variance génétique se trouve<br />

entre les populations.<br />

L’analyse en composantes principales (ACP), effectuée<br />

sur les fréquences alléliques pour tous les loci au niveau<br />

de l’ensemble des populations, a montré que les trois premières<br />

composantes absorbant 72,33 % de l’inertie totale<br />

(tableau 6). L’axe 1 (27,65 % de la variation) est défini<br />

par les locus Pgi-1 et Pgi-2. L’axe 2 (24,47 % de la variation)<br />

est corrélé aux loci Adh-1, Pgm-1, Pgm-2 et Pgi-8.<br />

L’axe 3 (20,20 % de la variation) est défini par les loci Icd-<br />

2, Icd-3 Mdh-1, Mdh-2 et Mdh-3. La projection des populations,<br />

sur le plan défini par les axes 1 et 2, révèle une<br />

forte structuration des populations selon l’axe 1 (figure<br />

2A). Le regroupement des populations selon leur proximité<br />

géographique ou leur appartenance à un même étage<br />

bioclimatique n’est observée que pour un nombre limité<br />

d’entre elles. La projection des populations selon le plan<br />

défini par les composantes 1 et 3 montre une structuration<br />

des populations globalement similaire à celle obtenue<br />

précédemment selon les axes 1-2, avec toutefois un regrou-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 23-37<br />

pement préférentiel de quatre populations (7, 11, 4, 9) de<br />

l’aride supérieur du côté négatif de l’axe 2 (figure 2B).<br />

Les indices d’identité génétique de Nei (1978) (I) entre<br />

les populations varient de 0,917 (populations 1 et 11) à<br />

1,000 (populations 1 et 12) (tableau 7). Toutefois, il n’y<br />

a pas de corrélation évidente entre ces valeurs, la proximité<br />

géographique des populations (cas des populations<br />

1-12, 8-9) ou l’étage bioclimatique (populations 1, 7 et<br />

11 par exemple). Le dendrogramme (figure 3) établi à<br />

partir de ces indices fait ressortir, à I = 0,940, deux grands<br />

groupes de populations :<br />

— le premier agrégat réunit des populations d’origines<br />

géographique et bioclimatique variables. Il peut être subdivisé<br />

(à I = 0,970) en deux sous-ensembles, comprenant<br />

respectivement quatre (1,12, 6 et 8) et trois (2, 3 et 5)<br />

populations (figure 3). Les populations du premier sousensemble<br />

se développent dans des sites très perturbés<br />

(terres de culture, oueds) et n’appartiennent pas au même<br />

étage bioclimatique. Celles du second sous-ensemble, géographiquement<br />

proches, appartiennent toutes à l’étage<br />

aride inférieur.<br />

— le second groupe est formé de 5 populations rassemblées<br />

en deux sous-groupes. Le premier est constitué<br />

de trois populations (4, 7 et 10) dont deux appartiennent<br />

à l’étage aride inférieur tandis que le second est formé de<br />

deux populations de l’aride supérieur (9 et 11).<br />

Au sein de chaque agrégation (premier ou second agrégat),<br />

l’éloignement géographique des populations ne<br />

semble pas jouer un rôle prédominant dans le regroupement<br />

des populations.<br />

33


34<br />

◆ ZAOULI YOSR, BEN FADHEL NAJEH, BOUSSAID MOHA<strong>MED</strong><br />

DISCUSSION<br />

L’analyse du polymorphisme isoenzymatique chez<br />

l’Argyrolobium uniflorum, espèce cléistogame, a révélé une<br />

diversité génétique intrapopulation importante. Le nombre<br />

moyen d’allèles par locus polymorphe (Ap = 1,625), le<br />

pourcentage de loci polymorphes (P = 57,64) et les taux<br />

d’hétérozygotie observée (H o = 0,103) et attendue (He =<br />

0,207) sous l’équilibre de Hardy-Weinberg sont plus élevés<br />

que ceux rapportés classiquement chez des taxons<br />

autogames voire cléistogames (Ap = 1,31 ; P = 20 % ;<br />

H e = 0,074 ; Ho = 0,159) (Hamrick & Godt, 1989 ;<br />

Matolweni et al., 2000 ; Linert et al., 2002). L’importante<br />

diversité intrapopulation indiquerait que la dégradation<br />

des sites n’a pas encore diminué la variabilité de l’espèce.<br />

Cette forte variation résulterait à la fois du niveau élevé<br />

de la diversité des populations ancestrales et de migration<br />

de graines (pouvant être faible mais continue au cours du<br />

temps) à partir de populations adjacentes. Aucune de ces<br />

hypothèses ne peut être rejetée pour le moment. Un marquage<br />

moléculaire (par des microsatellites, par exemple)<br />

de populations de petite taille (source et puits) (Wright,<br />

1978 ; Pernès, 1984), des mesures des flux potentiel et<br />

efficace des graines entre elles, et un suivi de la prolifération<br />

de la descendance au cours du temps devraient être<br />

réalisés pour tester ces hypothèses. Une grande variabilité<br />

génétique intrapopulation a été rapportée chez des<br />

espèces autogames, voire cléistogames, se développant sur<br />

des sites perturbés (Lewis & Crawford, 1995 ; Mc Donald<br />

& Hamrick, 1996 ; Evans et al., 2000 ; Linert et al., 2002).<br />

Figure 3. Dendrogramme établi à partir des indices<br />

d'identité génétique de Nei (1978) pour l'ensemble<br />

des populations analysées.<br />

Figure 3. Dendrogram based on Nei's (1978)<br />

unbiased genetic identities for all populations.<br />

* Populations<br />

** Étages bioclimatiques<br />

AI : aride inférieur,<br />

AS : aride supérieur<br />

Les populations d’Argyrolobium uniflorum se reproduisent<br />

sous un régime non panmictique (H o < H e). Le<br />

F IS moyen au sein des populations est élevé. Pour la majorité<br />

des loci polymorphes détectés (9/12), les valeurs des<br />

indices de fixation F de Wright sont égales à 1,000, indiquant<br />

une déviation totale par rapport à l’équilibre de<br />

Hardy-Weinberg par absence d’hétérozygotes (tableau 4).<br />

Le test χ 2 appliqué à ces indices révèle des différences<br />

hautement significatives (P < 0,001) par rapport à l’équilibre<br />

panmictique. Seuls trois loci Adh-1 (populations 1,<br />

6 et 8), Pgi-3 (pour toutes les populations) et Pgi-8 (populations<br />

3 et 6) ont montré une présence et un excès en<br />

hétérozygotes.<br />

Le niveau élevé d’homozygotie observé ne peut résulter<br />

d’un biais d’échantillonnage parce qu’il est révélé chez<br />

toutes les populations, en dépit de leur éloignement géographique<br />

et que notre analyse, pour une population donnée,<br />

a porté sur des individus collectés loin les uns des<br />

autres et dans une grande aire (> 3 ha). Il est dû au mode<br />

de reproduction cléistogame de l’espèce puisque le F ST<br />

calculé pour l’ensemble des populations subdivisées en<br />

six sous-groupes est faible (0,073). Nos résultats rappellent<br />

ceux observés par Tarayre et Thompsom (1997) chez<br />

Thymus vulgaris L. (espèce gynodioïque). La faible différenciation<br />

entre les subdivisions et la forte consanguinité<br />

des populations résulteraient du régime de reproduction<br />

de l’espèce.<br />

L’existence d’hétérozygotie chez Argyrolobium uniflorum,<br />

pour certains loci (3/12), peut trouver son origine<br />

suite à des mutations ayant affecté certains loci. L’excès<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


DIVERSITÉ GÉNÉTIQUE DES POPULATIONS NATURELLES DE L’ARGYROLOBIUM UNIFLORUM JAUB. ET SPACH… ◆<br />

d’hétérozygotes observé pour les loci Adh-1, Pgi-3 et Pgi-<br />

8 pourrait correspondre à un équilibre sélectif pour promouvoir<br />

certains loci afin de maintenir une hétérozygotie<br />

adaptative à certaines pressions du milieu (surpâturage,<br />

par exemple) (Falk & Holsinger, 1991).<br />

La cléistogamie de l’Argyrolobium uniflorum assure non<br />

seulement le maintien de la stabilité de la structure génétique<br />

des populations, mais permet aussi la pérennité des<br />

populations par la production des graines en dépit des<br />

conditions défavorables du milieu aride (irrégularité des<br />

facteurs physiques, rareté des pollinisateurs, etc.).<br />

Toutefois, l’homozygotie contribue à diminuer l’aptitude<br />

de résistance des individus aux variations du milieu et<br />

augmenterait le risque d’extinction des populations<br />

(Ellstrand & Elam, 1993 ;Young et al., 1996). Le maintien<br />

d’un niveau suffisant de variabilité génétique à l’intérieur<br />

des populations serait donc tributaire de l’ampleur<br />

du flux migratoire des graines et de la survie de leur descendance.<br />

Il est important de signaler qu’en milieu aride<br />

très perturbé de nombreux facteurs, dont le défrichement<br />

des parcours par des labours profonds, le surpâturage et<br />

les fortes érosions éolienne et hydrique, favorisent la dispersion<br />

des graines. Ces facteurs sont paradoxalement<br />

néfastes pour le milieu physique et l’installation des populations.<br />

Une différenciation substantielle entre les populations<br />

(F ST = 0,163) et une faible structuration régionale (F ST =<br />

0,073) sont observées. La différenciation des populations<br />

résulterait donc de la cléistogamie de l’espèce. La destruction<br />

continuelle des habitats contribue à réduire les<br />

flux géniques entre les populations. Saunders et al. (1999),<br />

Evans et al. (2000), Matolweni et al. (2000), Linert et al.<br />

(2002) ont rapporté des résultats similaires chez des<br />

espèces cléistogames ou autogames dont les populations<br />

ont été perturbées par les pressions anthropiques. L’analyse<br />

en composantes principales illustre bien la différenciation<br />

des populations. Celles-ci, selon les deux types de<br />

représentation (axes 1-2 ou axes 1-3), sont fortement dispersées<br />

sur les plans définis par les trois premiers axes de<br />

l’ACP. Leur regroupement selon la proximité géographique<br />

ou l’étage bioclimatique ne s’est opéré que pour<br />

un nombre limité d’entre elles.<br />

Une grande similarité génétique entre les populations<br />

d’Argyrolobium uniflorum est observée. Les indices d’identité<br />

génétique de Nei (I) sont très élevés et la plus faible<br />

valeur de I est de 0,915 entre les populations 9 (Houareb<br />

2) et 10 (Sbeitla) distantes de 70 km. Ceci indique une<br />

grande similarité génétique entre les populations tunisiennes<br />

d’Argyrolobium uniflorum. La variation des valeurs<br />

de I n’est pas toujours corrélée à la proximité géogra-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 23-37<br />

phique des populations (cas des populations 3 et 7 ; 8 et<br />

9). Ceci veut dire que ce critère ne doit pas toujours être<br />

considéré comme un bon prédicateur de la diversité génétique<br />

interpopulations (Godt & Hamrick, 1993 ;<br />

Mc Donald & Hamrick, 1996). Des faibles divergences<br />

génétiques, entre populations d’espèces cléistogames ou<br />

autogames, à habitats peu ou pas perturbés, sont rarement<br />

rapportées dans la littérature. En effet, ces caractéristiques<br />

semblent plutôt caractériser des espèces allogames<br />

(Bartlett et al., 2002).<br />

Le dendrogramme, établi à partir des indices d’identité<br />

génétique de Nei, a montré que le regroupement des<br />

populations s’opère avec des indices élevés (I comprise<br />

entre 0,930 et 1,000). Il peut être subdivisé à I = 0,940<br />

en deux sous-ensembles, sans corrélation franche entre le<br />

regroupement des populations, l’étage bioclimatique et la<br />

proximité géographique. En effet, l’agrégation des populations<br />

selon ces deux facteurs n’a été réalisé que pour un<br />

nombre limité d’entre elles.<br />

CONCLUSION<br />

L’analyse de la diversité génétique chez Argyrolobium<br />

uniflorum a montré que le polymorphisme intrapopulation<br />

est élevé et que certaines populations possèdent des<br />

allèles rares. Les populations sont fortement différenciées<br />

et présentent de faibles divergences génétiques entre elles.<br />

La forte structuration des populations et le niveau élevé<br />

de leur similarité génétique témoigneraient d’une fragmentation<br />

récente des populations sous l’influence des<br />

pressions anthropiques (surpâturage et défrichement).<br />

Ces facteurs, en milieu aride, jouent des rôles antagonistes<br />

mais importants dans la dynamique des populations. Le<br />

défrichement et le surpâturage détruisent le milieu et isolent<br />

progressivement les populations. Toutefois, la détérioration<br />

des habitats favorise l’action du vent et des eaux<br />

de ruissellement qui dispersent les graines. La migration<br />

des graines contribuerait à élever le niveau du polymorphisme<br />

intrapopulation.<br />

Les programmes de conservation de l’espèce ne peuvent<br />

s’appuyer uniquement sur l’analyse du polymorphisme<br />

isoenzymatique qui peut masquer d’autres variations<br />

pouvant être liées à des adaptations spécifiques à<br />

certaines conditions du milieu (physiologiques, morphologique<br />

et autres). Une estimation du polymorphisme<br />

moléculaire (AFLP, microsatellites, etc.) et une analyse<br />

plus précise de la dynamique des populations (distances<br />

de dispersion de semences, estimation de la taille effec-<br />

35


36<br />

◆ ZAOULI YOSR, BEN FADHEL NAJEH, BOUSSAID MOHA<strong>MED</strong><br />

tive des populations…) permettraient d’affiner notre<br />

étude.Toutefois, notre travail sur le polymorphisme enzymatique<br />

de l’Argyrolobium uniflorum constitue déjà un<br />

point de départ en vue de la conservation de l’espèce. Une<br />

priorité doit être accordée à la préservation des populations<br />

ayant présenté un niveau élevé de polymorphisme<br />

et des allèles rares. La conservation, ex-situ, doit s’appuyer<br />

sur la collecte d’un maximum d’individus à l’intérieur<br />

des populations et non entre les populations.<br />

Références bibliographiques<br />

ABDELKEFI A., BOUSSAID M. & MARRAKCHI M., 1990. Erosion<br />

génétique et aridité. Cas de trois légumineuses pastorales,<br />

susceptibles d’intervenir dans l’amélioration des parcours en<br />

zones arides. Biologia 1:91-98.<br />

ATTIA AH., 1986. Guide sur les parcours en zones arides.Programme<br />

ceinture verte de l’Afrique du Nord, Eds. ALECSO, Tunis,<br />

120 p.<br />

BARTLETT E., NOVAK S. J. & MACK R. N., 2002. Genetic variation<br />

in Bromus tectorum (Poaceae): Differentiation in the<br />

Eastern United State. Am. J. Bot. 89: 602-613.<br />

BATISTA F., BANARES A., CAUJAPÉ J., CARQUÉ E., MARRERO M.<br />

& SOSA P.A., 2001. Allozyme diversity in three endemic species<br />

of Cistus from the Canary islands: Infraspecific and interspecific<br />

comparison and implications for genetic conservation.<br />

Am. J. Bot. 88: 1582-1592.<br />

BEN FADHEL N., ZAOUALI Y. & BOUSSAID M., 2000. Aptitudes<br />

de deux légumineuses Pastorales (Hedysarum carnosum Desf.<br />

et Argyrolobium uniflorum Jaub. et Spach.) à la réhabilitation<br />

des parcours en milieu aride de la Tunisie, Cah.Opt.Med. 45 :<br />

95-98.<br />

BOUSSAID M., BEN FADHEL N., CHEMLI R. & BEN M’HA<strong>MED</strong> M.,<br />

1998. Structure of vegetation in Northern and Central Tunisia<br />

and protective measures, Cah. Opt. Med. 38: 295-302.<br />

CHATTI W.S., 1987. Analyse de la diversité génétique basée sur<br />

les caractères morphologiques et le polymorphisme enzymatique<br />

des espèces Hedysarum coronarium L. et<br />

Hedysarum carnosum Desf. Relations phylogénétiques avec<br />

le complexe Hedysarum spinosissimum L. Thèse 3e cycle.<br />

FS Tunis, 126 p.<br />

DAGNELLIE P., 1974. Théorie et méthodes statistiques. Vol. 2. Les<br />

Presses agronomiques de Gembloux (Belgique), 451 p.<br />

DARDOUR Z., BOUSSAID M. & MARRAKCHI M., 1989. Contribution<br />

à la connaissance de l’Argyrolobium uniflorum Jaub. et<br />

Spach. en Tunisie. 3e Journées de la Société tunisienne de Chimie<br />

biologique, Sousse : 3-4 avril 1989.<br />

ELLSTRAND N. C. & ELAM D. R., 1993. Population genetic consequences<br />

of small population size: Implications for plant<br />

conservation. Annu. Rev. Ecol. Syst. 24: 217-242.<br />

EMBERGER L., 1955. Une classification biogéographique des climats.<br />

Rev.Trav.Lab.Géol.Bota.et Zool.,Fac.Sci.Montpellier 7:<br />

3-43.<br />

EVANS M. E. K., DOLAN R. W., MENGENS E. S. & GORDON D.<br />

R., 2000. Genetic diversity and reproductive biology in Warea<br />

carteri (Brassicaceae), a narrowly endemic Florida scrub<br />

annual. Am. J. Bot. 87: 372-381.<br />

FALK D.A. & HOLSNIGER K.E., 1991. Genetics and conservation<br />

of rare plants. Oxford University Press, Oxford, UK.<br />

FLORET C. & PONTANIER R., 1982. L’aridité en Tunisie pré-saharienne<br />

: climat-sol-végétation et aménagement.Thèse de doctorat<br />

d’État, USTL, Montpellier. Travaux et documents de<br />

l’ORSTOM, n°150, 544 p.<br />

GAO LI,SONG GE & HONG DE-YUAN., 2000. Low levels of genetic<br />

diversity within populations and high differentiation among<br />

populations of wild Rice (Oryza granulata. Nees et Arn. Ex<br />

watt.) from China. Intern. J. Plant Sci. 161: 691-697.<br />

GOUNOT M., 1995. La végétation des parcours. In : Essai de synthèse<br />

sur la végétation et la phytoécologie tunisiennes. Vol. 5 et<br />

6. Eds. de l’IOR Tunisienne, p. 298-307.<br />

GODT M. J. & HAMRICK J. L., 1993. Genetic diversity and population<br />

structure in Tradescantia hirsuticaulis (Commelionaceae).<br />

Am. J. Bot. 80: 959-966.<br />

HAMRICK J. L. & GODT M. J., 1989. Allozyme diversity in plant<br />

species. In: Brown A.H.D., Clegg M.T., Kahler A.L., Weir<br />

B. S. (eds.), Plant population genetics, breeding and genetic<br />

ressources, Sinauer Associates, Sunderland, Mass. A.,<br />

p. 43-63.<br />

HANNAN G. L. & ORICK M. W., 2000. Isozyme diversity in Iris<br />

cristata and the threatened glacial endemic I. lacustris (iridiaceae).<br />

Am. J. Bot. 87:153-168.<br />

HEDRICK P.W., 2001. Conservation genetics: where are we now?<br />

Trends Ecol. Evol. 16: 629-636.<br />

LE FLOC’H E., 1995. Les écosystèmes des zones arides du Nord<br />

de l’Afrique : orientations pour l’établissement d’un réseau<br />

de Réserves de Biosphère. In : Essai de synthèse sur la végétation<br />

et la phytoécologie tunisiennes.Vol. 5 et 6. Eds. de l’IOR<br />

Tunisienne, p. 490-507.<br />

LE HOUÉROU H. N., 1969. La végétation de la Tunisie steppique<br />

(avec références au Maroc, en Algérie et la Libye). Ann. Inst.<br />

Nat. rech. Agr.Tunisie 42, 642 p.<br />

LE HOUÉROU H. N. & IONESCO T., 1973. Appétabilité des espèces<br />

végétales de la Tunisie steppique. Doc. travail ronéoté, 68 p.<br />

LEWIS P. O & CRAWOFORD D. J., 1995. Pleistocene refugium<br />

endemics exihit greater allozymic diversity than widespread<br />

congeners in the genus Polygonella (Polygonaceae). Am. J.<br />

Bot. 82: 141-149.<br />

LINERT J., FISHER M., SCHORELLER J. & DIEMER M., 2002.<br />

Isozyme variability of the Wetland. Swertia perennis (Gentianaceae)<br />

in relation to habitat size, isolation, and plant fitness.<br />

Am. J. Bot. 89: 801-811.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


DIVERSITÉ GÉNÉTIQUE DES POPULATIONS NATURELLES DE L’ARGYROLOBIUM UNIFLORUM JAUB. ET SPACH… ◆<br />

LWOGO A. B., 1984. Évaluation du plasma germinatif des pâturages<br />

tropicaux : stratégie et conceptions expérimentales. In :<br />

Recherche sur l’amélioration des pâturages en Afrique orientale<br />

et australe. Comptes Rend. Atelier Harare (Zimbabwe, 17-<br />

21/9/1984).<br />

MATOLWENI L. O., BALKWILL K. & MC LELLANT T., 2000.<br />

Genetic diversity and gene flow in the morphologically<br />

variable, rare endemic Begonia dregei and Begonia homonyma<br />

(Begoniaceae). Am. J. Bot. 87: 431-440.<br />

MC DONALD D. B. & HAMRICK J. L., 1996. Genetic variation in<br />

some plants of Florida scrub. Am. J. Bot. 83: 21-27.<br />

NABLI M. A., 1995. Essai de synthèse sur la végétation et la phytoécologie<br />

tunisienne. Le milieu physique et la végétation écologie<br />

végétale appliquée. Eds. de l’IOR tunisienne, 542 p.<br />

NEI M., 1978. Estimation of average heterozygosity and genetic<br />

distance from a small number of individuals. Genetics 89:<br />

583-590.<br />

OZENDA P., 1977. Flore du Sahara. (2e éd.). CNRS, Paris, 622 p.<br />

PASTEUR N., PASTEUR G., BONHOMME F., CATALON J. & BRITTON<br />

D. J., 1987. Manuel technique de génétique par électrophorèse des<br />

protéines. Eds. Tec. et Doc. Lavoisier, Paris, 217 p.<br />

PERNÈS J., 1984. Gestion des ressources génétiques des plantes, tome<br />

II. Eds. Tec. et Doc. Lavoisier, Paris, 346 p.<br />

POTTIER-ALAPETITE G., 1979. Flore de la Tunisie : Angiospermes<br />

Dicotylédones, Dialypétales. Eds. de l’IOR Tunisienne, 654 p.<br />

QUÉZEL P., 1991. Structure de végétation et flore en Afrique du<br />

Nord : leurs incidences sur les problèmes de conservation.<br />

In : Conservation des ressources végétales. Actes Éditions, Rabat,<br />

Maroc, 19-34.<br />

QUÉZEL P. & S ANTA S., 1962. Nouvelle flore de l’Algérie et des<br />

régions désertiques méridionales, tome I. CNRS, Paris, 565 p.<br />

RAYNAUD C., 1974. Éléments pour une flore pratique du Maroc.<br />

Légumineuses, tribu des Génistées. Série systématique, Edit.<br />

ENI, Rabat, Maroc, 76 p.<br />

SALHI-HANNACHI A., BOUSSAID M. & MARRAKCHI M., 1998.<br />

Genetic variability organisation and gene flow in natural<br />

populations of Medicago polymorpha prospected in Tunisia.<br />

Gen. Sel. Evol. 30, 121-135.<br />

SAS., 1990. SAS User’s Guide : SAS STAT, SAS BASIC.Version<br />

6, fourth edition. SAS incl, Box 8000. Cary, NC 27512-8000,<br />

Cary: NC: SAS institut Inc.<br />

SAUNDERS D. A., HOBBES R. J. & MARGULES C. R., 1999.<br />

Biological consequences of ecosystem fragmentation: a review.<br />

Conserv. Biol. 5, 18-32.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 23-37<br />

SCHOENENBERGER A., 1995. La végétation forestière. In : Essai<br />

de synthèse sur la végétation et la phytoécologie tunisiennes. Vol.<br />

5 et 6. Eds. de l’IOR Tunisienne, p. 215-255.<br />

STUBER C.W.,WENDEL J. F., GOODMAN M. M. & SMITH J. S. C.,<br />

1988.Techniques and scoring procedures for starch gel electrophoresis<br />

of enzymes from maize (Zea mays L.). Technical<br />

Bulletin 286, North Carolina State University, Raleigh, NC.<br />

SWOFFORD D. L. & SELANDER R. B., 1981. Biosys-1: A computer<br />

program for the analysis of allelic variation in genetics.<br />

University of Illinois, Urbana (USA).<br />

TARAYRE M. & THOMPSON J.D., 1997. Population genetic structure<br />

of gynodioecious Thymus vulgaris L. (Labiaceae) in<br />

Southern France. J. Evol. Biol. 10 : 157-174.<br />

THIAULT M., 1995. La végétation pastorale et la végétation messicole<br />

de la Tunisie tellienne. In : Essai de synthèse sur la végétation<br />

et la phytoécologie tunisiennes. Vol. 5 et 6. Eds. de l’IOR<br />

Tunisienne, p. 298-307.<br />

TRIFI-FARAH N., CHATTIW-S., MARRAKCHI M. & PERNÈS J, 1989.<br />

Déterminisme génétique de dix systèmes enzymatiques chez<br />

Hedysarum coronarium L. Agronomie 9:503-510.<br />

WAECHTER P, 1995. Comportement alimentaire au pâturage des<br />

animaux domestiques (ovins et caprins) en Tunisie aride. In :<br />

Essai de synthèse sur la végétation et la phytoécologie tunisiennes.<br />

Vol. 5 et 6. Eds. de l’IOR Tunisienne, p. 431-455.<br />

WENDEL J. & WEEDEN N., 1989.Visualization and interpretation<br />

of plant isozymes. In: D. Soltis & P. Soltis (eds.), Isozymes in<br />

plant biology, Advances in plant Sciences Series, vol. 4, 5-45,<br />

Dioscorides Press, Portland, USA.<br />

WORKMAN P. L. & NISWANDER J. D., 1970. Population studies<br />

on Southwestern Indian Tribes II. Local genetic differentiation<br />

in Papago. Am. J. Hum. Genet. 22: 24-49.<br />

WRIGHT S., 1951. The genetical structure of populations. Ann.<br />

Eugen. 15: 323-354.<br />

WRIGHT S., 1965.The interpretation of population structure by<br />

F-Statistics with special regard to systems of mating. Evolution<br />

19: 395-420.<br />

WRIGHT S., 1978. Evolution and the genetics of populations.Vol. 4:<br />

Var iability within and among natural populations. University<br />

of Chicago Press, Chicago, Illinois, USA.<br />

YOUNG A.G., BOGLE T. & BROWN T., 1996.The population genetic<br />

consequences of habitat fragmentation for plants. Trends<br />

Ecol. Evol. 11: 413-418.<br />

ZAOUALI Y., 1999. Polymorphisme isoenzymatique de douze<br />

populations naturelles d’Argyrolobium uniflorum Jaub. et<br />

Spach. en Tunisie. Mémoire de DEA. Fac. Sci. Tunis, 78 p.<br />

37


Browsing and nutritive value of the most common range species<br />

in Matruh area, a coastal <strong>Mediterranea</strong>n region, Egypt<br />

Broutage et valeur nutritive des espèces<br />

de parcours les plus communes dans la région de Matruh,<br />

une zone côtière méditerranéenne en Égypte<br />

S. Z. Heneidy<br />

Botany Department, Faculty of Science, University of Alexandria, Egypt.<br />

Abstract<br />

The grazing system in Matruh area is studied during the wet and<br />

dry seasons during two years. The study reveals that the area is occupied<br />

with 39 perennial and 43 annual species, Chamaephytes are<br />

the most dominant life-forms. The area (which include 32 sampling<br />

sites) composed of 5 different habitats. About 38 % of the<br />

studied sites are suffering from high grazing pressure. In addition<br />

66% of the habitats are found to be good pastures. Moreover, based<br />

on abundance and palatability, 5 to 7 species are considered as<br />

indicator (key) species in each site. Atriplex halimus and<br />

Asphodelus ramosus are the most common ones. Further about<br />

51 % of the perennial species are highly palatable.<br />

Grazing value was high for 69 and 49 % for the perennials and<br />

annuals species respectively. Therefore, the pasture condition and<br />

the quality in the studied area can be considered good. The average<br />

level of the digestible crude protein (DCP, 5.7 %), total digestible<br />

nutrients (TDN, 67 %), gross energy (GE, 3,976 kcal<br />

kg -1DM), digestible energy (DE, 3,274 kcal kg -1DM), and metabolizable<br />

energy (ME, 2,824 kcal kg -1DM), indicates that the present<br />

pasture is able to meet the feed demands of the grazing animals<br />

and these animals receive most of their energy requirements.<br />

The energy content is however equivalent 0.84 of the Scandinavian<br />

feed unit.<br />

Key-words<br />

Nutritive value, indicator species, pasture condition, Matruh area,<br />

Egypt.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 39-49<br />

Résumé<br />

Le système de pâturage a été étudié durant les périodes humides et<br />

sèches, pendant deux années, dans la zone de Matruh, en Égypte.<br />

Notre travail montre que cette zone est occupée par trente-neuf<br />

espèces pérennes et quarante-trois annuelles. Les chaméphytes sont<br />

les types biologiques dominants. La zone (qui comprend trente-deux<br />

sites d’échantillonnage) est composée de cinq habitats différents.<br />

38 % du site d’étude souffrent de surpâturage, mais 66 % des habitats<br />

peuvent être considérés comme de bonnes pâtures, 51 % des<br />

espèces pérennes sont fortement consommables. À partir de leur<br />

abondance et de leur appétence, 5 à 7 espèces ont été considérées<br />

comme espèces indicatrices (espèces clés) pour chaque site. Atriplex<br />

halimus et Asphodelus ramosus en sont les plus communes.<br />

La valeur nutritive pour le pâturage est forte, avec respectivement<br />

69 % et 49 % pour les espèces pérennes et les annuelles. Par conséquent,<br />

les conditions de pâturage et la qualité alimentaire de la<br />

zone étudiée peuvent être considérées comme bonnes. Le niveau<br />

moyen de digestibilité des matières azotées totales (DCP, 5.7 %),<br />

des cendres solubles (TDN, 67 %), de l’énergie totale (GE,<br />

3.976 Kcal kg -1 DM), de l’énergie digestible (DE 3.274 Kcal<br />

kg -1DM) et de l’énergie métabolisable (ME, 2.824 Kcal kg -1 DM),<br />

indique que l’état actuel des pâtures est capable de satisfaire la<br />

demande alimentaire des animaux au pâturage et que ces animaux<br />

reçoivent la plupart de leurs besoins énergétiques. La teneur en énergie<br />

est cependant équivalente à 0.84 de l’unité scandinave d’alimentation.<br />

Mots-clés<br />

Valeur nutritive, espèces indicatrices, conditions de pâturage, région<br />

de Matruh, Égypte.<br />

39


40<br />

◆ S. Z. HENEIDY<br />

INTRODUCTION<br />

The coastal <strong>Mediterranea</strong>n region, particularly the arid<br />

and semi-arid parts of Africa, has a long history of intensive<br />

land-use, mainly for grazing and rainfed farming.<br />

According to Duivenbooden (1985), rangeland, the major<br />

feed resources, occupies an estimated 90 % of the total<br />

area of such coastal zone. In this zone many areas are<br />

dominated by shrubs and sub-shrubs.The annuals occur<br />

only in a limited extent because of weather conditions and<br />

the increased animal density. According to Le Houérou<br />

(1980a), about 1.5 million heads of animals are normally<br />

present, and rangeland vegetation is insufficient throughout<br />

the year and therefore supplementary feed is a must.<br />

Subsequently sustainable development and improvement<br />

of the grazing system (including biomass, and nutritive<br />

values) are very important for continuity, particularly in<br />

the present <strong>Mediterranea</strong>n coastal area (e.g. Matruh).<br />

The present study is initiated to collect information<br />

about pasture in Matruh area which is one of the most<br />

richest area for grazing in the Egyptian coastal region. It<br />

is worthy to mention that in Egypt there are about 2.5<br />

million hectares of arid rangelands of the chamaephytic<br />

steppe type (Le Houérou, 1980a), 50 % of this area occured<br />

in its North west of the coastal region on the<br />

<strong>Mediterranea</strong>n sea.<br />

The objectives of this study are: (1) identification of<br />

range species and their life-forms, (2) estimation of the<br />

physical grazing characters of the rangeland species and<br />

assessment to their grazing value, and palatability (3)<br />

determination of the nutritive value of the plants in a purpose<br />

to evaluate the quality and quantity of the forage,<br />

and (4) focusing or throwing light on the role of the<br />

browse species in the grazing system (these species which<br />

are most effective in the range system, called key species).<br />

Key management species are however those on which the<br />

grazing management of a specific range is based.<br />

THE STUDY AREA<br />

The studied area is located about 200 km west of<br />

Alexandria. Its area extends to about 30 km southwords<br />

in the inland plateau and transversally to about 10 km<br />

from west to east parallel to the sea shore. It lies between<br />

the following longitudes and latitudes (30 o 30 - – 31 o 10 -N,<br />

27 o 30 - – 28 oE). The area includes different physiographic<br />

units leading to variety of habitats. Geomorphologically,<br />

it is a part of the Egyptian western desert and it<br />

is a plateau with numerous large and deep topographic<br />

depressions. Its soil is young, and essentially alluvial<br />

(Kamal, 1988).They are derived from two main sources<br />

(a)- the table land (inland plateau) composed essentially<br />

of limestone alternating with strata and shale; and (b)beach<br />

deposits composed of calcareous oolitic grains.The<br />

dominating land use is grazing by domestic animals.<br />

Climatologically, this area is classified as arid with mild<br />

winters and warm summers (UNESCO, 1977).<br />

METHODS<br />

The present study was carried out during the two climatic<br />

conditions (during two successive years) characterizing<br />

this area of the <strong>Mediterranea</strong>n region, the wet season<br />

(winter and spring) and dry season (summer and<br />

autumn).The study area was classified into five habitats:<br />

rocky plateau, flat plateau, rocky ridge, non-saline depression,<br />

and saline depression. At the beginning of the study<br />

several flocks of grazing animals (sheep, goats and camels)<br />

were observed several times in the studied sites (32 sites),<br />

to explore which species is preferred following the<br />

methods applied by Abdel-Razik et al. (1988a, b), Genin<br />

and Badan-Dangon (1989) and Heneidy (1992, 1996).<br />

The grazing pressure was assessed by determination of<br />

the stocking rate in each site, and according to the status<br />

and abundance of the vegetation (palatable species and<br />

its rate of consumption). These parameter were taken to<br />

estimate the grazing pressure and the pasture condition<br />

(El-Kady, 1983; Heneidy & Bidak, 1998). Floristic identification<br />

are according to Täckholm (1974) and the Latin<br />

names of species were updated following Boulos (1995).<br />

Pasture condition based on the distribution and the<br />

valuable of rangeland species were recorded in each site.<br />

Selection of the indicator (key) species<br />

The key-plant and key-area concepts have proven<br />

highly useful to managers in evaluating grazing effects on<br />

range vegetation (Holechek, 1988). A key species is one<br />

whose use indicates the degree of use of associated species<br />

and is important in the management program for<br />

study. The key species in the study area can withstand<br />

relative higher utilization level during the dry period.They<br />

also provide the bulk of the forage for the grazing animals<br />

within the pasture. They are abundant, productive, and<br />

palatable. In the present study the key species have been<br />

considered palatable for all types of livestock, and are<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


considered as a leading dominant species among the perenials.<br />

However, in each habitat, the key species (indicator<br />

species) are selected according to their abundance<br />

(cover and biomass) and palatability beside the above characters.<br />

In each site from five to seven species are recorded<br />

as key species (ranges from six to twenty species in<br />

the saline depression and in the flat plateau habitats respectively).<br />

Key management species are those on which<br />

the grazing management of a specific range is based.These<br />

species are ligneous plants and are considered as a<br />

constant biological resource, i.e. they are the main feeding<br />

source throughout the year (Holechek, 1988). The<br />

key species are also serve as indicators of management<br />

effectiveness. Generally, when the key species are considered<br />

properly used, the entire pasture is considered correctly<br />

used.<br />

Chemical analysis<br />

BROWSING AND NUTRITIVE VALUE OF THE MOST COMMON RANGE SPECIES IN COASTAL EGYPT ◆<br />

Chemical analysis were carried out on the samples collected<br />

during the wet season (the growing season) at the<br />

time of maximum consumption.These samples are eaten<br />

parts (grazeable parts) of each indicator species in each<br />

site. Before analysis the samples were cleaned and dried<br />

(carefully rinsed several times with tap and distilled water,<br />

then air dried and oven dried at 70 oC to constant weight).<br />

The analyses were to performed determine the following<br />

parameters: (a) total carbohydrates (nitrogen free extract,<br />

(NFE) according to Murata et al. (1968); (b) Crude protein<br />

CP; (c) ether extract (EE) and ash content according<br />

to Allen et al. (1974); (d) Crude fiber (CF) using the formula<br />

[CF = 100 – (CP + EE + NFE + ash)] after Le<br />

Houérou (1980b); (e) Digestible crude protein (DCP)<br />

using Demarquilly’s equations DCP = 0.93 CP – 3.52<br />

after De Ridder et. al. (1982).This equation is only valid<br />

in the case of nitrogen concentration among 3 to 0.61%<br />

(or CP > 3.81 %).TDN was calculated according to Abu-<br />

El-Naga and El Shazly (1971); (f) Energy contained in<br />

food (gross energy, or energy intake, GE) was determined<br />

according to Petrusewicz (1976); (g) The diet gross<br />

energy was calculated from its chemical analysis components<br />

multiplied by their average of heat of combustion<br />

(Lofgreen, 1951); (h) Digestible and metabolizable energy<br />

(DE and ME) were calculated according to Crampton<br />

and Harris (1969).<br />

Direct observation of animal behaviour, such as counting<br />

the number of bites from each plant species is a practical<br />

way of collecting dietary data in open pastures<br />

(Abdel-Razik et al., 1988a; Heneidy, 1992). In the present<br />

study numerous factors were considered when clas-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 39-49<br />

sifying a plant as palatable or not. Phenological stage, morphological<br />

form, odour, taste, chemical composition and<br />

its abundance were the main factors taken as the major<br />

effective parameters on the dietary data. However, through<br />

the long and continuous field observations, the degree of<br />

preference and the consumption rate were assessed. After<br />

getting the results of the above mentioned parameters<br />

palatability can be assessed (Heneidy, 1992, 1996;<br />

Heneidy & Bidak, 1996).<br />

RESULTS<br />

Palatability and pasture condition<br />

Eighty-two species were investigated in the study area<br />

(39 perennial and 43 annual species). Families, life-forms,<br />

palatability rate, grazing value, consumed parts, and types<br />

of livestock are represented in table 1. The most abundant<br />

life-forms of the species under study are chamaephytes.<br />

About 51 % of the perennial species are highly<br />

palatable, 38 % are palatable and about 10 % has shown<br />

low palatable. For annual species about thirty are highly<br />

palatable, 44 % palatable, 16 % are low palatability, and<br />

9% are unpalatable species for livestock.The grazing value<br />

is mostly dependent upon the high percentage of crude<br />

protein, mineral nutrients and low concentration of crude<br />

fibers. About 51 % of the perennial species have very high<br />

grazing value, 31 % attained high grazing value, and 18%<br />

have low grazing value. For annuals about 49 % attained<br />

high grazing value (5 % very highly grazing value), while<br />

the rest 51 % attained low grazing value.<br />

Table 2 shows the perennial indicator species, their<br />

habitats and their frequency in each site of the study area.<br />

The woody species Atriplex halimus is the most frequent<br />

indicator species (found in 59 % of the sites) in the study<br />

area. It is followed by Asphodelus ramosus and Salsola<br />

tetrandra in 56 and 50 % of the total sites respectively.<br />

Table 3 shows pasture condition (good, fair and poor),<br />

grazing pressure and local rainfall in different sites. About<br />

59 % of all sites are generally good pasture, depending<br />

upon the type of vegetation, (vegetation composition), the<br />

dry matter production, and the contribution of the palatable<br />

species. In the mean time twenty-two percent of the<br />

total sites are fair in pasture condition while 19 % are poor<br />

pastures. Generally, on habitat level, about 38 % of all<br />

habitats are suffering from highly grazing pressure while,<br />

34 % and 28 % are under moderate and low grazing pressure<br />

respectively. Fifty-nine percent of habitats are good<br />

41


42<br />

◆ S. Z. HENEIDY<br />

Species Family Life-form Palata-Bility Grazing Stock Consumed<br />

value type part<br />

Perennials<br />

Anabasis articulata (Forssk.) Moq Chenopodiaceae Ch. P VH CSG Yb, F<br />

Anabasis oropediorum Maire Chenopodiaceae Ch. VHP VH CSG Yb, F<br />

Artemisia herba-alba Asso. Compositae Ch. VHP VH SGC Yb, F<br />

Argyrolobium uniflorum (Decne.) Jaub & Spach Leguminosae Ch. VHP VH SGC Yb<br />

Asphodelus ramosus L. Liliaceae G. HP VH GS Yl,Dl,I<br />

Astragalus sieberi DC Leguminosae Ch. HP VH CGS Yb. Fl<br />

Astragalus spinosus (Forssk.) Muschl Leguminosae Ch. VHP VH CGS Yb, Fl<br />

Atriplex halimus L. Chenopodiaceae Ch/Ph HP H CSG Yb. L<br />

Carthamus lanatus L. Compositae Ch. P H CGS I,L<br />

Centaurea alexandrina Descr & Delile Compositae Ch. LP L1 CGS I<br />

Deverra tortuosa (Desf.) DC., Prodr Caryophyllaceae Ch. VHP L1 SGC Yb<br />

Echiochilon fruticosum Desf. Boraginaceae Ch. VHP VH SGC Yb, L<br />

Echinops spinosissimus Turra. Compositae Ch. P H CGS I,Yb<br />

Gymnocarpos decandrum Frossk. Caryophyllaceae Ch. VHP H SGC Yb, L<br />

Haloxylon scoparia Pomel Noum. Chenopodiaceae Ch. P VH CSG Yb, L<br />

Helianthemum kahiricum Del. Cistaceae Ch. HP VH SG Yb. L<br />

Helianthemum lippii (L.) Pres. Cistaceae Ch. VHP VH SG Yb, L<br />

Kickxia aegyptica Dumart. Scrophulariaceae Ch. P H CGS Yb, L<br />

Launaea nudicaulis (L.) Hook. Compositae Hc VHP VH SG Ab.Gr<br />

Lygeum spartum Loefl. Ex. L. Gramineae Hc P H SG Ab.Gr<br />

Lycium shawii Roem & Schult. Solanaceae Ch. HP VH SGC Yb, L<br />

Noaea mucronata (Frossk.) Asch.&Schwei. Chenopodiaceae Ch. HP VH CSG Yb, L<br />

Periploca angustifolia Labill. Asclepiadaceae Ch. VHP VH SGC Yb, L<br />

Plantago albicans L. Plantaginaceae Hc. VHP L1 SG Ab.Gr<br />

Polygonum equisetiforme Sm. Polygonaceae Ch. P H SG Ab. Gr<br />

Salvia aegyptiaca L. Labiatae Ch. P VH SG Yb, L<br />

Salvia lanigera L. Labiatae Ch. P VH SG Yb, L<br />

Salsola tetrandra Forssk. Chenopodiaceae Ch. P VH CGS Yb<br />

Salsola tetragona Del. Chenopodiaceae Ch. VHP VH SGC Br, L<br />

Salsola vermiculata Poir Chenopodiaceae Ch. P H CGS Yb, L<br />

Scorzonera undulata Vahl. Compositae G. VHP VH SG Ab. Gr<br />

Stipa parviflora Desf. Gramineae Hc. VHP VH SG Ab, Gr<br />

Suaeda pruinosa Lange. Chenopodiaceae Ch. P H CGS Yb, L<br />

Suaeda vermiculata Forssk. Ex. J. F. Gmel Chenopodiaceae Ch. P H CGS Yb, L<br />

Suaeda volkensii C.B.Clarke Chenopodiaceae Ch. P H CGS Yb<br />

Teucrium polium L. Labiatae Ch. LP L1 GS Yb, Fl<br />

Thymelaea hirsuta (L.) End. Thymelaeaceae Ch. P L1 GS Yb, Dbr<br />

Verbascum letourneuxii Asch.&Schwei. Scrophulariaceae Ch. LP L1 GSC Yb, L<br />

Zilla spinosa (Turra) Prantl. Cruciferae Ch. LP L1 CG Yb, L<br />

Annuals<br />

Adonis dentata Delile Compositae Th. LP L1 SG Fl.<br />

Aegilops bicornis (Forssk.) Jaub& Spach. Gramineae Th. VHP H SG +<br />

Aizoon canariense L. Aizoaceae Th. NP L1 -- --<br />

Anacyclus alexandrinus Willd. Compositae Th. P H GCS Fl.<br />

Anchusa aegyptiaca (L.) ADC. Boraginaceae Th. P H SG +<br />

Anthemis microsperma Boiss & Kotschy Compositae Th. P H SG Fl.<br />

Beta vulgaris L. Chenopodiaceae Th. P H GS +<br />

Brassica tournefortii Gouan. Cruciferae Th. P L1 CG +<br />

Bromus rubens L. Gramineae Th. HP H SG +<br />

Bupleurum semicompositum L. Umbelliferae Th. P L1 SG +<br />

Calendula arvensis L. Compositae Th. P L1 GS Fl.<br />

Carrichtera annua (L.) DC Cruciferae Th. P L1 SG +<br />

Centaurea glomerata Vahl. Compositae Th. P L1 SG +<br />

Chenopodium murale L. Chenopodiaceae Th. NP L1 -- --<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


BROWSING AND NUTRITIVE VALUE OF THE MOST COMMON RANGE SPECIES IN COASTAL EGYPT ◆<br />

Chrysanthemum coronarium L. Compositae Th. NP L1 -- --<br />

Cutandia dichotoma (Forssk.) Trab. Gramineae Th. VHP H SG +<br />

Didesmus aegyptius (L.) Desv. Cruciferae Th. LP L1 SG +<br />

Filago desertorum Pomel. Compositae Th. P L1 SG +<br />

Herniaria hirsuta L. Caryophyllaceae Th. LP L1 SG +<br />

Hippocrepis areolata Desv. Leguminosae Th. HP H SG +<br />

Hordeum leporinum (Link.) Arcong. Gramineae Th. HP H SGC +<br />

Ifloga spicata (Forssk.) Sch. Compositae Th. P L1 GS +<br />

Lobularia arabica (Boiss) Musch. Cruciferae Th. P H SG +<br />

Malva parviflora L. Malvaceae Th. P L1 SG +<br />

Matthiola livida (Delile) Maire Cruciferae Th. P L1 SG +<br />

Medicago laciniata (L.) Mill. Leguminosae Th. HP H SG +<br />

Medicago truncatula Gaertm. Leguminosae Th. HP H SG +<br />

Onobrychis crysta-galli (L.) Lam. Leguminosae Th. HP H SG +<br />

Ononis serrata Forssk. Leguminosae Th. LP H SG +<br />

Paronychia arabica (L.) DC. Caryophyllaceae Th. LP L1 GS +<br />

Phalaris minor Retz. Gramineae Th. HP H SG +<br />

Picris radicata (Frossk.) Less. Compositae Th. P L1 GS Fl.<br />

Plantago crypsoides Boiss. Plantaginaceae Th. HP H SG +<br />

Plantago ovata Frossk. Plantaginaceae Th. P H SG +<br />

Pseudorlaya pumila (L.) Grande Umbelliferae Th. P L1 SG +<br />

Reichardia tingitana (L.) Roth. Compositae Th. P L1 SG I<br />

Scabiosa rhizantha Viv. Dipsacaceae Th. LP L1 SG +<br />

Schismus barbatus (L.) Thell. Gramineae Th. HP H SG +<br />

Silene villosa Forssk. Caryophyllaceae Th. NP L1 -- --<br />

Spergularia marina (L.) Griseb. Caryophyllaceae Th. LP L1 SGC +<br />

Trifolium resupin atum L. Leguminosae Th. VHP VH SGC +<br />

Trigonella stellata Forssk. Leguminosae Th. VHP VH SGC +<br />

Valantia hispida L. Rubiaceae Th. LP H SG --<br />

Ch. = Chamaephytes, G. = Geophytes, Hc. = Hemicryptophytes, Ph. = Phanerophytes, Th. = Therophytes, VHP. = Very highly palatable, Hp. = Highly palatable,<br />

P. = Palatable, L.P = Low palatability, NP. = Unpalatable, VH.= Very high, H.= High, L1. = Low, S = Sheep, G. = Goats, C.= Camels, and + = Whole<br />

plant (annual species),Yb, = Young branch, Fl. = Flowers (without peduncle), L.= Leaves, Br.= Branch, I. = Inflorescence, Dl = Dead leaves, D br. = Dead<br />

branch, Ab. Gr. = Above-ground. (perennial herb species).<br />

Table 1. List of plant species, families, life-forms, palatability, grazing value, stock type, and consumed part in Matruh area.<br />

pastures while, 22 and 19 % may be considered to have<br />

fair and poor pasture condition respectively. The maximum<br />

rainfall is attained in site number 16 (173.3 mm),<br />

while the minimum rainfall is attained in site number 26<br />

(6.4 mm). On the other hand, the maximum average rainfall<br />

is observed in saline depression 107.3 ± 9.2 mm, followed<br />

by 77.8 ± 13.7 mm in the rocky plateau habitat<br />

(local meterological stations). Amount of rainfall is the<br />

most effective factors on the plant composition and the<br />

pasture condition, especially in the desert ecosystem where<br />

the growth of plant species is completely dependant of<br />

the amount and distribution of rainfall.<br />

Chemical analysis and nutritive value<br />

The individual indicator species in all sites were difficult<br />

to organize for chemical analysis so that all individuals<br />

at each habitat were pooled together in a compo-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 39-49<br />

site sample (different individuals of the same species) for<br />

chemical analysis as a habitat indicator species. Results of<br />

the chemical composition of the indicator species (grazeable<br />

parts) of each habitats (means and standard errors)<br />

during the season of maximum consumption are shown<br />

in table 4. Obviously the numbers of indicators species<br />

differs from one habitat to another.The table shows that<br />

the values of NFE ranges from a maximum of 34.1 ±<br />

3.9 % in the flat plateau to a minimum of 30.9 ± 4.7% in<br />

the non-saline depression. Also, the indicator species of<br />

the flat plateau attained the maximum ash content (14.2<br />

± 8.2 %). In addition the species of the non-saline depression<br />

attained the maximum of CP, CF, DCP, and GE (10.7<br />

± 4.3, 36.9 ± 9.4, 6.4 ± 4.0 % and 4.1 ± 0.23 Mcal kg -1<br />

respectively). Moreover, TDN values are the highest in<br />

the species of saline depression (67.2 ± 2.9 %). However,<br />

there was little variability in the percentages of chemical<br />

43


44<br />

◆ S. Z. HENEIDY<br />

Species Site number which species present Habitat* Total number % (from<br />

of sites all sites)<br />

Anabasis articulata 1,5,19,25 R.Pl.,F.Pl. 4 13<br />

Anabasis oropediorum 2,8,9,14,19,24,30 R.Pl.,F.Pl.,R.R. 7 22<br />

Artemisia herba-alba 3,25,27 R.Pl.,F.Pl. 3 9<br />

Argyrolobium uniflorum 22,31 N.Sal.Dep. 2 6<br />

Asphodelus ramosus 1,2,3,4,5,7,9,13,14,15,16,18,21,22,23,26,29,31 R.Pl.,F.Pl.,R.R.,N.Sal.Dep.,Sal.Dep. 18 56<br />

Atriplex halimus 6,7,8,10,11,12,13,15,16,17,19,20,21,23,26,28,29,30,32 R.Pl.,F.Pl.,R.R.,Sal.Dep. 19 59<br />

Carthamus lanatus 2,3,6,11,13,16,17,20 R.Pl.,F.Pl.,R.R.,N.Sal.Dep.,Sal.Dep. 8 25<br />

Deverra tortuosa 2,3,4,8,10,12,13,14,18,26,30,31,32 R.Pl.,F.Pl.,N.Sal.Dep. 13 41<br />

Echinops spinosissimus 1,2,5,6,9,14,21,23,31 R.Pl.,F.Pl.,R.R.,N.Sal.Dep. 9 28<br />

Gymnocarpos decandrum 2,4,6,7,8,9,14,23,24,25,29,32 R.Pl.,F.Pl.,R.R.,N.Sal.Dep. 12 38<br />

Haloxylon scoparia 1,3,5,6,7,13 R.Pl.,F.Pl.,R.R.,N.Sal.Dep. 6 19<br />

Helianthemum kahiricum 27 F.Pl. 1 3<br />

Helianthemum lippii 7 F.Pl. 1 3<br />

Lycium shawii 4,22,23,28 R.Pl.,F.Pl.N.Sal.Dep. 4 13<br />

Noaea mucronata 5,9,10,12,16,18,26 R.Pl.,F.Pl.,R.R. 7 22<br />

Periploca angustifolia 20,25,32 R.Pl.,R.R. 3 9<br />

Salvia aegyptiaca 6,11,12,22 F.Pl.,N.Sal.Dep. 4 13<br />

Salsola tetrandra 8,9,10,11,12,15,17,18,19,20, 21,24,25,26,27,30 R.Pl.,F.Pl.,R.R.,N.Sal.Dep.Sal.Dep. 16 50<br />

Salsola tetragona 2,10,11,15,17,21,24,25,26,27,29,30,32 R.Pl.,F.Pl.,Sal.Dep. 13 41<br />

Salsola vermiculata 10,28 F.Pl. 2 6<br />

Stipa parviflora 29 F.Pl. 1 3<br />

Suaeda pruinosa 8,19,20,28 R.Pl.,F.Pl.,R.R. 4 13<br />

Suaeda vermiculata 8 Sal.Dep. 1 3<br />

Thymelaea hirsuta 1,3,4,15,16,18,31 R.Pl.,N.Sal.Dep.,Sal.Dep 7 22<br />

Zilla spinosa 24 R.Pl. 1 3<br />

* R. Pl. = Rocky plateau, F.Pl. = Flat plateau, R. R. = Rocky ridge, N. Sal. Dep. = Non-saline depression, and Sal. Dep. = Saline depression.<br />

composition within different habitats. On the other hand,<br />

chemical analysis may differ from one habitat to another<br />

for the same species. For example A. ramosus (perennial<br />

herb) is common in all habitats, where the NFE content<br />

was high (31.3 %) in the rocky- plateau species, and low<br />

(29.4 %) in the flat plateau.<br />

Averages of digestible (DE) and metabolizable energy<br />

(ME) contents (Mcal kg -1 ) of the indicator species of different<br />

habitats of the study area are shown in table 5.The<br />

highest DE and ME values was attained by the grazeable<br />

parts of Periploca angustifolia (3.84 and 3.32 Mcal kg -1<br />

respectively) in the rocky ridge habitat while, the lowest<br />

contents were attained by Salsola tetragona from the saline<br />

depression and Salvia aegyptiaca in the non- saline depression<br />

habitats.<br />

DISCUSSION<br />

Table 2. Indicator species (perennial), the frequency, habitats and percentage of each species at the sampling sites of study area.<br />

Livestock of the study area is constituted by sheep,<br />

goats and camels, which have different behaviour and pre-<br />

ferences for plant species.Young branches and leaves are<br />

prefered but sometimes dead parts and inflorescences are<br />

also consumed by such livestock (Heneidy, 1992).<br />

Palatability of range plant species is a very complex notion.<br />

Moreover, it is difficult to generalize as it is linked to many<br />

factors that vary in time and space and is relative to what<br />

other alternative are available. Some of these variables are<br />

linked to the plant, others to the animals while a third<br />

category depends on various environmental factors (Le<br />

Houérou, 1980a; Heneidy, 1996; Heneidy & Bidak, 1996).<br />

As a rule, the palatability of a given taxon would<br />

increase with environmental aridity (Le Houérou, 1992).<br />

However, the same species may be palatable in some<br />

places and unpalatable in the others (Le Houérou, 1992;<br />

Heneidy, 1996). Palatable species typically put all their<br />

energy from photosynthesis into growth and reproduction,<br />

while unpalatable species put part of their energy<br />

into compounds that discourage defoliations. So the more<br />

productive and desirable vegetation decreased with an<br />

increase in the less palatable and poisonous plant species<br />

(Taylor & Ralphs, 1992). Most of the species in the study<br />

area are highly palatable (about 87% of perennials and<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


BROWSING AND NUTRITIVE VALUE OF THE MOST COMMON RANGE SPECIES IN COASTAL EGYPT ◆<br />

Site number Pasture condition Grazing pressure Habitat Mean of rainfall (mm)<br />

4 Low Non-saline depression 110<br />

5 Moderate Flat plateau 10<br />

6 Low Flat plateau 20<br />

8 Low Rocky ridge 74<br />

10 Moderate Flat plateau 87<br />

12 High Flat plateau 87<br />

14 Moderate Rocky plateau 102<br />

18 High Non-saline depression 97<br />

21 Low Rocky plateau 66<br />

22 Good Low Non-saline depression 84<br />

23 Moderate Rocky plateau 30<br />

24 Moderate Rocky plateau 55<br />

25 Low Rocky plateau 116<br />

26 Low Flat plateau 6<br />

27 Moderate Flat plateau 51<br />

29 Moderate Flat plateau 28<br />

30 Low Rocky plateau 30<br />

31 Moderate Non-saline depression 17<br />

32 Low Rocky ridge 42<br />

2 Moderate Rocky plateau 31<br />

7 High Flat plateau 62<br />

11 Moderate Flat plateau 78<br />

13 Fair High Rocky ridge 105<br />

15 High Saline depression 120<br />

19 High Flat plateau 162<br />

28 Moderate Flat plateau 28<br />

1 High Rocky plateau 90<br />

3 High Non-saline depression 67<br />

9 Poor High Rocky ridge 80<br />

16 High Rocky plateau 173<br />

17 High Saline depression 84<br />

20 High Rocky plateau 85<br />

Table 3.The pasture condition, [grazing pressure, habitats and average of rainfall (mm)] at the different sites of Matruh area.<br />

74% of annuals), which is consistent with the results recorded<br />

by Heneidy (1992) and Heneidy and Bidak (1998)<br />

in other grazing areas.<br />

The present study area is considered as one of the<br />

richest area for its forage resources and for the livestock<br />

on the coastal region. The most common life-forms are<br />

shrubby species (browse species) which are the skeleton<br />

part of this system.The contribution of the shrubby species<br />

is about 30% of perennial species some of them are<br />

woody, their heights reach to 3m, while about 41% of them<br />

are sub-shrubs (dwarf shrubs). However, the propagation<br />

of woody species in the study area is very important for<br />

improvement the present pasture. Where the fodder<br />

shrubs are essentially a supplementary or emergency feed<br />

(85% of shrubby species are highly palatable) for periods<br />

of grazing shortage that may occur in the dry season and/<br />

or prolonged droughts (Le Houérou, 1989).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 39-49<br />

The contribution of shrubby species is about 80% of<br />

the total rangeland production in the study area Heneidy<br />

(under publication). Shrubby species also play an important<br />

role in the protection of catchement basins, erosion<br />

control, fuel wood, and the strategy against desertification.<br />

Some of the shrubby species are able to improve the<br />

soil fertility and to increase the crop yield owing to the<br />

organic matter and nutrients, windbreak, to bring to the<br />

topsoil from deep layers of the subsoil and to smooth off<br />

the microclimatic condition (Le Houérou, 1989).<br />

The pasture condition is based on many factors, some<br />

of them are biomass, and accessible parts (actual production)<br />

for the livestock (Heneidy, 2000). On the other<br />

hand, there are factors like palatability, life-forms, performance,<br />

of plant species, nutritive value, topographic<br />

factor and soil type. However, the field observation is one<br />

of the main criteria to justify the status of the pasture.<br />

45


46<br />

◆ S. Z. HENEIDY<br />

According to Le Houérou (1993) the pasture conditions<br />

can be distinguished into three categories (good, fair, and<br />

poor). However, the pasture condition in the study area<br />

is generally good, as 59% of the sites is good pasture and<br />

about 22% is fair.<br />

The grazing value of the plant species is influenced by<br />

many factors. The main factors are chemical composition,<br />

morphology and floristic composition. However, in<br />

the present study, grazing value is mainly based on chemical<br />

composition especially, i.e. where the pasture has<br />

Percentage Mcal kg- 1<br />

Habitat NFE CP CF EE Ash DCP TDN GE<br />

Rocky plateau (18 species) 33.98 ± 4.32 9.6 ± 2.9 32.4± 10.2 8.0 ± 2.7 16.8 ± 8.1 5.4 ± 2.5 67.2 ± 1.9 4.0 ± 0.37<br />

Flat plateau (20 species) 34.1 ± 3.9 10.1± 2.5 30.2± 9.2 7.1 ± 1.9 19.2 ± 8.2 5.9 ± 2.3 66.6 ± 1.5 3.9 ± 0.39<br />

Rocky ridge (12 species) 33.7 ± 5.0 10.4± 2.7 32.2± 4.7 7.1 ± 1.5 16.9 ± 6.2 6.1 ± 2.5 66.7 ± 1.4 3.9 ± 0.32<br />

Non-saline depression (13 species) 30.9 ± 4.7 10.7 ± 4.3 36.9 ± 9.4 6.9 ± 2.5 13.7 ± 5.4 6.4 ± 4.0 66.4 ± 1.9 4.1 ± 0.23<br />

Saline depression (6 species) 31.6 ± 5.8 9.2±1.9 33.2 ± 13.4 7.6 ± 2.4 18.6 ± 6.9 4.9 ± 1.1 67.2 ± 1.9 3.9 ± 0.26<br />

Table 4. Means and standard error of chemical composition {percentage of nitrogen free extract (NFE), crude protein (CP), ether extract (EE), crude fibre (CF), ash, digistable<br />

crude protein (DCP), total digistable nutrients (TDN), and gross energy (GE Mcal kg-1)} of the grazeable parts of the indicator species at the different habitats of<br />

Matruh area.<br />

Habitats<br />

Species Non-saline depression Flat plateau Rocky plateau Rocky ridge Saline depression<br />

DE ME DE ME DE ME DE ME DE ME<br />

Anabasis articulta -- -- 3.1 2.7 3.3 2.8 -- -- -- --<br />

Anabasis oropediorum -- -- 2.7 2.3 2.9 2.5 2.9 2.5 -- --<br />

Artemisia herba-alba 3.5 3.0 3.5 3.0 3.7 3.2 -- -- -- --<br />

Asphodelus ramosus 3.2 2.8 3.4 2.9 3.5 3.0 3.3 2.8 3.4 2.9<br />

Atriplex halimus -- -- 3.1 2.7 3.1 2.7 3.1 2.7 3.1 2.6<br />

Carthamus lanatus 3.3 2.8 3.5 3.0 3.4 2.9 3.3 2.8 3.4 3.0<br />

Deverra tortuosa 3.5 3.0 3.4 2.9 3.6 3.1 3.4 2.9 -- --<br />

Echinops spinosissimus 3.5 3.0 3.6 3.1 3.1 2.7 3.0 2.6 -- --<br />

Gymnocarpos decandrum 3.2 2.8 3.2 2.8 3.5 3.1 3.5 3.0 -- --<br />

Haloxylon scoparia 3.5 3.0 3.4 2.9 3.4 2.9 3.0 2.6 -- --<br />

Helianthemum kahiricum -- -- 3.6 3.1 -- -- -- -- -- --<br />

Helianthemum lippii -- -- 3.0 2.6 -- -- -- -- -- --<br />

Lycium shawii 3.4 2.9 3.3 2.8 3.7 3.2 -- -- -- --<br />

Noaea mucronata 3.4 2.9 3.5 3.0 3.2 2.8 3.4 2.9 -- --<br />

Periploca angustifolia -- -- -- -- 3.7 3.2 3.8 3.3 -- --<br />

Salvia aegyptiaca 3.5 3.0 3.5 3.0 -- -- -- -- -- --<br />

Salsola tetrandra 2.9 2.5 2.9 2.5 2.8 2.4 3.4 2.9 3.0 2.6<br />

Salsola tetragona -- -- 2.7 2.3 2.9 2.5 -- -- 2.9 2.5<br />

Stipa parviflora -- -- 3.4 2.9 -- -- -- -- -- --<br />

Suaeda pruinosa -- -- 2.7 2.3 2.7 2.3 2.9 2.5 -- --<br />

Suaeda vermiculata -- -- 2.8 2.4 -- -- -- -- -- --<br />

Thymelaea hirsuta 3.4 2.9 -- -- 3.5 3.0 -- -- 3.5 3.0<br />

Zilla spinosa -- -- -- -- 3.3 2.8 -- -- -- --<br />

Table 5. Average of digestible energy (DE), and metabolizable energy (ME) (Mcal kg-1) in the indicator species in each habitat of Matruh area.<br />

higher CP content and minerals, and lower CF content.<br />

Eighty-two percent of the perennial species and 44% of<br />

the annual species in the present study have high grazing<br />

value.<br />

Overgrazing is one of the most important issue of arid<br />

land especially in rangeland areas (Ayyad, 1978; Abdel-<br />

Razik et al. 1988a,b; Le Houérou, 1989; Heneidy, 1992).<br />

Based on the long field observation of different flocks in<br />

each site, the degree of consumption of plant species,<br />

topography of the site, the ratio of unpalatable species<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


BROWSING AND NUTRITIVE VALUE OF THE MOST COMMON RANGE SPECIES IN COASTAL EGYPT ◆<br />

and the position of the settlements (i.e. whether they inhabit<br />

near by or far from the pasture), it is possible to estimate<br />

the grazing pressure in the different sites (El-Kady,<br />

1983; Heneidy, 1992; Heneidy & Bidak, 1998). In the present<br />

study 38 % of the sites have high grazing pressure,<br />

34 % have moderate and 28 % have low grazing pressure.<br />

Rangeland under high grazing pressure needs rotational<br />

grazing management to reduce the grazing intensity<br />

(depend on the status of the pasture). Actually, destocking<br />

deferred grazing and reduction of stocking rates are<br />

also effective means to regenerate these rangelands.<br />

The present study indicates that the indicator species<br />

have relatively high content of CP and low content of CF<br />

compared with many desert plants (Le Houérou, 1980b).<br />

CP and CF are viewed classically as an indicator of the<br />

nutritional value of plant as feed for ruminants (Bryant<br />

& Kuropat, 1983; Heneidy, 1992). In all habitats of the<br />

study area about 75% of the indicator species contained<br />

more than 8% CP and 50% of them contained less than<br />

30% CF. According to the percentage of CP the sites could<br />

be ranked into three groups.The first group includes 47%<br />

(15 sites) of the sites where CP > 10%, the second group<br />

includes 31% (10 sites) of the sites where CP > 8%, while<br />

the third group includes 22% (7 sites) of the sites where<br />

CP < 8%.The maximum average of CP is attained by the<br />

indicator plant species in the non-saline depression habitat<br />

(10.7 ± 4.3 %), while the minimum is attained by the<br />

indicator plant species in the saline depression (9.2 ±<br />

1.9 %).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 39-49<br />

% Kcal kg -1<br />

Fodder References DCP TDN GE DE ME<br />

Supplementary feed Soliman & El-Shazly, 1978<br />

berseem -- 56 -- -- -barley<br />

-- 64 -- -- -corn<br />

-- 68 -- -- --<br />

Western <strong>Mediterranea</strong>n Abdel-Salam, 1985 -- 66 -- -- --<br />

Desert Pasture (Omayed) Abdel-Razik et al., 1988 5.4 75 -- -- --<br />

Heneidy, 1992 4.9 72 3,606 2,610 2,278<br />

Aqaba Gulf Area Pasture (Sinai) Heneidy, 1996 4.6 66.5 4,100 2,650 2,200<br />

Bisha area (Saudi Arabia) Heneidy, 2000 8.8 74.8 3,974 3,292 2,874<br />

Tunisia (as arid zone) Sarson & El-Hamrouni, 1974 4-20 -- -- -- --<br />

East of Matruh Present study 5.7 67.0 3,993 3,274 2.824<br />

Table 6. A comparison between the present annual average percentage of total digestible nutrients (TDN), digestible crude protein (DCP), gross energy (GE), digestible<br />

energy (DE) and metabolizable energy (ME) in Kcal kg -1 with those supplementary feed and other pastures.<br />

The DCP percentage in the present study varied from<br />

a minimum 4.9 (as an average) in the saline depression<br />

habitat to a maximum 6.4 in the non-saline depression.<br />

This however is comparable with that recorded in the saltmarshes<br />

area of the <strong>Mediterranea</strong>n region where values<br />

ranged from 5.8 to 7.5 (Heneidy & Bidak, 1996), and<br />

with the values recorded by Heneidy (1992) in Omayed<br />

area 4.9 (table 6). This Table also contained the annual<br />

average DCP% in supplementary feed and other pastures.<br />

The mean DCP% in the study area (5.7) is slightly higher<br />

than that recorded in the plants of Aqaba Gulf in Sinai<br />

area by Heneidy (1996). But it is within the range 4 -20%<br />

of the dry matter of DCP% for woody steppe in arid zone<br />

(Sarson & El-Hamrouni, 1974).<br />

The term TDN (total digestible nutrients) is only an<br />

approximate measure of the food energy available to animal<br />

after the digestion losses have been deducted<br />

(Lofgreen, 1951).Therefore,TDN is a measure of energy<br />

requirement of animals and energy value of feeds. TDN<br />

values in the study area ranges from 66.4 ± 0.6% in the<br />

non-saline depression to 67.6 ± 1.1% in the flat plateau.<br />

The annual average percentage of TDN found in the present<br />

study is compared with that of the supplementary<br />

feed and other pastures as also shown in table 6.The TDN<br />

value in the study area is 67.0% comparing with that recorded<br />

in the natural forage from Aqaba Gulf of Sinai region<br />

(e.g. 66.5%) (Heneidy, 1996) and 66.2% in the halophytes<br />

of the <strong>Mediterranea</strong>n region (Heneidy & Bidak, 1996).<br />

The obtained results of TDN in the present study is also<br />

47


48<br />

◆ S. Z. HENEIDY<br />

higher than the average TDN (62.7%) in the supplementary<br />

feed (berseem, barley and corn) reported by<br />

Soliman & El-Shazly (1978). However, Abdel-Salam<br />

(1985) and Abdel-Razik et al. (1988a) reported that the<br />

average of TDN ranges from 66 to 75% in the coastal<br />

region.<br />

The average GE in the study area ranges from 3.9 ±<br />

0.1 Mcal kg -1 in the habitat of the rocky ridge, to 4.1 ±<br />

0.1 Mcal kg -1 in the non-saline depression. Lower GE<br />

values (3.6 and 3.5 Mcal kg -1 ) were recorded by Heneidy<br />

(1992) and Heneidy and Bidak (1996) in the western<br />

coastal region and in the halophytic species of the<br />

<strong>Mediterranea</strong>n region respectively.The average of DE in<br />

the study area was 3.3 ± 0.02 Mcal kg -1 which is less than<br />

2.6 Mcal kg -1 calculated by Heneidy (1992) in Omayed<br />

area of the coastal region.<br />

Generally the average of ME was 2.88 ± 0.04 in the<br />

non-saline depression, 2.8 ± 0.07 in the flat plateau, 2.84<br />

± 0.06 in rocky plateau, 2.79 ± 0.06 in the rocky ridge,<br />

and 2.81 ± 0.08 Mcal kg -1 in the saline depression habitat.<br />

Consequently, the average of ME in the whole study<br />

area was 2.82 ± 0.01 Mcal kg -1 which is compareable with<br />

2.2 Mcal kg -1 recorded by Heneidy (1996) in the natural<br />

forage in Aqaba Gulf.<br />

Le Houérou and Hoste (1977) reported that 1 SFU<br />

(Scandinavian Feed Unit) = 1,650 Kcal. Accordingly, the<br />

energy content in the study aera is equivalent to 0.84 SFU.<br />

Comparing with the energy content of forage in Aqaba<br />

Gulf pasture was 0.66 SFU (Heneidy, 1996), and 0.76<br />

SFU Heneidy and Bidak (1996) in halophytic pasture.<br />

Le Houérou et al. (1982) reported the energy content of<br />

one kg dry matter of the best fodder roughage and good<br />

fodder crops (alfalfa) was 0.8 and 0.6 SFU, respectively.<br />

It seems that the energy content of the pasture in the study<br />

area is higher than the good fodder crops and close to the<br />

best fodder roughage. However, this area is considered<br />

good pasture if compared with other pastures.<br />

CONCLUSION<br />

The problem which faces the pastoralists in range lands<br />

areas is low rainfall and how to prevent or reduced the<br />

land degradation.This is a matter of adjustment of stock<br />

numbers to the natural feed resources. Increasing the grazing<br />

pressure beyond the limits set by dry seasons reduces<br />

fodder resources and leads to depletion in forage<br />

resources. It is therefore appropriate to manage vegetation<br />

properly and to think in terms of encouraging or pro-<br />

pagation of the native species by using techniques which<br />

still have to be established. In particular the following<br />

points should be considered: 1- propagation of the palatable<br />

and high grazing value species (specially woody species<br />

) which produce quantity of seeds (e.g. Artemisia<br />

herba-alba, Anabasis oropediorum, Anabasis articulata,<br />

Atriplex halimus, Argyrolobium uniflorum, Gymnocarpos<br />

decandrum, Helianthemum lippii, Deverra tortuosa, Lycium<br />

shawii and Periploca angustifolia), 2- establishment of the<br />

shrub-land species (especially legume species), 3constructing<br />

a good policy to manage the use of grazing<br />

lands as a purpose to reduce the supplementary feed cost.<br />

AKNOWLEDGEMENTS<br />

The author wish to thank to Prof. F. Abdel-Khader,<br />

(the principal investigator of the Project of Water/Land<br />

Management Northwest Coast of Egypt), for facilitating<br />

during the field work.<br />

Sincere thanks are due to Professor K. H. Shaltout,<br />

and Professor El-Rayis, O. A. for their careful revision the<br />

manuscript although I should emphasize that outstanding<br />

errors are mine.<br />

References<br />

ABDEL-RAZIK M., AYYAD M. A. & HENEIDY S. Z., 1988a. Preference<br />

of grazing mammals for forage species and their nutritive<br />

value in a <strong>Mediterranea</strong>n desert ecosystem (Egypt). J.<br />

Arid Environ. 15: 297-300.<br />

ABDEL-RAZIK M.,AYYAD M.A. & HENEIDY S. Z., 1988b. Phytomass<br />

and mineral composition in range biomass of a <strong>Mediterranea</strong>n<br />

arid ecosystem (Egypt). Œcol. Plant. 9: 359-370.<br />

ABDEL-SALAM H., 1985. Grazing Capacity per Feddan at<br />

Omayed Grazing area “The Northern Coastal Zone”Western<br />

to Alexandria Ph. D.Thesis University of Alexandria. 166 p.<br />

ABU-EL-NAGA M. A. & EL-SHAZLY K., 1971.The prediction of<br />

the nutritive value of animal feeds from chemical analysis. J.<br />

Agric. Sci. 77: 25-37.<br />

ALLEN S. E., GRIMSHAY H. M., PARKINSON T. A. & QUARMBY,<br />

C., 1974. Analysis of Ecology Materials. Oxford, London:<br />

Blackwell Scientific Publications, 565 p.<br />

AYYAD M. A., 1978. A preliminary assessment of the effect of<br />

protection on the vegetation of the <strong>Mediterranea</strong>n desert ecosystems.<br />

Tackholmia 9:85-101.<br />

BOULOS L., 1995. Flora of Egypt:Checklist. Al-Hadra Publishing,<br />

Cairo, 286 p.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


BROWSING AND NUTRITIVE VALUE OF THE MOST COMMON RANGE SPECIES IN COASTAL EGYPT ◆<br />

BRYANT J. P. & KUROPAT P. J., 1983. Selection of winter forage<br />

by sub-arctic browsing vertebrates: the role of plant chemistry.<br />

Ann. Rev. Ecol. Syst. 11, 261-285.<br />

CRAMPTON E.W. & HARRIS L. E., 1969. Applied Animal Nutrition<br />

(2 nd ed.). San Francisco: W. H. Freeman, 753 p.<br />

DE RIDDER N., SROOSIJDER L., CISSE A. M. & VAN KEULEN H.,<br />

1982. A Study of the Soils, the Vegetation and the Exploitation<br />

of that Natural Resource. pps course book vol. 1.Wageningen:<br />

Agriculturral University, 231 p.<br />

DUIVENBOODEN N. VEN, 1985. Animal Husbandry in the Northwestern<br />

Coastal Zone of Egypt. Department of Theoretical<br />

Production Ecology, Agricultural University Wageningen.<br />

Report, 53 p.<br />

EL-KADY H.F., 1983. Animal Resources. In: AYYAD M. A. & LE<br />

FLOC’H E. (eds.). An Ecological Assessment of Renewable<br />

Resources For Rural Agricultural Development in the Western<br />

<strong>Mediterranea</strong>n Coastal Region of Egypt. Mapping Workshop<br />

of CNRS/CEPE: CNRS, p. 77-79.<br />

GENIN D. & BADAN-DANGON A., 1989. Rapport d’activites agropastorales<br />

en République du Niger. IEMVT Paris, 140 p.<br />

HENEIDY S. Z., 1992. An Ecological Study of the Grazing Systems<br />

of Mariut, Egypt. Submitted to UNESCO, 51 p.<br />

HENEIDY S. Z., 1996. Palatability and nutritive value of some<br />

common plant species from the Aqaba Gulf area of Sinai,<br />

Egypt. J. Arid Environ. 34: 115-123.<br />

HENEIDY S. Z., 2000. Palatability, chemical composition and<br />

nutritive value of some common range plants from Bisha,<br />

assir region, southwestern of Saudi Arabia. Desert Inst. Bull.,<br />

Egypt.<br />

HENEIDY S. Z. & BIDAK, L.M., 1996. Halophytes as a forage<br />

source in the western <strong>Mediterranea</strong>n coastal region of Egypt.<br />

Desert Inst. Bull., Egypt. (49), 2: 283-304.<br />

HENEIDY S. Z. & BIDAK L. M., 1998. Diversity of wadi vegetation<br />

in Matrouh region, Egypt. J. Union Arab.Biol.Cairo 6(B).<br />

Botany, 13-28.<br />

HOLECHEK J. L., 1988. An approach for setting the stocking rate.<br />

Rangelands 10(1): 10-14.<br />

KAMAL, S.A., 1988. A study of Vegetation and Land Use in the<br />

Western <strong>Mediterranea</strong>n Desert of Egypt. Ph. D. Thesis<br />

Alexandria University, 193 p.<br />

LE HOUÉROU H. N., 1980a. Browse in Northern Africa. In: Le<br />

Houérou H. N. (Ed.), Browse in Africa. Addis-Abiba: ILCA,<br />

p. 55-82.<br />

LE HOUÉROU H. N., 1980b. Chemical composition and nutritive<br />

value of browse in tropical West Africa. In: Le<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 39-49<br />

Houérou (Ed.), Browse in Africa. Addis-Abiba: ILCA, p. 261-<br />

289.<br />

LE HOUÉROU H. N., 1989. An assessment of the economic feasibility<br />

of fodder shrub plantation in Cm Mckell (ed.). In:<br />

The biology and utilization of shrubs, Ch. 30. Academic Press,<br />

New York, p. 603-630.<br />

LE HOUÉROU H. N., 1992.The role of salt bushes (Atriplex spp.)<br />

In: Arid Land Rehabilitation in the <strong>Mediterranea</strong>n Basin, a<br />

Review. Agroforestry Systems, 18: 107-148.<br />

LE HOUÉROU H. N., 1993. Grazing lands of the <strong>Mediterranea</strong>n<br />

Basin. In: Corpland. R. T. (ed.). Natural grassland, Eastern<br />

Hemisphere and Resume. Ecosystem of the World, vol. 8b.<br />

Amsterdam: Elsevier Science Publisher, p. 171-196.<br />

LE HOUÉROU H. N. & HOSTE C. H., 1977. Rangeland production<br />

and annual rainfall relations in the <strong>Mediterranea</strong>n basin<br />

and in the African Sahelo-Sudanian Zone. J. Range Manage.<br />

30: 181-189.<br />

LE HOUÉROU H. N., GINTZBURGER G. & AL-KHOJA M. M., 1982.<br />

Chemical composition and nutritive value of some range<br />

plants and fodder shrubs in Libya. Tripoli: FAO/Lib/018,<br />

Agricultural Research Centre, 14 p.<br />

LOFGREEN G. P., 1951. The use of digestible energy in the evaluation<br />

of feeds. J. Anim. Sci. 10: 344-351.<br />

MURATA T., AKOZAWA T. & SHIKIKO F., 1968. Enzymic mechanism<br />

of starch break down in germinating rice seeds. Plant<br />

Physiol. 43: 1-189.<br />

PETRUSEWICZ K., 1976. Suggested list of more important<br />

concepts in productivity studies (definition and symbols). In:<br />

Petrusewicz K. (ed.) Secondary Productivity of Terrestrial<br />

Ecosystems. Krakow: Warszawa, p. 51-58.<br />

SARSON M. & EL-HAMROUNI A., 1974.Valeur alimentaire de certaines<br />

plantes spontanées ou introduites en Tunisie. Note de<br />

recherche n° 2, Institut National Recherche Forestière,Tunis,<br />

78 p.<br />

SOLIMAN S. M. & EL-SHAZLY K., 1978. Increasing the productivity<br />

per feddan from total digestible nutrients. Alexandria<br />

J. Agric. Res. 26: 551-556.<br />

TÄCKHOLM V., 1974. Student’s Flora of Egypt, Cairo: Cairo<br />

University Press, 888 p.<br />

TAYLOR C. A. & RALPHS M. H., 1992. Reducing livestock losses<br />

from poisonous plants through grazing management. J. Range<br />

Manage. 45(1): 9-12.<br />

UNESCO, 1977. Map of the world distribution of arid regions.<br />

MAB Technical Notes 7. Paris.<br />

49


Epiphytic lichen vegetation<br />

of the Cedrus atlantica forests (Morocco)<br />

La végétation des lichens épiphytes<br />

des cédraies à Cedrus atlantica du Maroc<br />

Giuseppe Massari, Sonia Ravera<br />

Dipartimento di Biologia Vegetale. Università di Roma “La Sapienza”<br />

Piazzale Aldo Moro 5, 00185 Roma, Italia – email: giuseppe.massari@uniroma1.it<br />

Summary<br />

In Morocco, human influence induces actually transformation and<br />

destruction of forests. We conducted a three-year study (1998-2000)<br />

on the epiphytic lichen vegetation of the Cedrus atlantica forests<br />

of the mountain ranges of Rif and Central Atlas, comparing our<br />

findings with the syntaxonomical position of the syntaxa described<br />

in Europe and using, as reference, the epiphytic acidophilous communities<br />

of Hypogymnietea physodis Follm. 1974.<br />

Although most of the lichen species observed are common in the<br />

mountains of Europe, the Moroccan communities show substantial<br />

differences. These differences depend on climatic, ecological,<br />

and phytogeographic factors, which are strictly related to the<br />

<strong>Mediterranea</strong>n character of the Cedrus atlantica forests.<br />

The specificity of the epiphytic lichen vegetation of Moroccan Cedrus<br />

atlantica forests is represented by Pseudevernio-Hypogymnietum<br />

laminisorediatae ass. nov. which replaces the Pseudevernietum<br />

furfuraceae occurring in European coniferous woods.<br />

Key-words<br />

Phytosociology, epiphytic lichen vegetation, Morocco, Cedrus atlantica<br />

forests, syntaxonomy.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 51-57<br />

Résumé<br />

L’activité humaine et la dégradation sont en train de transformer<br />

la végétation forestière du Maroc. Pour étudier les communautés<br />

de lichens épiphytes qui présentent encore leur structure originale,<br />

nous avons entrepris une étude sur trois ans (1998-2000) de la<br />

végétation lichénique des forêts à Cedrus atlantica du Rif et de<br />

l’Atlas central. Les données marocaines ont été comparées aux unités<br />

syntaxonomiques des forêts à Larix decidua et Pinus cembra<br />

des montagnes européennes et, en particulier, aux communautés<br />

acidophyles épiphytiques de l’ Hypogymnietea physodis Follm.<br />

1974.<br />

Malgré une certaine similitude floristique, les communautés des<br />

lichens des cédraies marocaines diffèrent de celles des montagnes<br />

européennes : les différences dépendent surtout des facteurs climatiques,<br />

écologiques et phytogéographiques qui marquent les limites<br />

de la végétation épiphytique oroboréale avec la végétation oroméditerranéenne.<br />

La spécificité des communautés épiphytiques des cédraies du Maroc<br />

est représentée par le Pseudevernio-Hypogymnietum laminisorediatae<br />

ass. nov., vicariante du Pseudevernietum furfuraceae des<br />

montagnes européennes.<br />

Mots-clés<br />

Phytosociologie, végétation lichénique épiphytique, Maroc, forêts à<br />

Cedrus atlantica, syntaxonomie.<br />

51


52<br />

◆ GIUSEPPE MASSARI, SONIA RAVERA<br />

INTRODUCTION<br />

The lichen flora of the Cedrus atlantica forests of<br />

Morocco is apparently similar to that of the coniferous<br />

woods of the mountains of Europe. However, the epiphytic<br />

lichen vegetation of the Moroccan forests has never<br />

been thoroughly studied, a fact which is becoming of<br />

increasing importance in that these old forms are rapidly<br />

being eliminated or transformed by human influence and<br />

deforestation.<br />

To acquire knowledge on the lichen communities of<br />

the Moroccan cedar forests that still conserve traces of<br />

the original elements, we conducted a three-year study<br />

(1998-2000) on the epiphytic lichen vegetation of the<br />

Cedrus atlantica forests of the mountain ranges of Rif and<br />

Central Atlas, comparing our findings with the syntaxonomical<br />

position of the syntaxa described in Europe, using<br />

as reference the epiphytic acidophilous communities of<br />

Hypogymnietea physodis Follm. 1974 that have a subartic,<br />

subalpine and euatlantic distribution.<br />

We also aimed to expand the knowledge obtained from<br />

Morocco to the Taurus, Cyprus and West Himalayan<br />

forests of Cedrus libani, C. brevifolia and C. deodara,given<br />

that these forests are similar to those of Morocco in terms<br />

of flora and silviculture, and in light of the theory that all<br />

of these forests have a single common origin (Meusel,<br />

1971; Meusel & Schubert, 1971; Quézel 1978).<br />

STUDY AREA<br />

The study area is illustrated in figure 1. The lichen<br />

vegetation of the Cedrus atlantica forests of the Rif mountains<br />

was studied in the area surrounding the cities of Bab-<br />

Besen and Ketama, specifically, in snow stands located<br />

downwind from the principal mountain summits at altitudes<br />

ranging from 1,660m to 1,840m above sea level.<br />

The forests in this area grow on siliceous and marly-schistose<br />

substrata.<br />

The study areas in the Central Atlas mountains are<br />

located near the cities of Azrou (altitude: 1,270m),<br />

Michliffen (approximately 2,000m), and Ifrane (approximately<br />

2,000m), in areas exposed to prevalently oceanic<br />

winds. The forests grow on a calcareous substratum.<br />

The climate of these mountain ranges was described<br />

by Emberger (1939) with the apparently contradictory<br />

term “humid <strong>Mediterranea</strong>n”, with cold winters and a<br />

relatively continental climate. The rainfall varies from<br />

600mm to 1,000mm per year (Despois & Raynal, 1975),<br />

and summers are dry and brief, as observed in those areas<br />

of the <strong>Mediterranea</strong>n Basin with the highest rainfall.<br />

METHODS<br />

The epiphytic lichens were identified both on the<br />

cedars and in laboratory on fragments of bark which were<br />

collected, such as the unknown thalli, using the method<br />

of the “prélèvement partiel” (Clauzade & Roux, 1975;<br />

Roux, 1990).The cover-abundance indices were attributed<br />

according Braun-Blanquet’s scale, modified by Pignatti<br />

(1959), also used by Nimis (1981, 1985) in Canada. Other<br />

indices were integrated after identification of the collected<br />

samples in the laboratory, according to the method.<br />

Most of the relevés were of a qualitative type, both in<br />

order to make comparisons with the works on lichen sociology<br />

used as reference and because the epiphytic cedar<br />

communities consist of fruticose lichens too, which makes<br />

quantitative analysis difficult.<br />

We studied those Cedrus atlantica forests with large,<br />

centuries-old cedars, which were apparently suffering less<br />

from anthropic impact.<br />

Taxonomic references: the nomenclature used was that<br />

of Egea (1996) and Nimis (2000). Pertusaria was identified<br />

in accordance with Sipman (2000); Usnea lapponica<br />

was identified by Clerc, Lepraria elobata by Tønsberg.<br />

Authors’ abbreviations are according to Brummitt and<br />

Powell (1992), phytosociological nomenclature to<br />

Barkman et al. (1976). All samples are preserved in the<br />

Herbarium of the University of Rome “La Sapienza”<br />

(RO).<br />

Syntaxonomical position and phytogeographical references:<br />

we followed the system of Wirth (1995), proposed<br />

for Baden-Württembergs (Austria) and adopted in<br />

Spain by Sarriòn and Burgaz (1995) and Fos (1998).<br />

The syntaxonomic units of Hypogymnietalia that we<br />

used as reference were as follows. At the level of alliance,<br />

Hypogymnion physodis Beschel 1958 also referred to as<br />

Pseudevernion furfuraceae (Barkm.) James et al. 1977, first<br />

described in Great Britain, and to as Parmelion saxatilis<br />

Barkm. 1958 em. Delzenne & Géhu 1997 for France,<br />

rejected by Van Haluwyn & Schumacker (1988). At the<br />

level of association, Pseudevernietum furfuraceae (Hil.) Kalb<br />

1970, which is present throughout Europe, where it characterises<br />

the epiphytic vegetation of both coniferous and<br />

other types of woods of the montane subalpine belt.The<br />

typical association is a community with a central-<br />

European distribution that is non-aerohygrophilous and<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


which requires periods of drought.This syntaxon is common<br />

in the Italian Alps and Apennines; on the Iberian<br />

Peninsula, it has occasionally been found in the subhumid<br />

mesomediterranean belt (Atienza, 1990; Giralt,<br />

1996). At the level of association, we also used Letharietum<br />

vulpinae Gams 1927 as reference; this syntaxon occurs in<br />

subalpine woods of Larix decidua and Pinus cembra with<br />

high continental climate.<br />

All of these communities are acidophilous, and most<br />

of them are nitrophobous, toxiphobous, photophytic and<br />

ombrophilous. In terms of floristic composition, these<br />

communities are characterised by the dominance of the<br />

genera Parmelia s.l., Hypogymnia, Pseudevernia, Usnea,<br />

Alectoria, and Bryoria and by the presence of Letharia,<br />

Evernia, Cetraria, Platismatia, and Parmeliopsis.<br />

RESULTS<br />

We selected 55 relevés from the study areas of the Rif<br />

and Central Atlas mountains, which include a total of 47<br />

taxa, a low number in comparison by European Hypogymnietalia<br />

communities.<br />

The lichen species identified are common in the mountains<br />

of Europe, with the exception of, for example,<br />

Evernia illyrica, Lethariella intricata, Pertusaria multipunctoides,<br />

Hypogymnia laminisorediata, and Pertusaria tazzakensis,<br />

which is only known in Morocco. Nonetheless,<br />

the Moroccan communities are substantially different<br />

from those of Europe, in part for the lack of certain elements<br />

that are typical in the European communities, as<br />

described below.<br />

Hypogymnietalia:The following elements are absent in<br />

the Moroccan forests: i) the montane species with a typically<br />

arctic-alpine and circumboreal distribution, such as<br />

the epiphytic species of the Cetraria genus; ii) the species<br />

Imshaugia aleurites and Parmeliopsis hyperopta, which are<br />

diagnostic of Cetrario pinastri-Parmeliopsidetum ambiguae<br />

(Hil.) Van Haluwyn & Schumacker, 1988; and iii) Hypogymnia<br />

farinacea, H. bitteri, and H. physodes, which represent<br />

important elements of the order Hypogymnietalia.<br />

The finding that these species are lacking throughout the<br />

entire country of Morocco (Egea, 1996) could signify that<br />

there exists a climatic boundary between the oroboreal<br />

and the oromediterranean zone.<br />

Hypogymnion: All of the species of this alliance are present,<br />

obviously with the exception of Cetraria clorophylla.<br />

Pseudevernietum furfuraceae: Certain species that are<br />

faithful and characteristic of Pseudevernia communities<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 51-57<br />

EPIPHYTIC LICHEN VEGETATION OF THE CEDRUS ATLANTICA FORESTS (MOROCCO) ◆<br />

are lacking (e.g. Usnea hirta and the above-mentioned<br />

Hypogymnia).<br />

Letharietum vulpinae:With regard to the definition of<br />

Letharia communities, in accordance with Barkman<br />

(1954), we defined a community as such only when<br />

Letharia vulpina was dominant. In our relevés, various<br />

species that Frey (1952) and Kalb (1970) consider as<br />

characteristic and as companions of Letharietum vulpinae<br />

were missing.<br />

The most interesting lichens for the syntaxonomy of<br />

the cedar forests of Morocco can be summarised as follows:<br />

Hypogymnia laminisorediata, Pseudevernia furfuracea,<br />

and Parmelia saxatilis are the dominant species.<br />

Hypogymnia laminisorediata,whose distribution is limited<br />

to the Balkan mountains and the mountains of North<br />

Africa and Macaronesian region, belongs to the oromediterranean<br />

zone.<br />

Parmelia submontana is associated with Hypogymnia<br />

tubulosa, Phlyctis argena, Anaptychia ciliaris and Usnea lapponica<br />

in the forests of the Central Atlas mountains, which<br />

are humid and dense and where foliose and fruticose<br />

growth-forms prevail.The light is diffuse, despite the fact<br />

that the density of the tree foliage does not allow direct<br />

sunlight to reach the forest floor. Sexual reproduction is<br />

common, even among species that are generally sterile in<br />

Europe, such as Pseudevernia furfuracea, Hypogymnia laminisorediata,<br />

Parmelia saxatilis, and Parmelia submontana.<br />

Pertusaria tazzakensis was found in all of the relevés<br />

from the Rif mountains. A characteristic species combination<br />

of this forest is: Pertusaria tazzakensis associated<br />

with P. multipunctoides, Evernia illyrica, and Lethariella<br />

intricata.<br />

Letharia vulpina, which is easily recognised for its<br />

bright yellow colour, is present in the mountains at low<br />

altitudes, although it has a low degree of cover. In the<br />

Central Atlas mountains, for example in the cedar forests<br />

of Jebel Hebri near Azrou, there exist true communities<br />

of this lichen at 2,000 m, where there is a more marked<br />

continental climate.<br />

Table 1 shows the groups of relevés with the species,<br />

and in particular 21 taxa that are the most representative<br />

of the epiphytic lichen vegetation.<br />

Vegetation types<br />

(relevé type in table 2)<br />

Pseudevernio-Hypogymnietum laminisorediatae ass. nov.<br />

Association characterised by species that prefer an<br />

oceanic climate and by the dominance of Hypogymnia<br />

53


54<br />

◆ GIUSEPPE MASSARI, SONIA RAVERA<br />

Relevés group 1 2 3 4 5 6<br />

typical form var. with Pertusaria var. with Parmelia subass. letharietosum<br />

tazzakensis submontana<br />

Mean circumference of cedars (cm) 400 260 190 210 360 320 440<br />

Mean surface (m 2) 0.6 0.5 0.4 0.4 0.5 0.4 0.6<br />

Height above ground (cm) 100-130 110-140 100-150 90-100 90-120 80-100 90-110<br />

Prevailing exposure N NE NE N NE ENE-NE N-NE<br />

Mean coverage (%) 75 90 70 80 90 70 70<br />

Character-taxon of the association<br />

Pseudevernio-Hypogymnietum laminisorediatae ass. nov.<br />

Hypogymnia laminisorediata D. Hawksw. & Poelt V 1-3 V 2 IV 1-2 V +-2 IV 1-2 V +-2 V +-2<br />

Character-taxa of higher units<br />

Pseudevernia furfuracea (L.) Zopf V1-2 III +-1 V +-2 V1-3 V1-2 V +-3 V1-2 Parmelia saxatilis (L.) Ach. V +-3 IV +-1 IV +-1 V +-1 V1-3 IV + IV +-1<br />

Bryoria fuscescens (Gyeln.) Brodo & D. Hawksw. I + IIr II + III + IIIr-+ IV +-1<br />

Platismatia glauca (L.) W. L. Culb. & C. F. Culb. II1-3 V2-4 V1-3 III +-2<br />

Ochrolechia androgyna (Hoffm.) Arnold I + V +-1 V + IIr-+ III +<br />

Parmelia sulcata Taylor in J. Mackay I + I +<br />

Hypogymnia tubulosa (Schaer.) Hav. V +-1 V +-2<br />

Parmeliopsis ambigua (Wulfen) Nyl. II +<br />

Companions<br />

Lepraria elobata Tønsberg V + II + II + III + IV +<br />

Usnea lapponica Vainio II + III +<br />

var. with Pertusaria tazzakensis<br />

Pertusaria tazzakensis (Werner) Werner V + IV +-1<br />

Evernia illyrica Zahlbr. II r<br />

Lethariella intricata (Moris) Krog II r<br />

Pertusaria multipunctoides Dibben II +<br />

var. with Parmelia submontana<br />

Parmelia submontana Nadv. ex Hale IV +-1 V 1<br />

Phlyctis argena (Spreng.) Flot. V +-1 V +-1<br />

Anaptychia ciliaris (L.) A. Massal. V +-1 IV +-1 I r<br />

Pseudevernio-Hypogymnietum laminisorediatae letharietosum subass. nov.<br />

Letharia vulpina (L.) Hue V r-1 V 1-4 V 2<br />

Hypocenomyce scalaris (Liljeblad) M. Choisy V +-1 V +-2<br />

Lecanora varia (Hoffm.) Ach. I + V + I +<br />

Other taxa. Cladonia fimbriata (L.) Fr. +, Dendriscocaulon umhausense (Auersw.) Degel. +, Rinodina sp. +, Thelenella modesta (Nyl.) Nyl. +, in typicum. Candelariella xanthostigma<br />

(Ach.) Lettau +, in 1. Calicium trabinellum (Ach.) Ach. +, in 2. Ramalina fraxinea (L.) Ach. +, Lecanora albella (Pers.) Ach. +, Buellia griseovirens (Sm.) Almb. +, in 3. Cladonia<br />

fimbriata (L.) Fr. +, Physcia semipinnata (J.F. Gmel.) Moberg +, Candelariella xanthostigma (Ach.) Lettau +, Ramalina fraxinea (L.) Ach. +, Buellia griseovirens (Sm.) Almb. +,<br />

Trapeliopsis granulosa (Hoffm.) Lumbsch +, Lecanora chlarotera Nyl. +, Ramalina farinacea (L.) Ach. 1, Diplotomma alboatrum (Hoffm.) Flot. +, Physconia distorta (With.) J.R.<br />

Laundon +, Physconia perisidiosa (Erichsen) Moberg +, Parmelia glabra (Schaer.) Nyl. 1, Parmelia tiliacea (Hoffm.) Ach. 2, in 4. Parmelia exasperatula Nyl. 1, Xanthoria parietina<br />

(L.) Th. Fr r, Lecanora hagenii (Ach.) Ach +, Calicium trabinellum (Ach.) Ach. +, Lecanora albella (Pers.) Ach. +, Catillaria chalybeia (Borrer) A. Massal. +, Lecidella elaeochroma<br />

(Ach.) Haslz. +, Caloplaca cerinelloides (Erichsen) Poelt r, Physcia semipinnata (J.F. Gmel.) Moberg +, in 5. Chaenothecopsis pusilla (Ach.) A.F.W. Schmidt +, Lepraria sp. +, Cladonia<br />

fimbriata (L.) Fr. +, in 6.<br />

Localities. 1. BAB-BESEN, 1.750-1.840 m, exp. N, slope (°) 10; 2. KETAMA, 1.660-1.720, N, 40; 3. IFRANE, 1.750-1.800, 0; 4. AZROU, 1.750-1.800, 0; 5. JEBEL HEBRI,<br />

1.980-2.010, WSW, 10; 6. MICHLIFFEN, 1.960-2.040, NNW, 40.<br />

Table 1. Synthesis table (55 relevés) of the epiphytic lichen vegetation of the Rif and Central Atlas.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


ecologia mediterranea, tome 28, fascicule 2, 2002, p. 51-57<br />

EPIPHYTIC LICHEN VEGETATION OF THE CEDRUS ATLANTICA FORESTS (MOROCCO) ◆<br />

Pseudevernio-Hypogymnietum laminisorediatae typicum subass.<br />

var. with var. with letharietosum<br />

typical form P. tazzakensis P. submontana<br />

Stand Azrou Michliffen Bab-Besen Ketama Ifrane Azrou Michliffen Jebel Hebri<br />

UTM 50 zone 30, coordinates E 2 96 762 3 12 167 3 47 875 3 54 248 3 05 617 2 99 480 3 12 278 2 99 915<br />

N 36 99 913 37 06 118 38 70 051 38 65 423 37 10 633 36 99 784 37 06 229 36 94 280<br />

Altitude above sea level (m) 1 775 2 040 1 840 1 660 1 755 1 755 2 010 2 010<br />

Stand exposure 0 NNW N N 0 0 NNW WSW<br />

Slope (°) 0 40 10 40 0 0 40 10<br />

Canopy coverage (%) 50 100 40 100 60 90 50 50<br />

Number of relevé 29 53 8 5 12 40 42 16<br />

Date of relevé 2/01/99 31/12/00 2/01/98 2/01/98 7/01/98 6/01/00 27/12/00 28/12/98<br />

Circumference (cm) 555 340 245 180 190 385 435 258<br />

Surface (m 2) 0.63 0.42 0.49 0.48 0.42 0.49 0.5 0.63<br />

Height above ground (cm) 100 100 110 100 100 110 90 80<br />

Exposure N NW N ENE N NNW NNW ENE<br />

Coverage (%) 70 70 100 50 70 100 70 50<br />

Character-taxon of the association<br />

Hypogymnia laminisorediata 2 3 2 + + 1 1 +<br />

Character-taxa of higher units<br />

Pseudevernia furfuracea 1 1 • + 2 2 1 1<br />

Parmelia saxatilis 2 + + 1 1 2 • +<br />

Platismatia glauca • 1 4 2 • • 1 •<br />

Ochrolechia androgyna • + + + • • + +<br />

Bryoria fuscescens • • • • • + + r<br />

Parmelia sulcata • • • • • + • •<br />

Hypogymnia tubulosa • • • • + + • •<br />

Parmeliopsis ambigua • • • • • • + •<br />

Companions<br />

Lepraria elobata + + + • • • + •<br />

Usnea lapponica + • • • • + • •<br />

var. with Pertusaria tazzakensis<br />

Pertusaria tazzakensis • • + + • • • •<br />

Evernia illyrica • • r • • • • •<br />

Lethariella intricata • • r • • • • •<br />

Pertusaria multipunctoides • • + • • • • •<br />

var. with Parmelia submontana<br />

Parmelia submontana • • • • 1 1 • •<br />

Phlyctis argena • • • • 1 1 • •<br />

Anaptychia ciliaris • • • • 1 • • •<br />

Differential-taxa<br />

of the subassociation<br />

Letharia vulpina • • • 1 • • 2 2<br />

Hypocenomyce scalaris • • • • • • 1 1<br />

Lecanora varia • • • • • • • +<br />

Other taxa<br />

Dendriscocaulon umhausense + • • • • • • •<br />

Thelenella modesta + • • • • • • •<br />

Cladonia fimbriata + • • • • • • •<br />

Candelariella xanthostigma • • + • • • •<br />

Calicium trabinellum • • • + • • • +<br />

Ramalina fraxinea • • • • + r • •<br />

Buellia griseovirens • • • • + • • •<br />

Ramalina farinacea • • • • • 1 • •<br />

Chaenothecopsis pusilla • • • • • • + •<br />

Lecanora albella • • • • • • • +<br />

Catillaria chalybeia • • • • • • • +<br />

Table 2. Relevé type of the syntaxa.<br />

55


56<br />

◆ GIUSEPPE MASSARI, SONIA RAVERA<br />

laminisorediata which replaces in cedars forest the<br />

Hypogymnia species of the montane and subalpine coniferous<br />

woods. The community is a phytoclimatic indicator<br />

of the humid montane forests of Cedrus atlantica in<br />

the Central Atlas and Rif mountains, with a relatively<br />

continental climate.<br />

var. with Pertusaria tazzakensis. Geographic variant<br />

characterised by the endemic Pertusaria tazzakensis, by<br />

Platismatia glauca with a high degree of cover, by the<br />

dominance of Ochrolechia androgyna, and by the sporadic<br />

Evernia illyrica and Lethariella intricata, which are rare<br />

suboceanic species. It is present in the cedar forests of the<br />

Rif mountains, at an altitude of 1,660-1,840 m, on the<br />

northern slopes of the mountains.<br />

The aerohygrophilous aspect of the Pseudevernia communities<br />

of Ochrolechia androgyna and Platismatia glauca<br />

has been well documented for Central Europe and has<br />

been described as a distinct association [Ochrolechio<br />

androgynae-Platismatietum glaucae (Hil.) Delzenne et al.,<br />

1978] and subassociation (Parmelietum furfuraceae Hil.<br />

1925, sensu Ochsner 1928, ochrolechietosum androgynae<br />

Barkm. 1958). The combination of these species on<br />

cedars is typical; thus they cannot be categorised into<br />

these syntaxa.<br />

var. with Parmelia submontana. Geographic variant of<br />

the dense cedar forests with thick ground cover found in<br />

the Central Atlas mountains, characterised by the dominance<br />

of Parmelia submontana, Phlyctis argena, and<br />

Hypogymnia tubulosa, found at lower altitudes with respect<br />

to the communities of Letharia vulpina, in habitats<br />

in which certain species of Ramalina, Physconia and<br />

Parmelia, are present, most of which are nitrophilous.<br />

Pseudevernio-Hypogymnietum<br />

laminisorediatae letharietosum subass. nov.<br />

Stable subassociation, with a continental, xerophile,<br />

and heliophile affinity. This subassociation characterises<br />

the lower-cover and luminous cedar forests which are prevalently<br />

at the top of the mountains, at altitudes of<br />

2,000 m, on cedars with a mean circumference of 380 cm.<br />

The erosion caused by wind results in the fragmentation<br />

of the thalli and favours the growth of Hypocenomyce<br />

scalaris, which firmly adheres to bark, and the vegetative<br />

propagation of Letharia vulpina.<br />

Those species that are characteristic of higher units,<br />

which require greater humidity, are missing.<br />

Class: Hypogymnietea physodis Follm. 1974<br />

Order: Hypogymnietalia physodo-tubulosae Barkm. 1958 (!Parmelietalia<br />

physodo-tubulosae) / syn.: Alectorietalia (Dahl & Hada÷, 1944)<br />

Alliance: Pseudevernion furfuraceae (Barkm.) (James et al., 1977) / syn.:<br />

Hypogymnion physodis Beschel 1958 (!Parmelion physodis)<br />

Association: Pseudevernio-Hypogymnietum laminisorediatae ass. nov.<br />

var. with Pertusaria tazzakensis<br />

var. with Parmelia submontana<br />

subassociation: Pseudevernio-Hypogymnietum laminisorediatae letharietosum<br />

subass. nov.<br />

Figure 1. Study<br />

area in Morocco.<br />

Geographical locations:<br />

1. Azrou,<br />

2. Michliffen,<br />

3. Jebel Hebri,<br />

4. Ifrane,<br />

5. Bab-Besen,<br />

6. Ketama.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


CONCLUSION<br />

This analysis of the epiphytic lichen vegetation of the<br />

Moroccan cedar forests, together with the comparison<br />

with the floristically similar communities of the European<br />

mountains, allowed us to identify vegetation that is both<br />

original and specific. In particular, we observed a new<br />

association, Pseudevernio-Hypogymnietum laminisorediatae,<br />

a first variant with Pertusaria tazzakensis on the Rif<br />

mountains, a second variant with Parmelia submontana on<br />

the Central Atlas mountains, and the subassociation<br />

Pseudevernio-Hypogymnietum laminisorediatae letharietosum<br />

in the forests at high altitudes, with continental climate,<br />

where Letharia vulpina is the differential species<br />

with respect to the European Letharia communities.<br />

The differences between the Moroccan and the<br />

European communities obviously depend on climatic,<br />

ecological, and phytogeographic factors, which are strictly<br />

related to the <strong>Mediterranea</strong>n character of the Cedrus atlantica<br />

forests.These forests, although reduced to fragments<br />

and surrounded by areas where human activity is quite<br />

intense, constitute extremely interesting conserved habitats,<br />

which need to be actively protected.<br />

Bibliography<br />

ATIENZA V.E., 1990. Flora y vegetación liquénica epifítica de las<br />

comarcas de Els Ports y Baix Maestrat (Castellón) y territorios<br />

próximos. PhD Thesis. Universitat de València.<br />

BARKMAN J.J., 1954. Zur Kenntnis einiger Usneion-Assoziationen<br />

in Europa. Vegetatio 4: 309-333.<br />

BARKMAN J.J., MORAVEC J. & RAUSCHERT S., 1976. Code of phytosociological<br />

nomenclature. Vegetatio 32: 131-185.<br />

BRUMMITT R.K. & POWELL C.E. (EDS.), 1992. Authors of plant<br />

names. Royal Botanic Gardens, Kew. 732 p.<br />

CLAUZADE G. & ROUX C., 1975. Étude écologique et phytosociologique<br />

de la végétation lichénique des roches calcaires<br />

non altérées dans les régions méditerranéenne et subméditerranéenne<br />

du sud-est de la France. Bull. Mus. Hist. Nat.<br />

Marseille 35 : 153-208.<br />

DESPOIS J. & RAYNAL R., 1975. Géographie de l’Afrique du nordouest.<br />

Payot, Paris. 570 p.<br />

EGEA J.M., 1996. Catalogue of lichenized and lichenicolous fungi<br />

of Morocco. Bocconea 6: 19-114.<br />

EMBERGER L., 1939. Aperçu général sur la végétation du Maroc.<br />

In: Rübel E. & Lüdi W., Ergebnisse der Internationalen<br />

Pflanzengeographischen Exkursion durch Marokko und<br />

Westalgerien 1936. Verlag Hans Huber, Bern.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 51-57<br />

EPIPHYTIC LICHEN VEGETATION OF THE CEDRUS ATLANTICA FORESTS (MOROCCO) ◆<br />

FOS S., 1998. Líquenes epífitos de los alcornocales ibéricos.<br />

Correlaciones bioclimáticas, anatómicas y densimétricas con<br />

el corcho de reproducción. Guineana 4: 1-507.<br />

FREY E., 1952. Die Flechtenflora und -Vegetation des<br />

Nationalparks im Unterengadin.Die diskokarpen Blatt- und<br />

Strauchflechten. Ergebn. wiss. Untersuch. schweiz. Nat.-Parks<br />

3: 361-504.<br />

GIRALT M., 1996. Líquens epífits i contaminació atmosfèrica a la<br />

plana i les serralades litorals tarragonines. Institut d’Estudis<br />

Catalans. ASC, 113. Barcelona. 525 p.<br />

KALB K., 1970. Flechtengesellschaften der Vorderen Ötztaler<br />

Alpen. Dissertationes Botanicæ, 9.118 p.<br />

MEUSEL H., 1971. <strong>Mediterranea</strong>n elements in the flora and vegetation<br />

of the West Himalayas. In: Davis P.H., Plant Life of<br />

South-West Asia. Edinburgh.<br />

MEUSEL H. & SCHUBERT R., 1971. Beiträge zur Pflanzengeographie<br />

des Westhimalayas. Flora 160: 137-194, 373-432, 573-<br />

606.<br />

NIMIS P.L., 1981. Epigaeous lichen synusiae in the Yukon territory.<br />

Cryptogamie, Bryol. Lichénol. 2: 127-151.<br />

NIMIS P.L., 1982. The epiphytic lichen vegetation of the Trieste<br />

province (North Eastern Italy). Studia Geobot. 2: 169-191.<br />

NIMIS P.L., 1985. Phytogeography and Ecology of Epiphytic<br />

Lichens at the Southern Rim of the Clay Belt (N-Ontario,<br />

Canada). The Bryologist 88 (4): 315-324.<br />

NIMIS P.L., 2000. Checklist of the Lichens of Italy 2.0.,<br />

University of Trieste, Dept. of Biology, IN2.0/2<br />

(http://dbiodbs.univ.trieste.it/).<br />

PIGNATTI S., 1959. Fitogeografia. In: Cappelletti C., Trattato di<br />

Botanica. UTET, Torino.<br />

QUÉZEL P., 1978. Analysis of the flora of mediterranean and saharan<br />

Africa. Ann. Missouri Bot. Gard. 65: 479-534.<br />

ROUX C., 1990. Échantillonnage de la végétation lichénique et<br />

approche critique des méthodes de relevé. Cryptogamie,Bryol.<br />

Lichénol. 11: 95-108.<br />

SARRIÓN F.J. & BURGAZ A.R., 1995. Comunidades lignícolas del<br />

sector central de Sierra Morena (SW de España).<br />

Cryptogamie, Bryol. Lichénol. 16: 137-144.<br />

SIPMAN H., 2000. Lichen determination keys. Pertusaria in<br />

Europa and the <strong>Mediterranea</strong>n. Botanischer Garten und<br />

Botanisches Museum Berlin-Dahlem, Freie Universität Berlin<br />

(http://www.bgbm.fu-berlin.de/sipman/keys/Perteuro.htm).<br />

VAN HALUWYN C. & SCHUMACKER R., 1988. Les groupements<br />

à Parmeliopsis ambigua, Cetraria pinastri et C. sepincola dans<br />

le nord de la France, en Belgique et dans les régions limitrophes.<br />

Synthèse critique du Pseudevernion furfuraceae en<br />

Europe. Doc. Phytosociol., N.S. 11: 125-151.<br />

WIRTH V., 1995. Die Flechten Baden-Württembergs. Eugen Ulmer<br />

GmbHr & Co., Stuttgart. 1006 p.<br />

57


Population structure and habitat use in relation<br />

to body size in the grass snake (Natrix natrix) from<br />

a fragmented habitat of <strong>Mediterranea</strong>n central Italy<br />

Structure des populations et utilisation de l’habitat en relation<br />

avec la masse corporelle chez la couleuvre à collier (Natrix natrix)<br />

dans un habitat fragmenté d’Italie centrale méditerranéenne<br />

Lorenzo Rugiero 1, Massimo Capula 2, Luca Luiselli 3-4<br />

1. FIZV, via Domenico Cimarosa 13, I-00198 Roma, Italy<br />

2. Museo Civico di Zoologia, via Ulisse Aldrovandi 18, I-00197 Roma, Italy – email: m.capula@comune.roma.it<br />

3. Environmental Studies Centre Demetra and FIZV, via dei Cochi 48/B, I-00133 Roma, Italy<br />

4. Ecology Research Group, TSKJ Nigeria Ltd, 172 A Aba Road, Port Harcourt, Rivers State, Federal Republic of Nigeria<br />

Abstract<br />

Annual variation of the population structure and habitat use in<br />

relation to sex and body size of the grass snake, Natrix natrix, were<br />

studied in a fragmented agro-forest habitat of <strong>Mediterranea</strong>n central<br />

Italy. The study was carried out in two years (1995 and 1998).<br />

The population structure was different in the two years, as the proportion<br />

of subadults was much higher in 1995 than in 1998. Males<br />

were captured slightly more frequently than females, but the adult<br />

sex-ratio was similar in the two study years. The density of snakes<br />

felt dramatically of nearly 40% in the second study year, due to a<br />

clear decrement in the number of captured subadults. No specimen<br />

firstly marked in 1995 was recaptured in 1998. Natrix natrix<br />

individuals used the various basking sites available in the study<br />

area in a different way, depending on their body size. Some hypotheses<br />

to explain the size-based partitioning of the available habitat<br />

are presented and discussed.<br />

Key-words<br />

Natrix natrix, snakes, habitat fragmentation, population structure,<br />

sex-ratio, Italy<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 59-64<br />

Résumé<br />

Les variations annuelles dans la structure de population et l’utilisation<br />

de l’habitat en relation avec le sexe et la taille corporelle des<br />

individus ont été étudiées chez la couleuvre à collier Natrix natrix<br />

dans un milieu agro-forestier d’Italie centrale méditerranéenne.<br />

Cette étude a été conduite sur deux années de comparaison (1995<br />

et 1998). Les structures de population se sont avérées différentes<br />

entre les deux années, avec une proportion de subadultes nettement<br />

plus élevée en 1995 qu’en 1998. Les mâles ont globalement été capturés<br />

plus fréquemment que les femelles, mais le sex-ratio des adultes<br />

n’a pas varié entre les deux années d’étude. La densité des serpents<br />

a chuté de près de 40 % en 1998 du fait de la diminution nette<br />

des captures de subadultes. Aucun individu initialement marqué<br />

en 1995 n’a pu être recapturé en 1998. La fréquentation des sites<br />

de thermo-régulation disponibles dans la zone d’étude est variable<br />

en fonction de la taille corporelle des différents individus de Natrix<br />

natrix. Plusieurs hypothèses susceptibles d’expliquer le patron observé<br />

sont présentées et discutées.<br />

Mots-clés<br />

Natrix natrix, serpents, fragmentation de l’habitat, structure de<br />

population, sex-ratio, Italie<br />

59


60<br />

◆ LORENZO RUGIERO, MASSIMO CAPULA, LUCA LUISELLI<br />

INTRODUCTION<br />

In the last fifty years, enormous modifications of the<br />

pristine <strong>Mediterranea</strong>n landscape have been produced<br />

in the Italian peninsula, in particular by fragmenting the<br />

pristine wooded surface into a patchy mosaic of “wooded<br />

islands” interspersed inside a “sea” of areas modified<br />

by human impact (Mazzoleni & Spada, 1992). The<br />

extensive fragmentation of the natural landscape caused<br />

catastrophic consequences on several species of the vertebrate<br />

fauna, including several reptiles (see Andreone &<br />

Sindaco, 1998; Bologna et al., 2000). With regard to<br />

snakes, Filippi and Luiselli (2000) have pointed out that<br />

distribution-related factors, i.e. habitat loss and fragmentation,<br />

are the prevailing threats in Italy for more<br />

than 50% of the species.<br />

Since several years, we are studying the effects of habitat<br />

fragmentation both on the ecology of snakes of<br />

<strong>Mediterranea</strong>n central Italy (Luiselli & Capizzi, 1997)<br />

and the extinction risk of the various species (Filippi &<br />

Luiselli, 2000). In this paper, we analyse variation of some<br />

aspects of the population structure and habitat use of a<br />

population of grass snake (Natrix natrix Laurenti) inhabiting<br />

a typically fragmented and disturbed agro-forest<br />

habitat of <strong>Mediterranea</strong>n central Italy, in two different<br />

non-consecutive years. Our study aims to verify whether<br />

the population structure and the habitat use of Natrix<br />

natrix may vary substantially from year to year in a fragmented<br />

habitat. Natrix natrix was chosen as model species<br />

because (i) it is still relatively abundant in the study<br />

area despite the high levels of habitat disturbance, (ii) it<br />

is well known as a very adaptable species (Bruno &<br />

Maugeri, 1990; Luiselli & Capizzi, 1997), and (iii) it is<br />

one of the best known species of snakes in terms of ethoecology<br />

either in Italy (Luiselli & Rugiero, 1991; Gentilli<br />

& Zuffi, 1995; Luiselli, 1996; Luiselli et al., 1997) or in<br />

other European countries (Madsen, 1983, 1984; Madsen<br />

& Shine, 1993).<br />

MATERIALS AND METHODS<br />

The field study was carried out at an agro-forest area<br />

of central Italy (Vejo, province of Rome, approximately<br />

70 m a.s.l), that was also used by us for other snake studies<br />

(Capula et al., 2000). We investigated the structure<br />

of a Natrix natrix population by capture-mark-recapture<br />

procedure, in 1995 and 1998. In total, 166 days were spent<br />

in the field.<br />

The area inhabited by Natrix natrix was about 8.75<br />

ha. The area is sited inside a suburban sector of the city<br />

of Rome, and it is characterized by fragments of mixed<br />

oak forests surrounded by roads, houses, cultivations and<br />

grasslands. Natrix natrix occurs in a remnant thick oak<br />

forest, growing along the banks of a small stream. Inside<br />

this area, Natrix natrix individuals were constrained to<br />

bask only in five small sites (we named clearing “A”, “B”,<br />

“C”, “D”, “E”), that were subject to strong disturbance<br />

by humans and cars, especially at night.The general habitat<br />

conditions did not vary between the two study years.<br />

The five basking clearings had an almost identical grassy<br />

vegetation structure, but the clearings “A” and “B” were<br />

more wet than the clearings “C”, “D”, and “E”. This is<br />

clearly explained by the different average linear distance<br />

(m) between the various clearings and the stream banks<br />

(figure 1; one-way ANOVA: F = 8.710, d.f. = 4, 50, P =<br />

0.00002). Four clearings (“A”, “B”, “C”, “D”) were surveyed<br />

in both the study years, while the clearing “E” was<br />

surveyed only in 1998.<br />

Natrix natrix individuals were searched for by standardized<br />

routes conducted throughout each microhabitat<br />

type available in the study area.When captured, each<br />

individual was permanently marked by ventral scale-clipping<br />

(Brown & Parker, 1976). Then, it was painted with<br />

a white number on the back for easy individual identification<br />

within short timespans without necessity to further<br />

recapture it (see Luiselli, 1995; Luiselli et al., 1996, 1997).<br />

The captured individuals were routinarily measured to<br />

snout-vent length (SVL, to ± 0.5 cm precision) and sexed.<br />

For each capture event we recorded the exact spot of capture,<br />

the time of observation (Rome local time), and the<br />

type of activity exhibited by the snake.<br />

Population size was estimated, independently for the<br />

two study years, by using two different indexes, in a way<br />

to give a better approximation, according to Seber (1982).<br />

The two indexes are the following:<br />

[1] N = A x n / a (Lincoln-Petersen index);<br />

[2] N = A (n + 1) / (a + 1);<br />

where N is the population size, A is the total number<br />

of marked individuals, n is the total number of observed<br />

individuals, and a is the total number of recaptured individuals.<br />

Statistical analyses (χ2 test, ANOVAs followed by Tukey<br />

HSD post-hoc tests to identify direction trends of multiple<br />

comparisons) were performed using SPSS and STA-<br />

TISTICA for Windows PC package (version 4.3B). All<br />

tests were two-tailed, with alpha set at 5 %.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


POPULATION STUCTURE AND HABITAT USE IN RELATION TO BODY SIZE IN THE GRASS SNAKE (NATRIX NATRIX)… ◆<br />

RESULTS<br />

Population size, density and sex-ratio<br />

In 1995, 45 individuals were captured and marked (11<br />

males, 6 females, 28 subadults), 20 were observed and<br />

not captured, and 8 individuals were recaptured. In 1998,<br />

33 individuals were captured and marked (16 males, 8<br />

females, 9 subadults), 20 were observed, and 10 recaptured.<br />

Pooling data from the two years, the number of<br />

males exceeded that of females, but not at a statistically<br />

significant level (χ 2 = 3.12, d.f. = 1, P= 0.06).The apparent<br />

sex-ratio was similar in the two study years (1.8:1 in<br />

1995, 2:1 in 1998).The frequency of subadults was significantly<br />

higher in 1995 than in 1998 (χ 2= 27.58, d.f. = 1,<br />

P = 0.00001).<br />

The estimated rough population size was similar using<br />

both the indexes 1 and 2 ([1], N = 112 individuals in<br />

1995; N = 66 individuals in 1998; [2], N = 105 individuals<br />

in 1995; N = 63 individuals in 1998). From these<br />

estimates it seems that the Natrix natrix population experienced<br />

a very serious decline between 1995 and 1998<br />

(about 41% following the index [1], and 40% following<br />

the index [2]).<br />

The estimated rough density was 12-13 individuals ×<br />

ha -1 in 1995, and 7.2-7.5 individuals × ha -1 in 1998. If we<br />

consider only the adult specimens, the density was about<br />

5 individuals × ha -1 in 1995, and 4.5-5 individuals × ha -1<br />

in 1998. No specimen marked in 1995 was recaptured in<br />

1998.<br />

Habitat use in relation to sex and body size<br />

The frequencies of observation of males and females<br />

Natrix natrix in the five basking sites surveyed during the<br />

present study are presented in table 1. The inter-sexual<br />

difference in clearing utilization was not statistically significant<br />

in 1995 (χ 2 = 7.13, d.f. = 3, P = 0.068), while it<br />

Basking clearing 1995 1995 1998 1998<br />

males females males females<br />

A 9 (30%) 6 (35.3%) 2 (9.5%) 3 (21.4%)<br />

B 8 (26.7%) 6 (35.3%) 2 (9.5%) 3 (21.4%)<br />

C 5 (16.7%) 1 (5.9%) 6 (28.6%) 1 (7.1%)<br />

D 8 (26.7%) 4 (23.5%) 7 (33.3%) 6 (42.8%)<br />

E - - 4 (19%) 1 (7.1%)<br />

Table 1. Frequency of observation of Natrix natrix males and females<br />

in the five basking clearings in 1995 and in 1998.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 59-64<br />

was significant in 1998 (χ 2 = 27.2, d.f. = 4, P < 0.001).<br />

In the latter year, males especially used as basking sites<br />

the clearings “C” and “D”, while females the clearing “D”<br />

(table 1).<br />

The mean snout-vent length of the Natrix natrix individuals<br />

differed at a statistically significant level among<br />

the various basking clearings either in 1995 (Kruskal-<br />

Wallis ANOVA, P < 0.05) or in 1998 (Kruskal-Wallis<br />

ANOVA, P < 0.001; figure 2). A Tukey HSD post-hoc<br />

test revealed that (i) in 1995 we captured mainly large<br />

sized specimens at the clearing “D”, and (ii) in 1998 the<br />

large sized specimens were again captured frequently at<br />

the clearing “D”, but also at the clearing “E”.<br />

DISCUSSION<br />

Important aspects highlighted by our study are (i) the<br />

drastic decrement in the Natrix natrix population size<br />

from 1995 to 1998, and (ii) the remarkable inter-annual<br />

variation in the juvenile portion of the marked population.<br />

The drastic decrement of the population size<br />

(approximately 40%) was almost exclusively due to the<br />

latter factor, as the number of marked adults remained<br />

more or less constant in the two study years.<br />

Two years without surveys (i.e. 1996, 1997) were<br />

enough to have a complete turnover of the individuals, as<br />

no specimen captured in 1995 was recaptured three years<br />

later. This may be due (i) to the dispersal tendency and<br />

the wide home-range of the species (Madsen, 1984), (ii)<br />

to the regeneration of the “marked” ventral scales, that<br />

may confound identification of recaptured specimens<br />

when a long timespan passes throughout two capture<br />

events (Luiselli & Capula, unpublished data), (iii) to the<br />

different rate of birth in the years without surveys (1996,<br />

1997), and (iv) to unusually higher mortality rates due to<br />

habitat disturbance by man in the study area.<br />

Density of the Natrix natrix population was intermediate<br />

between that observed in populations from northwestern<br />

Italy (Gentilli & Zuffi, 1995) and that recorded<br />

in populations from central Germany (Mertens, 1995).<br />

The adult sex-ratio was similar to that observed in conspecific<br />

populations by other Authors (see, e.g., Gentilli &<br />

Zuffi, 1995; Mertens, 1995; Luiselli, 1996; Luiselli et al.,<br />

1997). It was also in good agreement with the adult sexratio<br />

expected in a species with an operational sex-ratio<br />

skewed towards males (see, e.g., Madsen & Shine, 1993;<br />

Luiselli, 1996). Considering that in both the study years<br />

we surveyed the area along the whole activity period of<br />

61


62<br />

◆ LORENZO RUGIERO, MASSIMO CAPULA, LUCA LUISELLI<br />

Figure 2. Averages (and dispersion measures)<br />

of the Natrix natrix snout-vent length (SVL)<br />

in the five basking clearings,<br />

in 1995 (A) and in 1998 (B).<br />

For statistical details, see text.<br />

Figure 1. Averages (and dispersion measures)<br />

of the linear distance (in m) between the five basking<br />

clearings and the banks of a stream.<br />

For statistical details, see text.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


POPULATION STUCTURE AND HABITAT USE IN RELATION TO BODY SIZE IN THE GRASS SNAKE (NATRIX NATRIX)… ◆<br />

the snakes, it is likely that the observed adult sex-ratio is<br />

a true reflection of the actual population structure (Parker<br />

& Plummer, 1987).<br />

The tendency toward a snake-size based selection of<br />

basking clearings is noteworthy and deserves further<br />

research.This trend was consistent in both the study years,<br />

which means that a given category of individuals tended<br />

to select a given clearing in both years. Apart for differences<br />

in mean humidity, the vegetation structure of the<br />

five basking clearings was practically identical, which led<br />

us to reject the possibility that the snake-size based habitat<br />

partitioning was simply the effect of diverging catchability<br />

patterns among snake body size classes.We can<br />

put forward the hypothesis that the different size categories<br />

of snakes selected different basking clearings in relation<br />

to their ecological needs, i.e. the need of encountering<br />

precise prey types (e.g., adults of Bufo bufo for the<br />

larger specimens, juveniles Bufo bufo and Rana italica for<br />

young individuals and smaller adults), and the need of<br />

finding good oviposition sites by the gravid females.The<br />

evidence for this hypothesis is that in the basking clearings<br />

“D” and “E”, where the larger Natrix natrix individuals<br />

were captured, adult specimens of Bufo bufo were<br />

much more common than in the other three clearings. On<br />

the other hand, Rana italica and newly metamorphosed<br />

Bufo bufo were especially abundant in the clearings “A”,<br />

“B” and “C” (Rugiero, unpublished data). At this regard,<br />

it must be stressed that the Italian Natrix natrix populations<br />

investigated to date are known to feed on tadpoles<br />

and small frogs during the first years of life, and on adult<br />

toads during the late maturity years (Luiselli et al., 1997).<br />

Concerning the inter-sexual differences in the utilization<br />

of the five basking clearings, the pattern emerging<br />

from our data is less obvious. In one of the study years<br />

(1995) the two sexes exhibited similar basking clearing<br />

selection, while in the other study year (1998) there was<br />

a significant inter-sexual difference in basking clearing<br />

utilization. We think that further research is necessary in<br />

order to give firm conclusions, especially in consideration<br />

of the fact that inter-sexual variation in life-history traits<br />

is one of the major problems in contemporary snake<br />

research (see, e.g., Shine, 1986, 1991, 1994; Parker &<br />

Plummer, 1987; Houston & Shine, 1993).<br />

ACKNOWLEDGEMENTS<br />

We thank Dr Dario Capizzi (Roma) and Dr Francesco<br />

M. Angelici (Roma) for helpful discussion on several<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 59-64<br />

issues of this paper. Dr Marco A.L. Zuffi (Pisa) kindly<br />

revised the final version of the manuscript. The field<br />

research was supported by a grant (X98NNP) by<br />

“Demetra SpA” and by MURST 40% funds (grant<br />

‘Variazione geografica e diversità a livello di specie, faune<br />

e zoocenosi: cause storiche ed ecologiche’).<br />

References<br />

ANDREONE F. & SINDACO R., (eds.) 1998. Erpetologia del Piemonte<br />

e della Valle d’Aosta. Monografie del Museo Regionale di<br />

Scienze Naturali, Torino<br />

BOLOGNA M.A., CAPULA M. & CARPANETO G.M., (eds.) 2000.<br />

Anfibi e Rettili del Lazio. Fratelli Palombi Editori, Roma.<br />

BROWN W.S. & PARKER W.S., 1976. A ventral scale clipping system<br />

for permanently marking snakes. J. Herpetol. 10: 247-<br />

249.<br />

BRUNO S. & MAUGERI S., 1990. Serpenti d’Italia e d’Europa.<br />

Editoriale Giorgio Mondadori, Milano.<br />

CAPULA M., FILIPPI E., LUISELLI L., AGUILAR J.R. & RUGIERO<br />

L., 2000. Body size and some demographic characteristics<br />

in two populations of Coluber viridiflavus in the countryside<br />

of Rome. In: Giacoma C. (ed.), Atti I Congresso Nazionale<br />

Societas Herpetologica Italica, Museo Regionale di Scienze<br />

Naturali, Torino: 435-438.<br />

FILIPPI E. & LUISELLI L., 2000. Status of the Italian snake fauna<br />

and assessment of conservation threats. Biol. Conserv. 93:<br />

219-226.<br />

GENTILLI A. & ZUFFI M.A.L., 1995. Structure of a Natrix natrix<br />

population from Northern Italy. In: Llorente G.A., Montori<br />

A., Santos X. & Carretero M.A. (eds.), Scientia Herpetologica,<br />

AGAL, Barcelona: 241-243.<br />

HOUSTON D.L. & SHINE R., 1993. Sexual dimorphism and niche<br />

divergence: feeding habits of the Arafura filesnake. J. Ani.<br />

Ecol. 43: 321-336.<br />

LUISELLI L., 1995. The mating strategy of the European adder,<br />

Vipera berus. Acta Œcol. 16: 375-388.<br />

LUISELLI L., 1996. Individual success in mating balls of the grass<br />

snake, Natrix natrix: size is important. J. Zool.(London) 239:<br />

731-740.<br />

LUISELLI L. & CAPIZZI D., 1997. Effects of area, isolation, and<br />

habitat features on distribution of snakes in <strong>Mediterranea</strong>n<br />

fragmented woodlands. Biodiv. Conserv. 6:1339-1351.<br />

LUISELLI L., CAPULA M. & SHINE R., 1996. Reproductive output,<br />

costs of reproduction and ecology of the smooth snake, Coronella<br />

austriaca, from the eastern Italian Alps. Œcologia 106: 100-110.<br />

LUISELLI L., CAPULA M. & SHINE R., 1997. Food habits, growth<br />

rates, and reproductive biology of grass snakes, Natrix natrix<br />

(Colubridae) in the Italian Alps. J. Zool. (London) 241:<br />

371-380.<br />

63


64<br />

◆ LORENZO RUGIERO, MASSIMO CAPULA, LUCA LUISELLI<br />

LUISELLI L. & RUGIERO L., 1991. Food niche partitioning by<br />

water snakes (genus Natrix) at a freshwater environment of<br />

central Italy. J. Freshwater Ecol. 6: 439-444.<br />

MADSEN T., 1983. Growth rates, maturation and sexual size<br />

dimorphism in a population of grass snakes, Natrix natrix,<br />

in southern Sweden. Oikos 40: 277-282.<br />

MADSEN T., 1984. Movements, home range size and habitat use<br />

of radio-tracked grass snakes (Natrix natrix) in southern<br />

Sweden. Copeia 1984: 707-713.<br />

MADSEN T. & SHINE R., 1993. Male mating success and body<br />

size in European grass snakes. Copeia 1993: 561-564.<br />

MAZZOLENI S. & SPADA F., 1992. Deciduous broadleaved versus<br />

evergreen sclerophyllous forest. Disturbance and local<br />

shifting dominance in <strong>Mediterranea</strong>n environments. In:Teller<br />

A., Mathy P. & Jeffers J.N.R. (eds.), Responses of Forest<br />

Ecosystems to Environmental Changes, Elsevier Science<br />

Publishers, Paris: 839-842.<br />

MERTENS D., 1995. Population structure and abundance of grass<br />

snakes, Natrix natrix, in central Germany. J. Herpetol. 29:<br />

454-456.<br />

PARKER W.S. & PLUMMER M.V., 1987. Population ecology. In:<br />

Seigel R.A., Collins J.T. & Novak S.S. (eds.), Snakes:<br />

Ecology and Evolutionary Biology, MacMillan, New York:<br />

253-301.<br />

SEBER G.A.F., 1982. The Estimation of Animal Abundance and<br />

Related Parameters. Charles Griffin & Co. Ltd., London.<br />

SHINE R., 1986. Sexual differences in morphology and niche utilization<br />

in an aquatic snake, Acrochordus arafurae. Œcologia<br />

69: 260-267.<br />

SHINE R., 1991. Intersexual dietary divergence and the evolution<br />

of sexual dimorphism in snakes. Am. Nat. 138: 103-122.<br />

SHINE R., 1994. Sexual size dimorphism in snakes revisited.<br />

Copeia 1994: 326-346.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


The colonisation of the Great Spotted Woodpecker<br />

(Picoides major L.) in Eucalypt woods<br />

and Popular Cultivations in Sicily 1<br />

La colonisation des plantations d’eucalyptus et de peupliers<br />

en Sicile par le pic épeiche (Picoides major L.)<br />

Tommaso La Mantia 1, Michele Spoto 2,Bruno Massa 3<br />

1. Dipartimento di Colture Arboree, Universit degli Studi, Viale delle Scienze 11, 90128 Palermo I; email: lamantia@unipa.it<br />

2. Stazione di Inalellamento, c/o Dipartimento Senfimizo, Vialle delle Scienze 13, 90128 Palermo I<br />

3. Entomologia, Acarologia and Zoologia, Dipartimento Senfimizo, Vialle delle Scienze 13, 90128 Palermo I; email: zoolappl@unipa.it<br />

Summary<br />

Thank to the forestry activity, in the last half of the century the<br />

wood surface in Sicily has increased. The species planted were<br />

<strong>Mediterranea</strong>n pine and Eucalyptus spp., the latter were used<br />

because considered more adapted to marginal soils. In comparison<br />

with some natural formations, Eucalyptus plantations appear quite<br />

poor either from floristic and faunistic point of view. Nevertheless,<br />

in the last years, some Australian species of insects, linked to<br />

Eucalyptus, colonised the European continent, contributing to<br />

increase the biodiversity of Eucalypt afforestations. The recent diffusion<br />

of the Great Spotted Woodpecker (Picoides major L.) in<br />

these woods is part of this process, which involved also other types<br />

of wood plantations, like poplar cultivation. In the Eucalypt afforestations,<br />

the Great Spotted Woodpecker exploits Eucalyptus, both<br />

extracting from them larvae of the Cerambycidae beetle Phoracantha<br />

semipunctata and drinking the sap oozing out through holes made<br />

along the trunk. This bird species is actually spreading in Pinus,<br />

Eucalytpus and Populus woods close to natural and semi-natural<br />

formations, as well as in isolated woods, using almond and olive<br />

orchards as biocorridors.<br />

Key-words<br />

Eucalyptus, Populus, Picoides major, Phoracantha semipunctata,<br />

Sicily<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 65-73<br />

Résumé<br />

Du fait d’une importante activité de reboisement durant la dernière<br />

moitié du siècle passé, la superficie des forêts en Sicile a considérablement<br />

augmenté. Les espèces utilisées pour le reboisement ont<br />

surtout été des pins méditerranéens et différentes espèces d’eucalyptus<br />

; ces dernières ayant souvent été préférées car considérées comme<br />

mieux adaptées à des conditions de sol difficiles. Comparées aux<br />

formations forestières naturelles, les plantations d’eucalyptus sont<br />

généralement plus pauvres tant d’un point de vue floristique que<br />

faunistique. Néanmoins, quelques espèces d’insectes australiens inféodés<br />

aux eucalyptus ont également colonisé le continent européen,<br />

contribuant ainsi à l’augmentation de la biodiversité de ces boisements.<br />

La colonisation récente de ces boisements d’eucalyptus ainsi<br />

que des plantations de peupliers par le pic épeiche (Picoides<br />

major L.) fait partie du même processus. Les facteurs qui ont favorisé<br />

cette colonisation du pic sont d’une part, la présence du coléoptère<br />

cérambycide Phoracantha semipunctata, dont les larves vivant<br />

dans les troncs des eucalyptus sont particulièrement recherchées par<br />

le Pic et, d’autre part, la consommation de la sève d’eucalyptus qui<br />

suinte au travers des trous creusés par l’oiseau le long du tronc. Cette<br />

espèce est actuellement en expansion dans les boisements de Pinus,<br />

Eucalyptus et Populus, à proximité de formations naturelles ainsi<br />

qu’au niveau de bosquets isolés où les vergers d’amandiers et d’oliviers<br />

servent de corridors biologiques.<br />

Mots-clés<br />

Eucalyptus, Populus, Picoides major, Phoracantha semipunctata,<br />

Sicile<br />

1. Paper carried out within the research program “Programma Operativo Multiregionale – Misura 2 – Innovazioni tecnologiche e trasferimento dei risultati<br />

della ricerca – Progetto B28: Nuove metodologie per la gestione sostenibile dei sistemi forestali complessi nell’Italia Meridionale”. Authors took equal parts<br />

in its fulfillment.<br />

65


66<br />

◆ TOMMASO LA MANTIA, MICHELE SPOTO, BRUNO MASSA<br />

Riassunto<br />

Nell’ultima metà del secolo scorso, la superficie forestale in Sicilia<br />

ha avuto un discreto incremento determinato dall’opera di rimboschimento.<br />

Le specie utilizzate sono state soprattutto pini mediterranei<br />

ed eucalitti. Questi ultimi sono stati utilizzati per la presunta<br />

elevata produttività e della effettiva capacità di adattamento alle<br />

situazioni pedologiche più difficili. I nuovi boschi di eucalitto hanno<br />

mostrato tutti i loro limiti in termini ecologici, non consentendo<br />

un’elevata diversità floro-faunistica. In questi ultimi anni, grazie<br />

anche alla colonizzazione da parte di specie di insetti proprie dell’eucalitto,<br />

è in atto un processo di naturalizzazione, che precorre una<br />

Photos 1 and 1bis. During the winter some Eucalyptus camaldulensis show a series of regularly spaced holes<br />

from which the Great Spotted Woodpecker obtain the sugary sap.<br />

probabile diversificazione dell’habitat. La diffusione del Picchio<br />

rosso maggiore negli eucalitteti si inserisce in questo processo; esso è<br />

stato osservato anche in altri tipi di rimboschimenti, come i pioppeti,<br />

la cui finalità è del tutto differente. Negli eucalitteti, il Picchio<br />

si nutre delle larve della Phoracantha semipunctata e della linfa<br />

ottenuta effettuando numerosi fori nel tronco di alcune piante. La<br />

specie si è diffusa in boschi di pini ed eucalitti, contigui a formazioni<br />

naturali e seminaturali ma anche in boschi isolati dimostrando<br />

una capacità dispersiva inattesa. Le osservazioni hanno<br />

accertato il ruolo di corridoi ecologici svolto dagli arboreti come<br />

mandorleti ed oliveti nel consentire questa espansione.<br />

Photo 2. Holes made by woodpeckers<br />

on a telephone pole, probably with<br />

the aim to excavate a nesting hole.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


THE COLONISATION OF THE GREAT SPOTTED WOODPECKER (PICOIDES MAJOR L.) IN EUCALYPT WOODS… ◆<br />

INTRODUCTION<br />

One of the main causes of the massive use of Eucalyptus<br />

species in the past decades has been the total absence of<br />

specific pests interfering significantly with the production<br />

processes. Nevertheless, in the last years, as international<br />

woody exotic species trade became more and more<br />

intense, the increasing import of entire trees and vegetative<br />

propagules enhanced the dispersal opportunities for<br />

a high number of insects linked to Eucalyptus species in<br />

their original country. As matter of the fact, the australian<br />

beetle Phoracantha semipunctata Fabricius became<br />

almost cosmopolite. In Europe it was recorded firstly in<br />

Sardinia (Tassi 1969; Piras et al., 1970) where, according<br />

to Cavalcaselle and Contini (1973) about 100,000 trees<br />

were already attacked in 1971; in the following years it<br />

was also reported from Sicily (Romano & Carapezza,<br />

1975) and Central-South Italy (Parenzan, 1976; Longo<br />

et al., 1993; Rastelli et al., 2001).<br />

Additionally, due to hazelnut economical crisis (Clerici<br />

& Asciuto, 1991), many poplars took the place of hazel<br />

orchards, or, in some pedologically favourable territories,<br />

previously uncultivated, a productive poplar cultivation<br />

started. Actually the surface of poplar in this area is about<br />

300 ha.<br />

Generally, Eucalypt plantations are very poor in bird<br />

species, but in the last ten years in Sicily they were colonised<br />

by the Great Spotted Woodpecker (Picoides major<br />

L.). Aims of this research were to ascertain the actual diffusion<br />

of this species into Sicilian Eucalypt and poplar<br />

plantations and its behavioural relationships with these<br />

trees.<br />

MATERIALS AND METHODS<br />

Between 1995 and 2001 all Eucalyptus woods of Sicily<br />

were visited to record: a) the possible presence of the<br />

Australian xylophagous Phoracantha semipunctata by its<br />

typical emerging holes or the presence of adults or larvae<br />

under the bark; b) the possible presence of the Great<br />

Spotted Woodpecker (Picoides major L.) by its calls, its<br />

typical druming (mechanical sound made by resonance<br />

of dead branches under rapid blows from bill) or other<br />

evidences of its activity on tree bark. Abundance of some<br />

woodpecker populations was estimated by the IKA<br />

method (Ferry & Frochot, 1958), which gives linear values<br />

of abundance referred to a measure unit (number of individuals<br />

per km). Holes were inspected to ascertain if used<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 65-73<br />

by woodpeckers to nest, drum or feed. Some data on its<br />

feeding habits were obtained through observations in the<br />

field with the aid of a 10 x 40 binocular.<br />

In the years 1999-2000 in five eucalypt forests seven<br />

sample areas (0.5 ha) were established, where the diameter<br />

of all trees (n = 1119) was measured by a calliper,<br />

and evidences of the woodpecker presence (nests, druming<br />

and feeding holes) were recorded.<br />

The main features of the studied Eucalyptus plantations<br />

are reported in table 1. Mustigarufi is the largest<br />

one, with three species of Eucalyptus (E. camaldulensis,<br />

E. occidentalis and E. globulus) mixed at patch. The plots<br />

have been selected within two homogeneous areas but the<br />

observations were carried out in the entire wood. Reina,<br />

Torretta and Cannatello woods are homogeneous and<br />

similar according to species, surface and age on the<br />

contrary Spagnolo wood shows some differences in species<br />

utilisation and plant size.<br />

Additionally, in the years 1999-2000, in the poplar cultivation<br />

of Piazza Armerina (Enna) some observations<br />

were carried out to ascertain which clones are preferred<br />

by woodpeckers for feeding and nesting; three sample<br />

areas 40 x 40 m were established to gather some biometrical<br />

parameters and to evaluate the percentage of trees<br />

(n = 206) exploited by woodpeckers.<br />

RESULTS AND DISCUSSION<br />

Spreading of the Great Spotted Woodpecker<br />

in eucalypt wood: toward<br />

a new association woodpecker-eucalypt?<br />

Woodpeckers are a bird group clearly defined and<br />

homogeneous, highly specialized for the arboreal life,<br />

widespread all over the world, with the exception of<br />

Australia, New Zealand and Madagascar. The Great<br />

Spotted Woodpecker (Picoides major) constitutes a superspecies<br />

with P. leucopterus, P. assimilis, P. himalayensis and<br />

P. syriacus and is well diffused in the Palearctic Region, in<br />

woodland, plantations and some urban parks (Del Hoyo<br />

et al., 2002). In Sicily it was known as uncommon and<br />

local resident breeder (Iapichino & Massa, 1989), living<br />

in all the naturally wooded areas. Lo Valvo et al. (1993)<br />

recorded it on 19.9 % of UTM 10 x 10 km sicilian<br />

squares; more recently, presumably from nearby natural<br />

woods, it colonized some woods, including Eucalyptus<br />

woods, poplar afforestations and riparian woods, with an<br />

67


68<br />

◆ TOMMASO LA MANTIA, MICHELE SPOTO, BRUNO MASSA<br />

Wood, Municipality Area Plants/ha Occurrence (%) Mean Ø Plants with holes made by woodpeckers for drinking sap Plants with holes for nests<br />

(Province) (in ha) of different species (± s.d.) in cm<br />

in the sample area<br />

n/ha % Mean Ø % of Eucalyptus species nests/ha Mean Ø % of Eucalyptus species with<br />

(± s.d.) in cm with woodpecker holes woodpecker holes<br />

Mustigarufi, 3,252 289 100 E. camaldulensis 24.3 ± 5.5 11 3.8 19.3 ± 4.3 100 E. camaldulensis* 2 29.2 100 E.camaldulensis*<br />

San Cataldo (CL)<br />

Reina, 210 350 95 E. camaldulensis 16.4 ± 4.4 13 3.7 15.6 ± 3.5 100 E. camaldulensis 3 24.7 100 E.camaldulensis<br />

Mussomeli (CL) 5 E. occidentalis<br />

Torretta, 76 750 95 E. camaldulensis 15.8 ± 4.4 10 1.5 17.0 ± 2.4 100 E. camaldulensis 2 28.6 100 E.camaldulensis<br />

Mussomeli (CL) 5 E. occidentalis<br />

Cannatello, 137 500 95 E. camaldulensis 16.5 ± 4.7 8 1.6 16.0 ± 2.2 100 E. camaldulensis 2 22,0 75 E.camaldulensis<br />

Mussomeli (CL) 5 E. occidentalis 17.0 ± 4.2 25 E.occidentalis 32,0<br />

Spagnolo, 226 350 56 E. occidentalis 11.1 ± 4.88 10 2.9 10.3 ± 2.3 100 E. camaldulensis 2 11,0 75 E. camaldulensis<br />

Milena (CL) 44 E. camaldulensis 10.8 ± 4.2 12,0 25 E. occidentalis<br />

* Outside sample areas some specimens of E. occidentalis with presence signs of woodpeckers were observed.<br />

Table 1. Some parameters of Sicilian Eucalyptus woods where woodpecker activity was observed and characteristics of plants where woodpeckers carried out their activity.<br />

amount of variation of another 6.4%, totally covering<br />

26.3% of UTM squares (fig. 1 and 2).<br />

This was unexpected, particularly because Eucalyptus<br />

afforestations are generally considered very poor in fauna<br />

and flora (La Mantia & Pasta, 2001); additionally, as<br />

reported by some authors (e.g. Hinsley et al.,1995), this<br />

woodpecker is usually lacking in small woods, like most<br />

Sicilian afforestations.<br />

Censuses of woodpeckers carried out in Sicilian natural<br />

woods let us to obtain densities remarkably higher than in<br />

afforestations, particularly of Eucalyptus. In oakwoods and<br />

beechwoods we found as much as 3-5 individuals/10 km,<br />

while in Eucalyptus afforestations only 0.5-1 individuals/10<br />

km; the difference was significant (Wilcoxon test: Z 8=1.94,<br />

P=0.05). During the winter we found many small holes<br />

(mean width 1.5 cm, depth 0,5 cm) on scattered trees; the<br />

result of its activity is a very high number of rings of holes<br />

around the trunk trees. Following its behaviour, we assumed<br />

that the food source was the sap and that the woodpecker<br />

used to exploit many times the same tree for this<br />

reason.<br />

Many species of woodpeckers, Great Spotted Woodpecker<br />

included, cover the bark of living trees with series<br />

of regularly spaced holes from which they obtain the<br />

sugary sap (Del Hoyo et al., 2002). Typically, the holes,<br />

known as “sap wells”, are arranged in horizontal rows,<br />

which, together with those made in previous years, can<br />

extend over rather distinctly large parts of a tree.<br />

Tree sap may be the key resource for survival in arid<br />

habitats.<br />

This behaviour indicates that the sap was not randomly<br />

searched for, but the same tree was exploited for a long<br />

time. Previously, other authors highlighted the importance<br />

of sap as food of P. major (cf. Cramp, 1985 and<br />

references therein); the rings of holes are made by woodpeckers<br />

to drink sap oozing out or, possibly, also to eat<br />

exposed cambium of tree. Similar situation were observed<br />

an individual tree of Ulmus glabra, showing over 400<br />

holes for sap extraction, thus certainly used for many<br />

years (Gatter, 1972).<br />

As regards the size of trees concerned by woodpecker<br />

activity it is variable. Woodpeckers seem to prefer trees<br />

characterized by a small trunk diameter (table 1 and fig. 3).<br />

The mean diameter of trees where woodpeckers excavated<br />

nesting holes resulted a bit larger, and dependent again<br />

on the mean size of trees (table 1 and fig. 3), but the scarce<br />

number of trees did not allowed a statistic test.<br />

Xylophagous larvae of P. semipunctata are also an<br />

important prey in winter. In Eucalyptus afforestations<br />

insects are scarce and uncommon, and larvae of P. semi-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


THE COLONISATION OF THE GREAT SPOTTED WOODPECKER (PICOIDES MAJOR L.) IN EUCALYPT WOODS… ◆<br />

Fig. 2. Map of Sicilian wood in Sicily<br />

(after Lo Valvo et al., 1993).<br />

punctata may provide an important part of the woodpecker’s<br />

diet (as reported by Mendel et al. (1984) for P. syriacus<br />

in Israel).We observed woodpeckers capturing insects<br />

living under bark in dead trees and in dead wood in the<br />

ground during the winter, and insects living on the surface<br />

of trees and on foliage in spring and summer.There<br />

are also consistent data reporting seeds, bird eggs and<br />

chicks as food of this woodpecker (Cuisin & Pesson, 1980;<br />

Cramp, 1985). Also Amat and Soriguer (1983) and Zenatello<br />

(1995) report as food of P. major in spring both xylophagous<br />

insects and other species collected on the leaves<br />

(particularly on the lower surface). Different authors have<br />

stressed the importance of xylophagous larvae in the diet<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 65-73<br />

Fig. 1. Map of distribution of the Great Spotted Woodpecker<br />

(Picoides major L.) in Sicily.<br />

In black the 10 x 10 km UTM squares where<br />

the woodpecker was present in 1979-93<br />

(after Lo Valvo et al., 1993), in white newly colonised areas.<br />

of this woodpecker, mostly in winter. Woodpeckers are<br />

efficient predators of overwintering moth larvae, frequently<br />

reducing the pest population in orchards to a level<br />

where other natural control agents were able to prevent<br />

the succeeding generation from damaging the fruit to an<br />

uneconomical degree (Mac Lellan, 1959).<br />

Adult and larvae of bark beetles form 67-99 % of the<br />

winter diet of woodpeckers in infected areas, and their<br />

activity results in 45-98 % decrease in beetle survival<br />

(Crockett & Hanoley, 1978). Woodpeckers have been<br />

recognized for many years as a major factor in the natural<br />

reduction of Engelmann spruce beetle populations<br />

(Knight, 1958). P. major is considered an important pre-<br />

69


70<br />

◆ TOMMASO LA MANTIA, MICHELE SPOTO, BRUNO MASSA<br />

Fig. 3. Mean diameter of trees in different Eucalypt woods (c = Eucalyptus camaldulensis;o=Eucalyptus occidentalis)<br />

and mean diameter of trees used by woodpeckers for nesting and for drinking sap.<br />

dator of overwintering larvae of Scolytus intricatus, being<br />

responsible of capturing up to 62.6 % of them (Pavlik,<br />

1999). Related species are also considered important predators<br />

of xylophagous insects. Mendel et al. (1984) observed<br />

that hole density of P. syriacus on Eucalyptus increase<br />

during the spring and summer, and that the average larval<br />

mortality of P. semipunctata due to predation of this<br />

woodpecker may reach as much as 29 %. They consider<br />

P. syriacus an efficient natural enemy of P. semipunctata,<br />

destroying larvae and consuming the eggs on standing<br />

trees.<br />

We dissected some trees bearing a high number of<br />

rings of holes (testifying the woodpecker activity for drinking<br />

sap), without finding any evidence of the presence<br />

of xylophagous insects. Conversely, only removing the<br />

bark of trees with wide and irregular woodpecker holes,<br />

we found the typical galleries of P. semipunctata.Thus, we<br />

may assume that woodpeckers excavate on the plant regu-<br />

lar and horizontal small holes to drink sap. As a matter of<br />

fact, only from parts of trees bearing wide and irregular<br />

woodpecker holes, preserved in laboratory, adults of<br />

P. semipunctata emerged.<br />

As regards eucalypt species, when woodpecker is searching<br />

for sap, it prefers Eucalyptus camaldulensis and E.globulus,<br />

probably because these species have a gum type<br />

deciduous bark (Moggi, 1964); species, with persistent<br />

bark, like E. occidentalis which present Stringybark type<br />

(Moggi, 1964), are not interested by woodpecker drinking<br />

activity. Conversely, the woodpecker activity for extracting<br />

larvae of P. semipunctata is independent from the<br />

eucalypt species and the bark type.<br />

During our research, we found rings of holes made by<br />

woodpeckers also on some Aleppo Pine (Pinus halepensis),<br />

Cypress (Cupressus sempervirens) and Elm (Ulmus<br />

minor) and, exceptionally, also on a telephone pole, probably<br />

with the aim to excavate the nesting hole. Several<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


THE COLONISATION OF THE GREAT SPOTTED WOODPECKER (PICOIDES MAJOR L.) IN EUCALYPT WOODS… ◆<br />

Clones plants/ha Mean diameter Height Plants bearing Nesting or resting holes<br />

% n (cm) (m) woodpecker Plants with holes Mean N. Height range of<br />

feeding holes (%) of holes per plant holes (m)<br />

I 614 12 50 41 30.0 25* 63** 3.5 5-7<br />

Armerino 88 381 24 25.7 – – – –<br />

* feeding holes and nesting or resting holes were found in different plants<br />

** % is referred to the total number of plants of the clone<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 65-73<br />

Table 2. Characteristics of poplar cultivations and woodpecker activity at Piazza Armerina (Enna).<br />

authors have reported damages caused by woodpeckers.<br />

Evers (1982) noticed in Germany and Norway some cases<br />

of Norway Spruce power line poles (preserved with carbolineum),<br />

exploited as sites for the excavation of nesting<br />

holes by P. major, while Moran (1977) reported cases of<br />

attack to polyethylene irrigation pipes in Israel orchards<br />

by P. syriacus. According to Allegro (1993), the benefits<br />

of P. major (which feeds on a wide range of insect pests)<br />

outweigh any damage to the tree bark and stems which<br />

may be caused. As regards eucalypt woods, probably trees<br />

injured by woodpecker activity are more easily submitted<br />

to pathogen and P. semipunctata attacks, but certainly the<br />

woodpecker results to be an important natural enemy of<br />

the xylophagous insect. Indeed, we found many trees killed<br />

by the attacks of P. semipunctata, while we did not find<br />

dead trees, bearing typical rings of holes caused by woodpecker<br />

searching for sap.<br />

In the wood of Mustigarufi we observed the highest<br />

activity of the woodpecker against larvae of P. semipunctata<br />

in plots where trees, according to Fierotti et al. (1995),<br />

grow under very unfavourable conditions. Evidence of its<br />

activity was the presence of irregular holes spread on the<br />

bark. On the contrary, in plots where trees grow under<br />

quite favourable conditions, we mostly observed the typical<br />

woodpecker activity for extracting sap, testified by the<br />

presence of small regular rings of holes.<br />

Concerning sampled eucalypt woods, they (with the<br />

exception of Mustigarufi) are small sized and well isolated<br />

from other woods. It seems that P. major has colonised<br />

these woods through areas with scattered trees, like<br />

almond orchards, that it exploits in autumn-winter for<br />

feeding reasons (A. Cairone, pers. comm.) and suggests<br />

the important role of small woods and biocorridors. At<br />

local scale, some studies (La Mantia 1997; Massa & La<br />

Mantia, 1997) have already highlighted the key-role played<br />

in Sicily by biocorridors and hedgerows-windbreaks in<br />

conserving and increasing species richness.<br />

Colonisation of the poplar cultivation<br />

of Piazza Armerina by the Great Spotted<br />

Woodpecker<br />

Recently, the Great Spotted Woodpecker has colonised<br />

the poplar cultivations of Piazza Armerina, in central<br />

Sicily (fig. 2).This area has been cultivated with a mosaic<br />

distribution of Eucalyptus, poplar, Pinus pinea, and riparial<br />

vegetation which increased the diversity. Nowadays,<br />

at Piazza Armerina the following poplar clones are cultivated<br />

(Garfì & La Mantia, unpublished data): 1)<br />

‘Armerino’, now naturalised, which is a hybrid of uncertain<br />

origin between Populus nigra and euroamerican<br />

hybrids introduced to Piazza Armerina. Despite its lower<br />

productivity in comparison with other clones, it shows a<br />

higher fitness to the environment and very good technological<br />

characteristics to produce matches that are the<br />

main local product; 2) ‘I 214’, an euroamerican clone very<br />

diffused in the modern poplar cultivation for its high productivity,<br />

at Piazza Armerina it has been decreasing<br />

because its wood is not suitable for matches production;<br />

3) ‘Luisa Avanzo’, also introduced for its high productivity,<br />

but, for the same reasons of ‘I 214’, has become rare.<br />

As a consequence of the availability of newly afforested<br />

territories (both of eucalypts and poplars), the Great<br />

Spotted Woodpecker has recently colonised the above<br />

cited poplar plantations, showing evident preferences for<br />

some clones. In particular, in poplar stands with mixed<br />

clones, it does not excavate holes in trees of ‘Armerino’,<br />

while it does in ‘Luisa Avanzo’ and ‘I 214’ clones, both<br />

for feeding and nesting. Our observations (table 2) let us<br />

to conclude that the presence of few trees of ‘I 214’ or<br />

‘Luisa Avanzo’ (preferred by woodpeckers) possibly<br />

consents to protect them from xylophagous attacks,<br />

without be detrimental to technological characteristics of<br />

‘Armerino’ trees. Some farmers report negative effects of<br />

woodpeckers on poplars, but scattered ‘I 214’ and ‘Luisa<br />

71


72<br />

◆ TOMMASO LA MANTIA, MICHELE SPOTO, BRUNO MASSA<br />

Avanzo’ clones may play the same role of dead trees and<br />

natural woods bordering poplar cultivations of North Italy<br />

(Allegro, 1991). Concerning the positive role of this woodpecker<br />

within poplar stands, Camerini & Quadrelli (1991)<br />

report the interesting observation of its active predation<br />

on Ostrinia nubilalis living within stems of corn that area<br />

cultivated at the edge of a young poplar, where it regularly<br />

preyed the xylophagous Cossus cossus, Saperda carcharias<br />

and Paranthrene tabaniformis (cf. also Allegro 1991,<br />

1993). In winter, in riparian forests of poplar this woodpecker<br />

may destroy about 10-50% of larvae of Saperda<br />

carcharias by pecking them out of their galleries (Srot,<br />

1983; Allegro, 1991 and 1996). According to Quadrelli<br />

(1984), in poplars of North Italy it removes about 20%<br />

of larvae of Cossus cossus, showing a higher foraging activity<br />

in trees with diameter between 40-50 and 80-90 cm.<br />

CONCLUSIONS<br />

The recent diffusion of the Great Spotted Woodpecker<br />

suggests the important role of small woods and biocorridors.Their<br />

function has been recognised at an international<br />

level and led to the realisation of the “European Ecological<br />

Network” (E<strong>ECO</strong>NET) (Benett, 1991, 1994).<br />

The results of this work, focusing on the role of great<br />

spotted woodpecker and Phoracantha semipunctata, show<br />

an increase of ecological interactions within these artificial<br />

woods. For this reasons several researches are now<br />

involved in the identification of the best approach to<br />

increase their naturalisation process (La Mantia et al.,<br />

2000). In particular, the adoption of more adequate selvicultural<br />

techniques should enhance the diffusion of<br />

many autochthonous trees and berry-producing shrubs.<br />

The so-called “re-naturalisation actions” (La Mantia &<br />

Pasta, 2001) could be applied in different ways, depending<br />

on the wood-type; besides, they should allow an<br />

increase and a maintainance of species-richness (Ciancio,<br />

2000).Thus, it is necessary to fit wood implantation and<br />

management techniques to the new role that wood itself<br />

plays in the western countries.<br />

AKNOWLEDGEMENTS<br />

We thank the staff of the Forestry Office of Sutera and<br />

Mustigarufi, and particularly Mr. S. Butera, Mr. G. Farina<br />

and Mr. C. Platania (ISFA Catania), who allowed and<br />

facilitated our investigations in the poplar implantation of<br />

Piazza Armerina; Mr. P. Colina for his kind help,<br />

A. Cairone, V. Di Dio, R. Ientile, M. Lo Valvo and<br />

O. Vitrano for providing their personal data on Great<br />

Spotted Woodpecker. Thanks to I. Sparacio for bibliographic<br />

researches, G. Garfì and M. Cassarà for field<br />

investigations at Piazza Armerina.We are grateful to Miss<br />

N. Callea, for maps’ editing. This research was partially<br />

funded by the Assessorato Agricoltura e Foreste Regione<br />

Sicilia to Stazione di Inanellamento.<br />

The authors are thankful to P. Villard for the critical<br />

and careful reading of this paper.<br />

References<br />

ALLEGRO G., 1991. Il picchio rosso maggiore (Picoides major L.)<br />

nella limitazione naturale delle popolazioni della Saperda maggiore<br />

del pioppo (Saperda carcharias L.). Avocetta 15: 33-41.<br />

ALLEGRO G., 1993. Un prezioso alleato della pioppicoltura: il<br />

Picchio rosso maggiore. Informatore Agrario 49: 73-76.<br />

ALLEGRO G., 1996. Osservazioni sul comportamento del picchio<br />

rosso maggiore, Picoides major L., in pioppeto durante<br />

il periodo riproduttivo. Riv. Ital. Orn. 66: 17-27.<br />

AMAT J.A. & SORIGUER R.C., 1983. Utilizacion del habitat por<br />

el Pico picapinos (Dendrocopos major) en la Sierra de Cazorla.<br />

Alytes 1: 347-361.<br />

BARBERA G., LA MANTIA T., LA MELA VECA D.S., MARCHETTI<br />

M., SCALZO G., 2001. Productivity of Eucalyptus spp. in different<br />

environmental conditions and silvicultural systems in Sicily<br />

- an updating description. International Conference “Eucalyptus<br />

in the mediterranean basin: perspectives and new utilization”<br />

October 15-19, 2000 Taormina, Italy. Ed. Centro Promozione<br />

Pubblicità, Firenze, Italy, p. 291-299.<br />

BENETT G. (ED.), 1991.Towards a Europaen Ecological Network.<br />

Institute for European Environmental Policy, Arnhem, 80 p.<br />

BENETT G. (ED.), 1994. Conserving Europe’s natural heritage.<br />

Towards a European ecological Network. Proceedings of the<br />

international conference held in Maastricht, 9-12 November<br />

1993. Graham and trotman/Martinus Nijhof, London,<br />

Dordrecht Boston, 334 p.<br />

CAMERINI G. & QUADRELLI G., 1991. Ruolo del Picchio rosso<br />

maggiore, Picoides major, nella predazione della Piralide del<br />

Mais, Ostrinia nubilalis. Riv. Ital. Orn. 61: 43-47.<br />

CAVALCASELLE B. & CONTINI C., 1973. Osservazioni preliminari<br />

sugli insetti xilofagi dell’eucalitto in Italia. Boll. Soc. Sard. Sci.<br />

Nat., a.VII vol. 12 : 125-136.<br />

CIANCIO O., 2000. Gestione forestale e sviluppo sostenibile. – Atti<br />

II Congr. Naz.“Selvicoltura per il miglioramento e la conservazione<br />

dei bochi italiani” (Venezia dal 24-27 giugno 1998),<br />

vol. 3: 131-187.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


THE COLONISATION OF THE GREAT SPOTTED WOODPECKER (PICOIDES MAJOR L.) IN EUCALYPT WOODS… ◆<br />

CLERICI E. & ASCIUTO G., 1991. La pioppicoltura di Piazza<br />

Armerina: aspetti auxometrici ed economici. Cellulosa e Carta<br />

42 (2): 31-42.<br />

CRAMP S., 1985. The Birds of the Western Palearctic.Vol. IV. Oxford<br />

Univ. Press.<br />

CROCKETT A.B. & HANOLEY P.L., 1978. Apparent response of<br />

Picoides Woodpeckers to outbreaks of the pine bark beetle.<br />

Western Birds 9: 67-70.<br />

CUISIN M. & PESSON P. (ED.), 1980. L’influence des pics sur les<br />

milieux forestiers. In : Actualités d’écologie forestière. Sol, flore,<br />

faune, Gauthier-Villars, Paris, p. 407-414.<br />

DEL HOYO, J.,ELLIOTT, A.& SARGATAL, J., 2002. Handbook of<br />

the Birds of the World. Vol. 7. Lynx ed., Barcelona.<br />

EVERS,W., 1982. Eigenartiges Verhalten von Spechten. Forst und<br />

Holzwirt 37: 497-499.<br />

FERRY C. & FROCHOT B., 1958. Une méthode pour dénombrer<br />

les oiseaux nicheurs. Terre et Vie 12: 85-102.<br />

FIEROTTI G., DAZZI C., RAIMONDI S., BELLANCA A., MONTE-<br />

LEONE S., NER,L.& PIPITONE G., 1995. I suoli del complesso<br />

boscato “Mustigarufi” (CL) e la loro adattabilità all’Eucalyptus<br />

camaldulensis. Azienda Foreste Demaniali della Regione<br />

Siciliana.<br />

GATTER W., 1972. Ringing by woodpeckers. J. Ornithol. 113:<br />

207-213.<br />

HINSLEY S.A., BELLAMY P.E. & NEWTON I., 1995. Bird species<br />

turnover and stochastic extinction in woodland fragments.<br />

Ecography 18: 41-50.<br />

IAPICHINO C. & MASSA B., 1989. The Birds of Sicily. British<br />

Ornithologists’Union. Check-list n.11, London.<br />

KNIGHT F.R., 1958.The effects of woodpeckers on populations<br />

of the Engelmann spruce beetle. J. Econ. Entomol. 51: 603-<br />

607.<br />

LA MANTIA T, CULLOTTA S., GARFÌ G., LA MELA VECA S.D.,<br />

MARCHETTI M., SCALZO G., BARBERA G., 2000. Rinatu-ralizzazione<br />

delle piantagioni di eucalitti e di pini mediterranei.<br />

Relation presented at the meeting on “Nuove Metodologie<br />

per la Gestione dei Sistemi Forestali Complessi nell’Italia<br />

Meridionale” 28 Maggio 2001, UCEA – Roma.<br />

LA MANTIA T., 1997. Il ruolo degli elementi diversificatori negli<br />

agroecosistemi mediterranei: valorizzazione e relazioni con<br />

le popolazioni di vertebrati. Naturalista sicil., 12 (suppl.): 175-<br />

211.<br />

LA MANTIA T. & PASTA S. 2001. La rinaturalizzazione negli<br />

impianti forestali artificiali: proposte metodologiche e ipotesi<br />

di intervento nella Riserva Naturale “Grotta di S. Ninfa”.<br />

Naturalista sicil. 25 (suppl.): 299-323.<br />

LO VALVO M., MASSA B. & SARÀ M. (RED.), 1993. Uccelli e paesaggio<br />

in Sicilia alle soglie del terzo millennio. Naturalista sicil.<br />

17 (suppl.): 1-373.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 65-73<br />

LONGO S., PALMERI V. & SOMMARIVA D. 1993. Sull’attività di<br />

Avetianella longoi Siscaro, ooparassitoide di Phoracantha semipunctata<br />

Fabr. nell’Italia meridionale. Redia 76: 223-239.<br />

MAC LELLAN C.R., 1959.Woodpeckers as predators of the codling<br />

moth in Nova Scotia. Can. Entomol. 91: 673-680.<br />

MASSA B., LA MANTIA T., 1997. Benefits of hedgerows-windbreaks<br />

for birds and their valorisation in sustainable agriculture.<br />

Agricoltura <strong>Mediterranea</strong>, 127: 332-341.<br />

MENDEL Z., GOLAN Y. & MADAR Z., 1984. Natural control of<br />

the Eucalyptus borer, Phoracantha semipunctata (F.)<br />

(Coleoptera: Cerambycidae), by the Syrian woodpecker. Bull.<br />

ent. Res. 74: 121-127.<br />

MOGGI G., 1964. Guida al riconoscimento degli Eucalitti coltivati<br />

in Italia. Pubblicazioni del Centro di Sperimentazioni agricole<br />

e forestali 7: 147-220.<br />

MORAN S., 1977. Distribution and characteristics of the damage<br />

of the Syrian Woodpecker, Dendrocopos syriacus (Hemp. and<br />

Ehr.) (Aves: Picidae), in polyethylene irrigation pipes in fruit<br />

orchards. Phytoparasitica 5: 127-139.<br />

PARENZAN P., 1976. La Phoracantha semipunctata Fabr.<br />

(Coleoptera: Cerambycidae) nell’Italia meridionale.<br />

Entomologica, 12: 9-20.<br />

PAVLIK S., 1999. Predation of overwintering larvae of the<br />

European oak bark beetle (Scolytus intricatus Ratz.) by woodpeckers<br />

at logging residues in oak forests. Acta Facultatis<br />

Forestalis Zvolen Slovakia 41: 305-313 (in Slovakian, English<br />

summary).<br />

PIRAS L., CONTINI C. & PISANO P., 1970. Sulla introduzione in<br />

Europa di un Cerambicide australiano parassita di Eucalyptus:<br />

Phoracantha semipunctata Fabr. (Col. Cerambycidae). Boll.<br />

Soc. ent. ital. 102: 58-63.<br />

QUADRELLI G., 1984. Presenza ed alimentazione del Picchio<br />

rosso maggiore Dendrocopus major nelle coltivazioni di pioppo.<br />

Avocetta 8: 83-89.<br />

RASTELLI S., ABBRUZZESE E. & RASTELLI M., 2001. Cerambycidae<br />

d’Italia. Atlante fotografico dei Coleotteri Cerambicidi<br />

d’Italia. Museo Civico di Storia Naturale di Carmagnola e<br />

Centro Studi Ecologici Appenninici. CD-ROM.<br />

ROMANO F.P. & C ARAPEZZA A. 1975. Sulla presenza di<br />

Phoracantha semipunctata Fabr. in Sicilia (Coleoptera<br />

Cerambycidae). Boll. Soc. ent. ital. 107: 91-92.<br />

SROT M., 1983. Factors reducing the population density of the<br />

large poplar borer (Saperda carcharias L.). Lesnictvi 29: 785-<br />

804 (in Czech, English summary).<br />

TASSI F., 1969. Un longicorne australiano per la prima volta in<br />

Italia (Coleoptera Cerambycidae). Boll. Ass. romana Entom.<br />

24: 69-71.<br />

ZENATELLO M., 1995. Note sull’alimentazione del Picchio rosso<br />

maggiore, Picoides major, nella Foresta del Cansiglio (Prealpi<br />

Venete). Riv. Ital. Orn. 65: 29-36.<br />

73


Interactions between sympatric snakes, Coluber viridiflavus<br />

and Elaphe longissima: are there significant inter-annual<br />

differences in coexistence patterns?<br />

Interactions entre deux serpents sympatriques,<br />

Coluber viridiflavus et Elaphe longissima :existe-t-il<br />

des variations dans les modalités de coexistence inter-annuelles ?<br />

Lorenzo Rugiero 1, Dario Capizzi 2, Luca Luiselli 3<br />

1. FIZV, via Domenico Cimarosa 13, I-00198 Rome, Italy; email: lrugiero@hotmail.com<br />

2. INFS, via Ca’Fornacetta 14, Ozzano Emilia (Bologna), Italy; email: dario.capizzi@tin.it<br />

3. FIZV, via Olona 7, I-00198 Rome, Italy, and Institute of Environmental Studies “Demetra”,<br />

Largo Rina Morelli 15/III, I-00139 Rome, Italy; email: lucamlu@tin.it<br />

Abstract<br />

The dynamics of coexistence between two large-sized colubrid snakes<br />

(Elaphe longissima and Coluber viridiflavus) were studied in a<br />

mixed-oak forest in central Italy during eleven consecutive years<br />

(1986-1996). 511 adult C. viridiflavus (270 males, 241 females)<br />

and 294 adult E. longissima (153 males, 141 females) were captured<br />

and marked. Adult sex-ratio did not depart significantly from<br />

equality in either species. Males of both species were significantly<br />

larger (snout-vent-length) than females. The abundance ratio between<br />

the two species was always skewed towards C. viridiflavus<br />

(mean ratio: 1.65: 1, Standard Deviation = 0.32), although there<br />

were some annual fluctuations and inter-habitat differences in<br />

abundance patterns. Mean adult densities were 2.32 ± 1.05 specimens<br />

for C. viridiflavus and 1.34 ± 0.89 specimens x ha -1 for<br />

E. longissima. C. viridiflavus was abundant in sunny habitats<br />

and tended to avoid habitats with closed canopy, which were clearly<br />

preferred by E. longissima. Both species were, however, generalists<br />

in terms of macro-habitat preferences. There were no major interannual<br />

differences in the patterns observed. At all ages, the dietary<br />

spectrum of C. viridiflavus was wider than that of E. longissima<br />

and included rodents, lizards and birds (common prey for both species)<br />

as well as frogs and invertebrates, which were not taken at all<br />

by E. longissima. In addition, C. viridiflavus more often preyed<br />

upon Clethrionomys glareolus than on Apodemus flavicollis,<br />

whereas the reverse was true for E. longissima. Male C. viridiflavus<br />

consumed a significantly higher number of lacertids than rodents<br />

but females consumed an almost identical number of each; no such<br />

inter-sexual differences occurred in E. longissima. The diet composition<br />

of the juveniles of the two species essentially consisted of<br />

lacertid lizards, but C. viridiflavus also ate invertebrates. ‘Preysize-predator-size’<br />

relationships were similar in the two species. Body<br />

temperature (T b) was higher in C. viridiflavus (x = 25.4 ± 3.1 °C)<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

than in E. longissima (x = 23.4 ± 1.8 °C), but in both species T b<br />

was positively correlated with air temperature (T a), and the regressions<br />

T a-T b were similar and both differed significantly from ideal<br />

poikilothermy. Body temperature did not correlate with snout-vent<br />

length in either species. The relative stability shown across years in<br />

this study system is likely to be due to the mature and stable conditions<br />

of the environment, the low level of habitat fragmentation<br />

and the high quality of habitat along the buffer zones. Species-specific<br />

eco-physiological requirements appear to explain the coexistence<br />

patterns of the two species better than any community-based<br />

synecological factors.<br />

Key-words<br />

Colubridae, Coluber viridiflavus, Elaphe longissima, Interspecific<br />

relationships, Ecology, Population biology, Italy<br />

Résumé<br />

La dynamique de coexistence entre deux espèces de serpents colubridés<br />

de grande taille (Elaphe longissima et Coluber viridiflavus)<br />

a fait l’objet d’un suivi, onze années durant (1986-1996), dans une<br />

forêt mixte de chênes en Italie centrale. 511 individus adultes de C.<br />

viridiflavus (270 mâles, 241 femelles) et 294 individus adultes de<br />

E. longissima (153 males et 141 femelles) ont ainsi été capturés, et<br />

marqués. Le sex-ratio des adultes ne diffère pas significativement de<br />

l’égalité pour les deux espèces. Les mâles des deux espèces se sont avérés<br />

être significativement plus gros que les femelles.<br />

Le rapport d’abondance entre les deux espèces a toujours été en<br />

faveur de C. viridiflavus (taux moyen : 1.65 :1, écart-type = 0.32)<br />

malgré quelques fluctuations annuelles et des différences inter-habitat<br />

dans les patrons d’abondance. Les densités moyennes d’individus<br />

adultes furent 2.32 ± 1.05 spécimens pour C. viridiflavus et<br />

1.34 ± 0.89 spécimens x ha-1 pour E. longissima. C. viridiflavus<br />

75


76<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

est une espèce particulièrement abondante dans les habitats ensoleillés<br />

qui tend à éviter les milieux trop fermés, lesquels sont clairement<br />

préférés par E. longissima. Les deux espèces se sont cependant<br />

avérées être relativement généralistes en termes de macro-habitats.<br />

Quel que soit l’âge, le spectre alimentaire de C. viridiflavus<br />

est apparu plus large que celui de E. longissima et inclut des rongeurs,<br />

des lézards et des oiseaux (proies communes pour les deux<br />

espèces) ainsi que des amphibiens et des invertébrés (chez E. longissima<br />

seulement). En outre, C. viridiflavus prédate plus fréquemment<br />

Cle-thrionomys glareolus que Apodemus flavicollis<br />

alors que c’est l’inverse pour E. longissima. Les mâles de C. viridiflavus<br />

consomment un nombre significativement plus élevé de<br />

lézards que de rongeurs alors que les femelles en consomment un<br />

nombre identique. Une telle différence entre les sexes n’a pas été<br />

mise en évidence chez E. longissima. Le régime alimentaire des<br />

juvéniles des deux espèces est essentiellement composé de lézards<br />

Lacertidés, mais C. viridiflavus consomme également des invertébrés.<br />

La relation “taille des proies-taille des prédateurs” est simi-<br />

laire chez les deux espèces. La température corporelle (T b) est plus<br />

élevée chez C. viridiflavus (x = 25.4 ± 3.1 °C) que chez E. longissima<br />

(x = 23.4 ± 1.8 °C), mais chez les deux espèces, T b est<br />

positivement corrélée avec la température de l’air (T a), et les régressions<br />

T a-T b présentent des valeurs identiques et différent de la poïkilothermie<br />

idéale. La température du corps n’est corrélée avec la<br />

longueur du corps chez aucune des deux espèces. La relative stabilité<br />

au cours des années est probablement due aux conditions stables<br />

et matures de l’environnement, le faible degré de fragmentation et<br />

la qualité de l’habitat le long des zones-tampons. Les besoins écophysiologiques<br />

spécifiques de chaque espèce semblent mieux expliquer<br />

leurs patrons de coexistence que l’ensemble des paramètres synécologiques<br />

des communautés.<br />

Mots-clés<br />

Colubridae, Coluber viridiflavus, Elaphe longissima, relations<br />

inter-spécifiques, écologie, biologie des populations, Italie<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


INTRODUCTION<br />

INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

The study of community organization and species’<br />

interactions is one of the prominent issues of modern ecology.<br />

It has for a long time been dominated by an “epistemiological<br />

misdirection” towards equilibrium hypotheses<br />

(sensu Barbault, 1991), where snapshots in space and time<br />

of given assemblages have been considered as reliable indicators<br />

of that community’s organization, i.e. with an underestimation<br />

of the role of physical factors and of spatial<br />

and temporal variability in the dynamics of natural communities<br />

(e.g., see Wiens, 1977; Schoener, 1986; Barbault,<br />

1991). Considering that spatial and temporal variability<br />

no doubt exists in most (all?)natural systems (e.g., see<br />

Blondel, 1990), it is evident that the data-sets best suited<br />

to interpret ecological patterns and processes should come<br />

from long-term studies (Magnuson, 1990) and / or from<br />

comparative studies of the same ecological system (i.e.<br />

the interaction between the same species) under different<br />

environmental conditions (e.g., different habitat types; see<br />

Harvey & Pagel, 1991). Both these approaches, however,<br />

are time consuming and logistically difficult and have only<br />

infrequently been used to study species’ interaction dynamics<br />

in the field (Schoener, 1983); this shortfall is especially<br />

true of snake studies (but see Luiselli et al., 2002).<br />

In the present paper, we use a long-term approach to<br />

study the interaction dynamics of two large-sized colubrid<br />

snakes from the <strong>Mediterranea</strong>n region of central Italy<br />

(Coluber viridiflavus and Elaphe longissima). We investigated<br />

whether the mechanisms of coexistence of these<br />

two species show any notable variation across years and<br />

whether coexistence dynamics are maintained through<br />

competitive relationships or via species-specific eco-physiological<br />

constraints (see Barbault, 1991; Barbault &<br />

Stearns, 1991). We address these issues using data collected<br />

over 11 consecutive years in a mature Oak forest<br />

in central Italy.This long-term approach permits a much<br />

deeper analysis of the ecological relationships between<br />

these two species than has previously been possible. It is<br />

noteworthy that both these species have been subject to<br />

detailed auto-ecological studies (e.g., see Naulleau, 1984,<br />

1987, 1989, 1992a, 1992b; Ciofi & Chelazzi, 1991, 1994;<br />

Luiselli & Rugiero, 1993; Bonnet & Naulleau, 1996a,<br />

1996b; Capula & Luiselli, 1995, 2002; Capula et al.,1995,<br />

1997, 2000; Filippi, 1995; Naulleau & Bonnet, 1995;<br />

Rugiero & Luiselli, 1995; Capizzi et al., 1996; Pillet, 1997;<br />

Bonnet et al., 1999; Kammel, 1999; Najbar, 1999a, 1999b,<br />

2000a, 2000b; Springolo & Scali, 1999; Agrimi, 2000;<br />

Cattaneo & Capula, 2000; Hofer, 2001a, 2001b; Zuffi,<br />

2001; Gomille, 2002), but their synecological relation-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

ships have been studied in less detail (but see Scali & Zuffi,<br />

1994; Capizzi & Luiselli, 1996a; Luiselli & Capizzi, 1997;<br />

Bretscher, 1999; Filippi & Luiselli, 2001, 2002; Hofer,<br />

2001a, 2001b).<br />

Here, we attempt to address the following main questions<br />

concerning sympatric populations of C. viridiflavus<br />

and E. longissima:<br />

1. Do the populations of these snakes exhibit annual<br />

fluctuations in <strong>Mediterranea</strong>n habitats? If so, to what<br />

extent, i.e. within what range do these fluctuations<br />

occur?<br />

2. Do these species exhibit any resource partitioning<br />

in terms of a) habitat utilization and b) dietary habits,<br />

and do any patterns occur consistently among years<br />

over a long-term period?<br />

3. Given that differences in thermal ecology may<br />

constitute niche differences among sympatric populations<br />

of different species (Avery, 1979), do these snakes<br />

differ each from the other in terms of thermal ecology<br />

attributes?<br />

4. Can the eventual interspecific differences observed<br />

be interpreted as a result of competitive relationships<br />

or via species-specific eco-physiological constraints?<br />

These questions are crucial to understanding the proper<br />

structure of species’ interaction because they link<br />

population dynamics with community structure (Strong<br />

et al., 1984).<br />

MATERIALS AND METHODS<br />

Study area<br />

Data presented in this paper were gathered from the<br />

“Bosco Macchia Grande di Manziana” (province of<br />

Rome, approx. 300-350 m a.s.l.), which is situated<br />

approximately 5 km west-northwest of Bracciano, and 2<br />

km south-west of Manziana.The study area, which has a<br />

surface of approximately 530 ha, is the last remnant piece<br />

of the ancient Sylva Mantiana, an enormous mixed forest<br />

(with oaks, lime-trees and maples) which previously covered<br />

the whole territory of the Sabatini, Ceriti, and Tolfa<br />

Mountains.The study area is situated inside the sub-montane<br />

deciduous oak forest climatic region, and is dominated<br />

by Quercus cerris, Q. frainetto, Acer campestre and<br />

Tilia sp., and has populations of Carpinus betulus and C.<br />

orientalis in cooler parts. Populations of Q. ilex are present<br />

in the drier and sunny sites, and populations of<br />

77


78<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

Corylus avellana, Fraxinus ornus, and Ulmus minor, are<br />

present in a few spots with optimal humidity conditions.<br />

The borders of the woodlot are colonized by <strong>Mediterranea</strong>n<br />

macchia vegetation with Cytisus scoparius, whereas<br />

the typical undergrowth consists of species of central-european<br />

origin (Sorbus domestica, S. aucuparia,<br />

Mespilus germanica,Crataegus monogyna,Rubus caesius,R.<br />

ulmifolius, Prunella vulgaris, Prunus spinosa, Ruscus aculeatus,Ilex<br />

aquifolium,Rosa canina,R.arvensis,Hedera helix,<br />

Daphne laureola, Pteridium aquilinum, etc.). Grassy areas<br />

are dominated by Dactylis glomerata, Luzula forsteri, Poa<br />

trivialis, Lolium perenne,Trifolium repens, Plantago lanceolata,<br />

and Galium vernum. Further details of the vegetation<br />

of the area are published elsewhere (e.g., Montelucci,<br />

1977; Spada, 1977).<br />

The fauna is rich and diverse. Among reptiles, apart<br />

from the two species studied here, the following species<br />

are found: Vipera aspis,Natrix natrix,N.tessellata,Coronella<br />

austriaca, Elaphe quatuorlineata, Lacerta bilineata, Podarcis<br />

sicula, P. muralis, Chalcides chalcides,Anguis fragilis,Testudo<br />

hermanni (Bruno, 1977; Luiselli et al., unpublished observations;<br />

Filippi, 1995).<br />

General methods<br />

Most of the data are original, apart from some of the<br />

dietary data, which the authors have published elsewhere<br />

(Luiselli & Rugiero, 1993; Capula & Luiselli, 2002).<br />

Snakes were searched along standardized routes<br />

conducted throughout the study area. Each snake was<br />

measured for snout-vent length (SVL, to the nearest ±<br />

0.1, cm) and body mass (to ± 0.1 g precision by an electronic<br />

balance), and individually marked by ventral scale<br />

clipping. Snakes were palpated in the abdomen until<br />

regurgitation of ingested food or defecation occurred, and<br />

prey items were identified to the lowest taxon possible.<br />

Female pregnancy status and numbers of eggs were determined<br />

by abdominal palpation. Subadults (which were<br />

discriminated from the adults by their different dorsal<br />

livery; see Bruno & Maugeri, 1977) were included in our<br />

dietary analyses, but were excluded from our analyses of<br />

population abundance.This was because their frequency<br />

appeared to be very different between sites and years and<br />

it was impossible to establish whether this occurred<br />

because of true site- or year-specific differences in abundance,<br />

or because of observer bias resulting from differences<br />

in vegetation and habitat structure among sites.<br />

Each year, from 1986 to 1996, a total of 21 to 32 man/<br />

days was spent in the field surveying the study area. We<br />

attempted to standardise effort as much as possible year-<br />

by-year and between macro-habitat types. Four well-separates<br />

snake macro-habitats were surveyed within the study<br />

area:<br />

1. Forest clearings with grassy-sandy soil (CLR).The<br />

typical phytosociological communities are included in<br />

the Thero-Brachypodetalia Br.-Bl. (1931) 1936 and the<br />

Brometalia Br.-Bl. (1936);<br />

2. Dense oak forest (OAK) (mainly Quercus cerris,<br />

Q. pubescens, Ostrya carpinifolia and Carpinus orientalis<br />

(canopy approx. > 70 %);<br />

3. Sparse oak forest (ROV) (mainly Quercus cerris,<br />

Q. frainetto and also Acer monspessulanum, Cercis siliquastrum<br />

and Paliurus spina-christi) with bushes (Rubus<br />

ulmifolius and Cytisus scoparius) (canopy approx.<br />


of the number of captured specimens of one species and<br />

of the number of captured specimens of the other species.<br />

Each snake was counted only once within each year,<br />

regardless of whether it was recaptured in subsequent<br />

years. By applying this procedure, a total number of 730<br />

data entries for C. viridiflavus and 448 for E. longissima<br />

were obtained. In order to evaluate yearly variations in<br />

habitat use of the two species, we used data only from the<br />

first capture of each specimen within each year under the<br />

assumption that records of habitat usage from recaptures<br />

within short-time spans could be biased by the disturbance<br />

effect of previous captures.<br />

All statistical tests, using SPSS and STATISTICA<br />

(versions for Windows), were two-tailed and alpha was<br />

set at 5 %. Means are presented followed by ± one standard<br />

deviation.<br />

RESULTS<br />

INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

● Do the populations of Coluber viridiflavus and Elaphe<br />

longissima exhibit annual fluctuations in <strong>Mediterranea</strong>n<br />

habitats?<br />

▲ In total, 511 adult C. viridiflavus (270 males, 241<br />

females) and 294 adult E. longissima (153 males, 141<br />

females) were captured. Adult sex-ratio did not depart<br />

significantly from equality in either species (Fisher exact<br />

test, NS). Males attained significantly larger SVL than<br />

females in both species (C. viridiflavus: males, x = 83.8<br />

± 11.1 cm, N = 254; females, x = 74.3 ± 9.1 cm, N =<br />

233; differences significant at P < 0.000001, t-test with<br />

df = 485; — E. longissima: males, x = 95.8 ± 15.2 cm,<br />

N = 149; females, x = 73.8 ± 14.4 cm, N = 140; differences<br />

significant at P < 0.000001, t-test with df = 287).<br />

The yearly ratio between captured specimens of the two<br />

species was always skewed towards C. viridiflavus, and<br />

was relatively stable across years at a value > 1.5: 1 (mean<br />

ratio: 1.65: 1, Standard Deviation = 0.32). However, in<br />

1991 there was a considerable excess of C. viridiflavus<br />

versus E. longissima (2.49: 1), and in 1994 and 1996 the<br />

number of C. viridiflavus was just slightly higher than<br />

that of E. longissima (1.37: 1, and 1.29: 1, respectively).<br />

If we pool data for four habitat types, the mean density<br />

per year of C. viridiflavus adults was 2.32 ± 1.05 specimens<br />

x ha -1, and the mean density of E. longissima<br />

adults was 1.34 ± 0.89 specimens x ha -1.<br />

● Do Coluber viridiflavus and Elaphe longissima exhibit<br />

any resource partitioning in terms of habitat utilization,<br />

and any patterns occur consistently among years over a<br />

long-term period?<br />

▲ The numbers of snake sightings in the various macrohabitats,<br />

year-by-year from 1986 to 1996, are presented<br />

in table 1. Coluber viridiflavus was observed essentially<br />

in habitats CLR (31.6 % of the total observations,<br />

N = 730) and CUL (45.1 %), whereas E.longissima were<br />

found in CLR (31.6 %, N = 448) and OAK (38.4 %).<br />

YEAR CLR CLR OAK OAK ROV ROV CUL CUL<br />

E.l. C.v. E.l. C.v. E.l. C.v. E.l. C.v.<br />

1986 11 23 14 6 10 7 7 33<br />

1987 16 28 13 3 13 18 8 38<br />

1988 18 21 19 5 8 13 4 36<br />

1989 9 19 12 7 7 12 8 31<br />

1990 14 22 17 3 4 6 4 35<br />

1991 7 26 16 4 2 15 4 27<br />

1992 13 17 13 1 6 16 5 23<br />

1993 20 20 14 1 2 11 2 30<br />

1994 15 24 19 4 7 13 7 25<br />

1995 10 16 17 3 5 10 7 29<br />

1996 13 15 18 4 6 8 4 22<br />

Total 146 231 172 41 70 129 60 329<br />

% Total 32.6 31.6 38.4 5.6 15.6 17.7 13.4 45.1<br />

Table 1. Number of snake sightings in the various macro-habitats of the “Macchia di Manziana” area, year-by-year from 1986 to 1996.<br />

Symbols: C.v. = Coluber viridiflavus; E.l. = Elaphe longissima; CLR = Forest clearings; OAK = Dense oak forest;<br />

ROV = Sparse oak forest with bushes (Rubus, Cytisus); CUL = Cultivations with bushes.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

79


80<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

A factorial General Linear Model (GLM) with year as<br />

the fixed factor, habitat types as random factors, and the<br />

ratio ‘numbers of C. viridiflavus / numbers of E. longissima’<br />

as the dependent variable, revealed a significant<br />

effect for habitat type (F3,30 = 21.389, P < 0.0001), but<br />

not for year (F3,10 = 1.594, P = 0.156; intercept: F1,3 =<br />

4.294, P = 0.130). When applying the same GLM to<br />

E. longissima as the dependent variable, with year as the<br />

fixed factor, and habitat types as random factors, we obtained<br />

the same result, i.e. the frequency of the study species<br />

did not vary among years (F10,30 = 1.121, P = 0.379)<br />

but depended upon the habitat type (F3,30 = 31.649,<br />

P < 0.0001; intercept: F1,3 = 12.191, P = 0.026).<br />

Repeating the same statistical model for C. viridiflavus,<br />

we obtained an almost identical result, i.e. the year did<br />

not have any statistically significant effect (F10,30 = 2.048,<br />

P = 0.063), whereas the habitat type did (F3,30 = 113.169,<br />

P < 0.000001; intercept: F1,3 = 8.559, P = 0.061).<br />

The general patterns were:<br />

(i) C. viridiflavus was abundant in sunny habitats and<br />

tended to avoid habitats with closed canopy (OAK);<br />

(ii) E. longissima clearly preferred habitats with closed<br />

canopy (OAK);<br />

(iii) both species were generalists, as they were found<br />

in all the macro-habitats available at the study area and<br />

exhibited values of habitat niche breadth typical for<br />

generalist species, when a series of four habitat types<br />

are available (table 2);<br />

(iv) the values of habitat niche breadth did not differ<br />

significantly either between species (one-way ANOVA:<br />

F1,20 = 2.054, P = 0.167), or among years (F10,11 =<br />

1.193, P = 0.385);<br />

(v) the proportion of snakes observed in the various<br />

macro-habitats was not significantly related to the field<br />

effort performed in each macro-habitat type, either<br />

considering year-by-year separately (C. viridiflavus:<br />

Spearman’s r = 0.218, N = 44, P > 0.532; E. longissima:<br />

r = 0.313, N = 44, P > 0.455) or pooling all the years<br />

together (C. viridiflavus: r = 0.795, N = 4, P = 0.205;<br />

E. longissima: r = 0.698, N = 4, P = 0.302).<br />

The values of habitat niche overlap (table 2) indicated<br />

a clear-cut habitat partitioning between species (0.5<br />

< O jk < 0.72) in most years, although in 1989 the interspecific<br />

habitat overlap was relatively high (O jk = 0.822),<br />

and in 1996 it was extremely low (O jk = 0.240).<br />

● Do Coluber viridiflavus and Elaphe longissima exhibit<br />

any resource partitioning in terms of food preferences?<br />

▲ Prey type. — The food items recorded from the two<br />

species at the study area are presented in tables 3 and<br />

4. Data are expressed on the basis of the sex and the<br />

age (juveniles or adults) of the specimens examined.<br />

Concerning C. viridiflavus, 211 prey items were recorded.The<br />

bulk of the diet of juveniles consisted of lacertid<br />

lizards, with invertebrates and newborn rodents also<br />

occasionally eaten. Adults of both sexes preyed mainly<br />

upon small sized vertebrates, i.e. lacertids, rodents, and<br />

birds. Nonetheless, there were remarkable inter-sexual<br />

differences (c 2 test with df = 2, P < 0.001), in that males<br />

consumed a significantly higher amount of lacertids than<br />

rodents, whereas females consumed an almost identical<br />

amount of rodents and lacertids. Birds were also<br />

preyed upon in similar rates by the two sexes (fig. 2A,<br />

p. 82). Frogs were occasionally consumed.<br />

127 prey items were recorded from E. longissima specimens<br />

(table 4).The bulk of the diet of juveniles consisted<br />

of lacertid lizards, but newborn rodents were also<br />

occasionally consumed. Adults of both sexes preyed only<br />

upon small sized vertebrates, mainly rodents, birds, and<br />

lacertid lizards.The ontogenetic change in dietary composition<br />

was very significant (c 2 test with df = 1,<br />

P < 0.00001), with juveniles taking more lizards and<br />

less mammals and birds than adults of both sexes. If we<br />

consider the adults, there were no statistically significant<br />

inter-sexual differences in dietary composition<br />

(c 2 test with df = 2, P > 0.05), although males tended<br />

to consume more lacertids than females. Birds were also<br />

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996<br />

C.v. = B s 1.672 1.718 1.706 1.776 1.578 1.778 1.739 1.645 1.773 1.633 1.745<br />

E.l. = B s 1.946 1.949 1.776 1.961 1.715 1.608 1.855 1.481 1.834 1.814 1.629<br />

O (C.v.-E.l.) 0.718 0.792 0.617 0.822 0.523 0.526 0.680 0.535 0.720 0.710 0.240<br />

Table 2.Values of macro-habitat niche breadth (Bs) and niche overlap (O) between Coluber viridiflavus (C.v.) and Elaphe longissima (E.l.),<br />

at a mature Oak Forest in central Italy (“Macchia di Manziana”).Values are calculated, year-by-year between 1986 and 1996, by Simpson’s (1949)<br />

diversity index (niche breadth values) and by Pianka’s (1986) symmetric equation (niche overlap values).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

PREY TYPE JUVENILES ADULT ADULT<br />

MALES FEMALES<br />

Oligochaeta<br />

Lumbricus sp.<br />

Arthropoda<br />

1 (1) – –<br />

Lithobius sp. 1 (1)<br />

Orthoptera<br />

Amphibia<br />

4 (3) 1 (1) 1 (1)<br />

Rana italica<br />

Reptilia<br />

– – 1 (1)<br />

Elaphe longissima (juv.) – – 1 (1)<br />

Anguis fragilis – 1 (1) –<br />

Chalcides chalcides – 1 (1) –<br />

Lacerta bilineata 3 (3) 17 (15) 13 (11)<br />

Podarcis muralis 33 (26) 21 (16) 17 (13)<br />

Podarcis sicula<br />

Aves<br />

2 (2) 4 (3) 3 (3)<br />

Passeriformes (ad) – 8 (7) 9 (6)<br />

Passeriformes (nestlings)<br />

Mammalia<br />

– 6 (2) 7 (4)<br />

Clethrionomys glareolus – 14 (11) 19 (15)<br />

Apodemus flavicollis 2 (2) 4 (4) 6 (4)<br />

Rodentia indet. – 3 (3) 4 (4)<br />

Crocidura sp. – 1 (1) 2 (2)<br />

Insectivora indet. – 1 (1) –<br />

Table 3. Prey eaten by Coluber viridiflavus at a mature Oak Forest<br />

in central Italy (“Macchia di Manziana”), years 1986-1996.<br />

Both the absolute numbers of food items and the numbers of individuals<br />

from which a given prey type have been found are indicated (in parentheses).<br />

PREY TYPE JUVENILES ADULT ADULT<br />

MALES FEMALES<br />

Reptilia<br />

Lacerta bilineata 2 (2) 5 (5) 4 (4)<br />

Podarcis muralis 9 (7) 8 (5) 5 (4)<br />

Podarcis sicula<br />

Aves<br />

3 (3) – –<br />

Parus caeruleus – 2 (2) 1 (1)<br />

Sylvia atricapilla – 1 (1) –<br />

Serinus serinus – – 2 (2)<br />

Carduelis chloris – 3 (1) 1 (1)<br />

Passeriformes indet. (ad) – 2 (2) 2 (2)<br />

Passeriformes (nestlings)<br />

Mammalia<br />

– 4 (2) 5 (2)<br />

Clethrionomys glareolus 2 (2) 8 (8) 12 (10)<br />

Apodemus flavicollis 3 (3) 17 (12) 19 (15)<br />

Rattus rattus – 1 (1) 4 (4)<br />

Crocidura sp. – 1 (1) –<br />

Insectivora indet. – 1 (1) –<br />

Table 4. Prey eaten by Elaphe longissima at a mature oak forest<br />

in central Italy (“Macchia di Manziana”), years 1986-1996.<br />

Both the absolute numbers of items and the numbers of individuals<br />

from which a given prey type has been found are indicated (in parentheses).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

preyed upon in similar rates by the two sexes (fig. 2B,<br />

p. 82). In all classes of individuals, and in both species,<br />

the preponderance of Podarcis muralis versus P. sicula<br />

in the diets depended merely on the much higher abundance<br />

in the field of the former species versus the latter<br />

(Rugiero, Capizzi & Luiselli, unpublished data).<br />

In terms of taxonomic dietary composition, there were<br />

significant inter-specific differences amongst adults<br />

(c 2 test, P < 0.00005):<br />

(i) the dietary spectrum of C. viridiflavus was wider<br />

than that of E. longissima at all ages, and included frogs<br />

and invertebrates which were not taken at all by the latter<br />

species;<br />

(ii) C. viridiflavus preyed much more frequently than<br />

E. longissima on lacertids and other reptiles, and also<br />

occasionally consumed snakes and snake-shaped lizards<br />

(Chalcides chalcides and Anguis fragilis);<br />

(iii) C.viridiflavus preyed less frequently than E.longissima<br />

on rodents and other small mammals;<br />

(iv) C. viridiflavus preyed more frequently on Clethrionomys<br />

glareolus than on Apodemus flavicollis, however,<br />

the opposite was true for E. longissima;<br />

(v) The two snake species consumed a similar amount<br />

of small sized birds (Passeriformes).<br />

The numbers of food items recorded for each species<br />

were not enough to allow for any search of consistent<br />

patterns across years.<br />

▲ Prey size. — Unfortunately, our data on prey size are<br />

less comprehensive than that of prey type because in<br />

several instances the weight of the ingested bolus was<br />

not recorded.This prevented us from calculating intersexual<br />

or inter-annual differences in terms of ‘predator-size-prey-size’<br />

relationships.<br />

We recorded prey size and predator size for 31 C. viridiflavus<br />

(snake mass = 151.5 ± 71.5 g; prey mass =<br />

11.4 ± 7.7 g, N = 31), and 21 E.longissima (snake mass<br />

= 186.4 ± 101.4 g; prey mass = 27.3 ± 20.0 g, N = 21).<br />

Mean snake mass was not significantly different between<br />

the species (t = 1.55, df = 50, P = 0.154) but<br />

E. longissima preyed upon significantly heavier prey than<br />

C. viridiflavus (t = 4.01, df = 50, P = 0.0002). Prey<br />

mass/predator mass ratio averaged 0.075 ± 0.11 g in<br />

C. viridiflavus,and 0.147 ± 0.19 g in E.longissima.These<br />

averages were not significantly different (t = 1.70, df =<br />

50, P = 0.0958). In both species there was a positive<br />

relationship between prey mass and predator mass<br />

(C. viridiflavus:Spearman’s r = 0.739, N = 31, ANOVA<br />

F 1,29 = 32.63, P < 0.0001; regression equation — Prey<br />

Mass = -0.673 + 0.08 x Body Mass; figure 3A, p. 83;<br />

81


82<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

Figure 2. Proportion of the main types<br />

of vertebrates (rodents, lacertids, and birds)<br />

as prey of juveniles, adult males<br />

and adult females of Coluber viridiflavus<br />

(A) and Elaphe longissima (B)<br />

at a mature Oak Forest in central Italy<br />

(“Macchia di Manziana”).<br />

Figure 1.Total number<br />

of specimens of<br />

Coluber viridiflavus<br />

and Elaphe longissima,<br />

observed year-by-year<br />

(1986-1996) at a mature<br />

Oak Forest in central Italy<br />

(“Macchia di Manziana”).<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

Figure 3. Relationships between prey mass<br />

and predator mass in (A) Coluber viridiflavus<br />

and (B) Elaphe longissima at a mature<br />

Oak Forest in central Italy<br />

(“Macchia di Manziana”).<br />

For statistical details, see the text.<br />

E. longissima: r = 0.701, N = 21, ANOVA F 1,19 = 18.4,<br />

P = 0.004; regression equation – Prey Mass = 1.554 +<br />

0.138 x Body Mass; fig. 3B). The regression line for<br />

C. viridiflavus did not differ from that for E. longissima<br />

in terms of slope (ANCOVA: F 1,48 = 2.747, P = 0.104,<br />

pooled slope = 0.114) but did for elevation (ANCOVA<br />

F 1,49 = 15.233, P = 0.0003).<br />

It is well known that, in snakes, prey size is dependent<br />

on predator size not only in terms of absolute size<br />

(length and mass), but also in terms of relative head size<br />

(see Arnold, 1993). Concerning our species, there were<br />

not only significant inter-sexual differences in mean<br />

head sizes (e.g. Springolo & Scali, 1999), but also a<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

reversed condition in comparison to absolute sizes:<br />

although E. longissima attained on average a bigger size<br />

(longer SVL and heavier mass) than C. viridiflavus, it<br />

nevertheless had a smaller head size (relatively to body<br />

length) than the latter species (Rugiero et al., unpublished<br />

data from several localities of <strong>Mediterranea</strong>n central<br />

Italy). However, we did not measure head length in<br />

most of the snake specimens captured at the study area,<br />

and so we could not directly analyse the relationships<br />

between head size and prey size. To handle this problem,<br />

we retained for the analyses those cases in which<br />

the head size of the fed snakes was measured (N = 8 in<br />

C. viridiflavus, N = 10 in E. longissima), and predicted<br />

83


84<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

(by the use of partial correlation analysis) the head<br />

length of all the other fed specimens.To do this, we calculated<br />

the slope of the regression SVL against head<br />

length for 117 C. viridiflavus and 64 E. longissima examined<br />

by us in the area of Rome (years 1992-1997)<br />

(the relative statistics are not presented here for brevity,<br />

but are available from authors), and calculated from the<br />

slopes of these regressions the predicted head length of<br />

our specimens for their known sex and SVL.Then, we<br />

run (i) the predicted measures of head length against<br />

the measures of prey mass, and (ii) the SVL of the<br />

various snakes against the prey mass, and finally we<br />

compared slopes of these two regressions by an heterogeneity<br />

of slopes test. In C. viridiflavus, we obtained<br />

significant positive relationships between both head<br />

length and SVL versus prey mass (in all cases, Spearman’s<br />

r > 0.55, N = 31, P < 0.001), and the slopes of<br />

these two regression lines were similar (F = 4.325, P ><br />

0.21).The same was not true for E. longissima, in which<br />

the slope of the regression line between SVL and prey<br />

mass differed significantly from that of the regression<br />

line between head length and prey mass (Spearman’s r<br />

= 0.216, N = 21, P > 0.1; heterogeneity of slopes test:<br />

F = 21.443, P < 0.0001). Practically speaking, whereas<br />

Figure 4. Relationships between<br />

body and air temperatures in<br />

(A) Coluber viridiflavus and<br />

(B) Elaphe longissima at a mature<br />

Oak Forest in central Italy<br />

(“Macchia di Manziana”).<br />

For statistical details, see the text.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

in C. viridiflavus the specimens with a bigger head tended<br />

strongly to prey upon bigger organisms, the same<br />

relationships was not evident in E. longissima. It is certainly<br />

a surprising result in view of other snake studies<br />

published to date, where there is a strong parallel between<br />

snake body size, head size and prey size (e.g.<br />

Forsman, 1991a, b, 1992, 1993; Werner, 1994).<br />

● Do Coluber viridiflavus and Elaphe longissima differ<br />

from each another in terms of thermal ecology attributes?<br />

▲ A total of 33 Tb readings were taken from adult<br />

C. viridiflavus, and 21 from adult E. longissima. Tb averaged<br />

25.4 ± 3.1°C in C. viridiflavus, and 23.4 ± 1.8 °C<br />

in E. longissima. These averages were significantly different<br />

(t = 2.68, df = 52, P = 0.0099).Ta averages were<br />

23.6 ± 4.1 °C relating to captured C. viridiflavus, and<br />

22.2 ± 3.6 °C for E.longissima. These averages were not<br />

significantly different (t = 1.28, df = 52, P = 0.206).<br />

Tb was positively correlated to Ta in both species (figure<br />

4; for C. viridiflavus: Spearman’s r = 0.855, ANOVA<br />

F1,31 = 84.38, P < 0.0001; regression equation – Tb =<br />

10.303 + 0.640 x Ta; for E. longissima: r = 0.857,<br />

ANOVA F1,19 = 52.65, P < 0.0001; regression equation<br />

– Tb = 13.723 + 0.438 x Ta). The regression line for<br />

C. viridiflavus did not differ from that for E. longissima<br />

in terms of slope (ANCOVA: F1,50 = -0.099, P = 0.993,<br />

pooled slope = 1.240) or elevation (ANCOVA F1,51 =<br />

2.713, P = 0.106). In both species, the regression Ta- Tb differed significantly from the line of ideal perfect<br />

poikilothermy (i.e. the line of absolute correlation between<br />

Ta and Tb) (for C. viridiflavus: ANCOVA on<br />

slopes, F1,62 = 26.618, P < 0.0001; for E. longissima:<br />

ANCOVA on slopes, F1,38 = 86.536, P < 0.0001).<br />

Considering that in some snake species (both from<br />

<strong>Mediterranea</strong>n and tropical regions) Tb is influenced by<br />

body size (SVL) (Luiselli & Zimmermann, 1997;<br />

Luiselli & Akani, 2002), we tested whether the same<br />

relationship occurred in our study species. Tb was not<br />

significantly correlated to SVL in either species (fig. 5;<br />

for C. viridiflavus: r = 0.201, ANOVA F1,31 = 1.305,<br />

P > 0.35; regression equation – Tb = 19.726 + 0.067 x<br />

SVL; for E.longissima:r = -0.051, ANOVA F1,19 = 0.049,<br />

P > 0.82; regression equation – Tb = 24.235 - 0.08 x<br />

SVL).The regression line for C.viridiflavus did not differ<br />

from that for E. longissima in terms of slope<br />

(ANCOVA: F1,50 = 1.066, P = 0.308, pooled slope =<br />

1.240) but did for elevation (ANCOVA F1,51 = 7.159,<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

P = 0.01). In C. viridiflavus, a few gravid females with<br />

a SVL of 85-95 cm had higher body temperatures<br />

(> 30 °C) than other individuals (fig. 5, p. 86).<br />

DISCUSSION<br />

● Do the populations of Coluber viridiflavus and Elaphe<br />

longissima exhibit annual fluctuations in <strong>Mediterranea</strong>n<br />

habitats?<br />

▲ We demonstrated that, apart from minor variations<br />

(likely due to random effects in the samples) both the<br />

absolute apparent abundance and the relative abundance<br />

of the two species (i.e. the ratio between the abundances<br />

of one species versus the other) did not vary<br />

significantly between years.This suggests a relative stability<br />

of our study system over the years, which in turn<br />

is likely to be due to the mature and stable conditions<br />

of the general environment, which is currently experiencing<br />

only moderate disturbance from forestry activities.<br />

In addition, habitat fragmentation is relatively<br />

scarce, and there is good quality habitat along the buffer<br />

zones; both these factors probably contribute to the<br />

observed stability in the abundance patterns of the two<br />

species.<br />

The average density of C. viridiflavus was slightly lower<br />

at our study area than at a thermo-<strong>Mediterranea</strong>n coastal<br />

locality of central Italy (Castel Fusano), where it was<br />

approximately 3.5 specimens x ha-1 (Filippi & Luiselli,<br />

2001). On the other hand, the density of E. longissima<br />

is nearly identical in these two sites (being approximately<br />

1.5 specimens x ha-1 at Castel Fusano (Filippi &<br />

Luiselli, 2001).This is likely to be dependent upon the<br />

more continental bio-climate of Manziana compared to<br />

Castel Fusano.There are very few data on the densities<br />

of these snakes from other European regions. In southern<br />

Switzerland, Bretscher (1999) found a density of<br />

E. longissima nearly identical to that observed at our<br />

Italian sites (i.e. 1.3 specimens x ha-1 ), and a density of<br />

C. viridiflavus that was slightly lower (1.4 specimens x<br />

ha-1 ). His study area was, however, more heterogeneous<br />

than our own. Other studies from southern Switzerland<br />

present density estimations of 46-83 specimens x ha-1 for E. longissima (Pillet, 1997) and 40 specimens x ha-1 for C. viridiflavus (Hofer, 2001a). These estimations<br />

cannot be compared directly with our own because they<br />

are based on counts along a line transect over a longterm<br />

period, and do not take into account mortality,<br />

immigration and emigration rates.<br />

85


86<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

In addition, our study demonstrated a sexual size dimorphism<br />

in both species studied here, where males were<br />

longer than females.This is in full agreement with previous<br />

studies on Italian conspecific populations of both<br />

C. viridiflavus (Springolo & Scali, 1999; Capula et al.,<br />

2000; Scali & Montonato, 2000) and E.longissima (Scali<br />

& Montonato, 2000) and these unsurprising results do<br />

not merit further discussion.<br />

● Do Coluber viridiflavus and Elaphe longissima exhibit<br />

any resource partitioning in terms of habitat utilization,<br />

and any patterns occur consistently among years over a<br />

long-term period?<br />

Figure 5. Relationships between body temperatures<br />

and body length in (A) Coluber viridiflavus and<br />

(B) Elaphe longissima at a mature Oak Forest<br />

in central Italy (“Macchia di Manziana”).<br />

In the plot, relative to C. viridiflavus, note that<br />

the six individuals with SVL of approximately<br />

85-95 cm and high body temperatures<br />

(around 30-33 °C) were gravid females.<br />

For statistical details, see the text.<br />

▲ At our study area, the two species appeared different<br />

from each another in terms of macro-habitat selection.<br />

In contrast, the populations of the two species from southern<br />

Switzerland showed a habitat overlap of approximately<br />

81 % (Hofer, 2001b), although E.longissima was<br />

again more linked than C. viridiflavus to sites with closed<br />

canopy (Bretscher, 1999; Hofer, 2001b). A strong<br />

preference for cool, forested spots was also observed in<br />

E.longissima from Neckar-Odenwald region (Germany)<br />

(Gomille, 2002).<br />

The inter-specific differences in the use of macro-habitat<br />

types, although predictable on the basis of previous<br />

accounts from both Italy (Luiselli & Capizzi, 1997) and<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

elsewhere (Vuagniaux, 1977; Naulleau, 1984; Bretscher,<br />

1999; Hofer, 2001a, 2001b), may be related to either<br />

synecological issues (i.e. habitat partitioning between<br />

sympatric competitors to reduce inter-specific competition<br />

- see Saint Girons, 1975 for another case of sympatric<br />

European snakes) or eco-physiological factors (i.e.<br />

external proximate constraints on the life-history of the<br />

two species – cf. Barbault, 1991; Barbault & Stearns,<br />

1991). Our data cannot exclude either hypothesis, and<br />

it is likely that both may contribute to explain the observed<br />

patterns (see also Filippi & Luiselli, 2001).To resolve<br />

which of the two issues contributes more to the observed<br />

pattern, specific ad-hoc experiments (e.g. controlled<br />

removal) are needed. To properly consider synecology<br />

we must include the effects of interference from other<br />

competitors (e.g. other snakes, mustelids, raptors, etc.)<br />

that have proven to interact with C. viridiflavus and<br />

E. longissima elsewhere (e.g., Capizzi & Luiselli, 1996).<br />

The general preference of C. viridiflavus for open<br />

(drier and sunnier) macro-habitats is most likely related<br />

to its specific thermal and physiological tolerances,<br />

and this correlates well with the overall distribution of<br />

this species in Europe, especially considering that it<br />

can be found in the very hot regions in the extreme<br />

south of Italy (Bruno & Maugeri, 1977). At the same<br />

time, the preference of E. longissima for forested spots<br />

is likely to be related to the fact that this is a northernmore<br />

species and central Italy (southern Latium) is its<br />

southernmost limit. Indeed, immediately south of<br />

Latium, E. longissima is replaced by Elaphe lineata,<br />

which was until recently considered a subspecies of<br />

longissima but considerable genetic and morphological<br />

divergence has now promoted it to the rank of full<br />

species (Lenk & Joger, 1994; Lenk & Wüster, 1999).<br />

Similar habitat utilisation patterns to the ones found<br />

in this study occur in a coastal locality of <strong>Mediterranea</strong>n<br />

central Italy, where C. viridiflavus was found<br />

in dry and sunny places and E. longissima in cooler and<br />

shadier places (Filippi & Luiselli, 2001). Thus, external<br />

eco-physiological cons-traints (e.g. climate) may<br />

well explain the local macro-habitat selection of the<br />

two species, as well the temporal constancy of this<br />

apparent resource partitioning pattern which was<br />

observed at our study area.<br />

● Do Coluber viridiflavus and Elaphe longissima differ<br />

from each another in terms of thermal ecology attributes?<br />

▲ Our data on the thermal ecology of the two species<br />

also support this suggestion because C.viridiflavus maintains<br />

higher body temperatures than sympatric E. lon-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

gissima. Considering that (i) the slopes of the regressions<br />

T b against T a are similar between species, and that (ii)<br />

in both cases they are significantly different from the<br />

ideal line of perfect poikilothermy, it is likely that the T bs<br />

of C. viridiflavus were higher because they were captured<br />

in sunnier spots than specimens of E. longissima. It<br />

is also evident that E.longissima maintained its T b at lower<br />

values by both avoiding direct exposure to the sun (the<br />

same as in several tropical snakes, e.g. Akani et al.,2002)<br />

and also becoming active earlier in the morning (Rugiero<br />

et al., unpublished data). In practice, the thermoregulatory<br />

ability of the two species should be physiologically<br />

similar but, given their different thermal requirements,<br />

the two species adopt diverging behavioural and microhabitat<br />

selection strategies to maintain their T bs around<br />

their species-specific optimal values. In contrast to semiaquatic<br />

snakes from both Europe (cf. Luiselli &<br />

Zimmermann, 1997) and tropical Africa (Luiselli &<br />

Akani, 2002), there was no influence of body size (SVL)<br />

on T bs of either species, and this can be interpreted either<br />

as inefficient thermoregulation or the lack of a physiological<br />

trade-off between body size and T b. In this regard<br />

it is, however, important to remark that a few gravid C.<br />

viridiflavus had higher T bs than other individuals (males<br />

and non-gravid females).This may reflect a shift towards<br />

selecting higher T bs in gravid individuals (Beuchat,<br />

1986), as has been observed in another European oviparous<br />

snake (Luiselli & Zimmermann, 1997). If this is<br />

the case, it is evident that gravid females should select<br />

higher T bs by being active at high T as (> 30 °C, see figure<br />

4). However, our data are too few to draw any firm<br />

conclusions, especially considering that there is no general<br />

rule for squamate reptiles (cf. Gibson & Falls, 1979;<br />

Shine & Lambeck, 1990; Schwarzkopf & Shine, 1991;<br />

Akani et al., 2002; Luiselli & Akani, 2002). Ad-hoc experiments<br />

are to be planned to resolve this issue.<br />

● Do Coluber viridiflavus and Elaphe longissima exhibit<br />

any resource partitioning in terms of food preferences?<br />

▲ As previously observed in another agro-forest environment<br />

of <strong>Mediterranea</strong>n central Italy (Capizzi &<br />

Luiselli, 1996), the feeding niche of C. viridiflavus was<br />

remarkably different from that of E.longissima.This can<br />

be better seen by separately examining (a) adults and<br />

juveniles, and (b) prey types and prey sizes.<br />

The major inter-specific differences were observed when<br />

comparing the taxonomic dietary composition of juveniles<br />

of the two species. Indeed, C. viridiflavus juveniles<br />

87


88<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

were much more eclectic predators than juvenile E.longissima,<br />

preying upon many lizards and newborn mice<br />

as well as several invertebrates (earthworms and arthropods),<br />

which were totally avoided by E. longissima.<br />

Considering that snakes respond to both chemical and<br />

visual stimuli (Arnold, 1993) and most are relatively<br />

specialized in their prey preferences (Mushinsky, 1987),<br />

this indicates that C.viridiflavus may respond to an unusually<br />

wide range of stimuli and, hence, have an unusually<br />

wide dietary spectrum.<br />

The same trend, although less pronounced, was also<br />

observed in the adults; both C. viridiflavus and E. longissima<br />

foraged on lizards, rodents and birds, but<br />

C. viridiflavus occasionally also fed on frogs and<br />

arthropods, which were not consumed at all by E. longissima<br />

(Luiselli & Rugiero, 1993; Rugiero & Luiselli,<br />

1995). Other spectacular examples of high dietary<br />

generalism are found among tropical elapids, e.g.<br />

cobras and tree cobras (Pawvels et al., 1999; Luiselli<br />

& Angelici, 2000; Luiselli et al., 2002). It is also noteworthy<br />

that both species had several birds in their stomachs,<br />

which reflects the good climbing ability, viz.<br />

semi-arboreal habits, of these snakes (Naulleau, 1984,<br />

1987, 1989). A case of direct predation of the one species<br />

(C. viridiflavus) versus the other (E. longissima)<br />

was also recorded, but it is very unlikely that this kind<br />

of interference competition may be of any relevance<br />

for the dynamics of the studied system. It must be noticed<br />

that C. viridiflavus has been sometimes claimed to<br />

exercise a strong pressure on the populations of sympatric<br />

snakes (e.g., Cattaneo & Carpaneto, 2000, for<br />

the case of Elaphe quatuorlineata), but there is no<br />

reliable field dataset to confirm this assertion.<br />

If we consider that offspring and subadult snakes<br />

are often prey-limited and relatively emaciated when<br />

captured in the wild (Filippi, 1995; Luiselli &<br />

Angelici, 1996), it may be assumed that the interspecific<br />

differences observed may greatly help<br />

coexistence of these species via reducing inter-specific<br />

competition, which is likely more intense<br />

among juvenile snakes than among adult snakes, at<br />

least in the <strong>Mediterranea</strong>n environments (Luiselli &<br />

Angelici, 1996). On the other hand, we do not<br />

believe that the taxonomic diet difference between<br />

the adults of these species reflects competitive pressure<br />

to reduce dietary niche overlap.This is because<br />

adult C. viridiflavus has a diet composition completely<br />

different from that of E. longissima even in<br />

areas where the latter species is not present, e.g. in<br />

Montecristo island (Thyrrenian sea, Tuscan archi-<br />

pelago – where its main prey are anuran amphibians<br />

(Zuffi, 2001). The fact that C. viridiflavus preyed<br />

more upon Clethrionomys glareolus than Apodemus<br />

flavicollis, whereas the reverse was true for E. longissima,<br />

does not again imply any resource partitioning,<br />

but merely reflects the different habitat selection<br />

of these two rodents. Clethrionomys glareolus are<br />

found in clearings, which is better habitat for C. viridiflavus<br />

than E. longissima; and Apodemus flavicollis<br />

occur in mature forests, which are better habitat for<br />

E. longissima than C. viridiflavus (Amori et al., 1986;<br />

Capizzi & Luiselli, 1996b). So, we definitely think<br />

that the differences in taxonomic dietary composition<br />

reflect diverging species-specific preferences in<br />

habitat utilization (and hence in prey communities<br />

available) rather than community-based mechanisms<br />

of resource partitioning.<br />

Regarding the relationship of prey-size relative to predator-size,<br />

we observed that C. viridiflavus occasionally<br />

foraged on very small organisms (arthropods)<br />

and E. longissima occasionally ate relatively large animals<br />

(rats), but in most cases the two species showed<br />

a similar pattern.<br />

● Can the interspecific differences observed be interpreted<br />

as a result of competitive relationships or via species-specific<br />

eco-physiological constraints?<br />

▲ In conclusion, our study demonstrated that: (1)<br />

these snake species do exhibit a clear-cut pattern of<br />

resource partitioning in terms of habitat utilization,<br />

and, in consequence of it, also in terms of dietary<br />

habits, and these patterns occur consistently among<br />

years over a long-term period; (2) the populations of<br />

these snakes do not exhibit remarkable annual fluctuations<br />

in <strong>Mediterranea</strong>n habitats; (3) these snakes<br />

differed clearly each from the other in terms of thermal<br />

ecology attributes, and it is most probably linked<br />

to interspecific differences in micro-habitats (Filippi<br />

& Luiselli, 2001) and macro-habitats (Luiselli &<br />

Capizzi, 1997; this study). In general, we suggest that<br />

species-specific eco-physiological requirements appear<br />

to explain the coexistence patterns of the two species<br />

better than any community-based synecological factors.<br />

Indeed, only the niche differences in prey types<br />

of subadults may be explained with competition pressure<br />

to reduce dietary niche overlap. However, it is<br />

obvious that field experiments designated ad-hoc<br />

should be used to test specifically for the various questions<br />

posed by this study, and we are currently working<br />

on these experimental procedures.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


ACKNOWLEDGEMENTS<br />

We are indebted to the “Università Agraria di Manziana”<br />

for several kinds of assistance when in the field.<br />

Many researchers, and especially U. Agrimi, M. Capula,<br />

and E. Filippi, were of invaluable help in collecting data<br />

and discussing with us about the subject of this study.<br />

Two anonymous referees, G.C. Akani, M. Capula, F.M.<br />

Angelici, E. Filippi, E. Vidal, and especially J. Lea commented<br />

upon a preliminary draft of this manuscript, and<br />

gave also linguistic assistance.<br />

References<br />

INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

AGRIMI U., 2000. Coluber viridiflavus (Lacépède, 1789). In:<br />

Bologna, M.A., Capula, M. & Carpaneto, G.M. (eds), Anfibi<br />

e Rettili del Lazio. Fratelli Palombi Editori, Rome: 90-91.<br />

AKANI G.C., ENIANG E.A., EKPO I.J., ANGELICI F.M. &<br />

LUISELLI L., 2002. Thermal and reproductive ecology of the<br />

snake Psammophis phillipsi from the rainforest region of southern<br />

Nigeria (West Africa). Herpetological Journal, 12: 63-67.<br />

AMORI G., CRISTALDI M. & CONTOLI L., 1986. Sui roditori<br />

(Gliridae, Arvicolidae, Muridae) dell’Italia peninsulare ed<br />

insulare in rapporto all’ ambiente bioclimatico mediterraneo.<br />

Animalia,11 (1984): 217-269.<br />

ARNOLD S.J., 1993. Foraging theory and prey-size-predator-size<br />

relations in snakes. In: Seigel, R.A. & Collins, J.T. (eds.),<br />

Snakes - Ecology and Behavior, McGraw-Hill, New York:<br />

87-116.<br />

AVERY R.A., 1979. Lizards - A study in thermoregulation.E.Arnold<br />

Ltd., London.<br />

BARBAULT R., 1991. Ecological constraints and community dynamics:<br />

linking community patterns to organismal ecology.The<br />

case of tropical herpetofaunas. Acta Œcologica, 12: 139-163.<br />

BARBAULT R. & STEARNS S.C., 1991. Towards an evolutionary<br />

ecology linking species interactions, life-history strategies and<br />

community dynamics: An introduction. Acta Œcologica, 12:<br />

3-10.<br />

BEUCHAT C.A., 1986. Reproductive influences on the thermoregulatory<br />

behaviour of a live-bearing lizard. Copeia, 1986:<br />

971-979.<br />

BLONDEL J., 1990. Faunes, peuplements, populations: une perspective<br />

hiérarchique. <strong>Ecologia</strong> mediterranea, 16: 395-401.<br />

BONNET X. & NAULLEAU G., 1996a. Are body reserves important<br />

for reproduction in male dark green snakes (Colubridae:<br />

Coluber viridiflavus)? Herpetologica, 52: 137-146.<br />

BONNET X. & NAULLEAU G., 1996b. Catchability in snakes:<br />

consequences for estimates of breeding frequency. Canadian<br />

Journal of Zoology, 74: 233-239.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

BONNET X., NAULLEAU G. & SHINE R., 1999. The dangers of<br />

leaving home: dispersal and mortality in snakes. Biological<br />

Conservation, 89: 39-50.<br />

BRETSCHER A., 1999. Populationsschätzung von Zornatter (Coluber<br />

viridiflavus) und Aesculapnatter (Elaphe longissima) im<br />

Mendrisiotto. Unpublished Diploma Thesis, ETH Zurich, 37 p.<br />

BRUNO S., 1977. Gli Anfibi e i Rettili dei Monti della Tolfa<br />

(Antiappennino Laziale). In: (no eds.), Ricerche ecologiche,flogistiche<br />

e faunistiche nel comprensorio Tolfetano-Cerite-Manziate,<br />

Accademia Nazionale dei Lincei, Roma: Quaderno 227:<br />

89-124.<br />

BRUNO S. & MAUGERI S., 1977. Rettili d’Italia. Giunti-Martello<br />

Editore, Firenze.<br />

CAPIZZI D., CAPULA M., EVANGELISTI F., FILIPPI E., LUISELLI L.<br />

& TRUJILLO V.J., 1996. Breeding frequency, clutch size, reproductive<br />

status and correlated behaviours in sympatric females<br />

Elaphe quatuorlineata and Elaphe longissima (Reptilia, Colubridae).<br />

Revue d’Écologie (Terre et Vie), 51: 297-311.<br />

CAPIZZI D. & LUISELLI L., 1996a. Feeding relationships and competitive<br />

interactions between phylogenetically unrelated predators<br />

(owls and snakes). Acta Œcologica, 17: 265-284.<br />

CAPIZZI D. & LUISELLI L., 1996b. Ecological relationships between<br />

small mammals and age of coppice in an oak-mixed<br />

forest in central Italy. Revue d’Écologie (Terre et Vie), 51:<br />

277-291.<br />

CAPULA M., FILIPPI E. & LUISELLI L., 1995. Annual mating in<br />

female colubrid snakes with irregular reproductive frequency<br />

(Squamata: Colubridae). Herpetozoa, 8:11-15.<br />

CAPULA M., FILIPPI E., LUISELLI L., AGUILAR J.R. & RUGIERO<br />

L., 2000. Body size and some demographic characteristics<br />

in two populations of Coluber viridiflavus in the countryside<br />

of Rome. In:Giacoma, C. (ed.), Atti del I Congresso Nazionale<br />

della Societas Herpetologica Italica, Torino, Museo regionale<br />

di Scienze Naturali: 435-438.<br />

CAPULA M., FILIPPI E., LUISELLI L. & TRUJILLO V.J., 1997. The<br />

ecology of the Western Whip Snake, Coluber viridiflavus<br />

(Lacépède, 1789) in <strong>Mediterranea</strong>n central Italy. Herpetozoa,<br />

10: 65-79.<br />

CAPULA M. & LUISELLI L., 1995. Hierophis viridiflavus (Western<br />

Whip Snake). Communal nesting. Herpetological Review, 26:<br />

38-39.<br />

CAPULA M. & LUISELLI L., 2002. Feeding strategies of Elaphe<br />

longissima from contrasting <strong>Mediterranea</strong>n habitats in central<br />

Italy. Italian Journal of Zoology, 69: 153-156.<br />

CATTANEO A. & CAPULA M., 2000. Elaphe longissima (Laurenti,<br />

1768). In: Bologna, M.A., Capula, M. & Carpaneto, G.M.<br />

(eds), Anfibi e Rettili del Lazio.Fratelli Palombi Editori, Rome:<br />

96-97.<br />

CATTANEO A. & CARPANETO G.M., 2000. Elaphe quatuorlineata<br />

(Lacépède, 1789). In: Bologna, M.A., Capula, M. &<br />

Carpaneto, G.M. (eds), Anfibi e Rettili del Lazio. Fratelli<br />

Palombi Editori, Rome: 98-99.<br />

89


90<br />

◆ LORENZO RUGIERO, DARIO CAPIZZI, LUCA LUISELLI<br />

CIOFI C. & CHELAZZI D., 1991. Radiotracking of Coluber viridiflavus<br />

using external transmitters. Journal of Herpetology,<br />

25: 37-40.<br />

CIOFI C. & CHELAZZI D., 1994. Analysis of homing pattern in<br />

the colubrid snake Coluber viridiflavus.Journal of Herpetology,<br />

28: 477-484.<br />

FILIPPI E., 1995. Aspetti dell’ ecologia di due comunità di Colubridi<br />

e Viperidi (Reptilia: Serpentes) di un’area dell’ Italia centrale<br />

(Monti della Tolfa, Lazio). Unpublished Thesis, University of<br />

Rome “La Sapienza”, 143 p.<br />

FILIPPI E. & LUISELLI L., 2001. Use of microhabitat and substratum<br />

types by sympatric snakes in a <strong>Mediterranea</strong>n area<br />

of central Italy. <strong>Ecologia</strong> <strong>Mediterranea</strong>, 27: 141-153.<br />

FILIPPI E. & LUISELLI L., 2002. Negative effect of the wild boar (Sus<br />

scrofa) on the populations of snakes at a protected mountainous<br />

forest in central Italy. <strong>Ecologia</strong> <strong>Mediterranea</strong>, 28 (1): 93-98.<br />

FORSMAN A., 1991a.Variation in sexual size dimorphism and<br />

maximum body size among adder populations: effects of prey<br />

size. Journal of Animal Ecology, 60: 253-267.<br />

FORSMAN A., 1991b. Adaptive variation in head size in Vipera<br />

berus L. populations. Biological Journal of the Linnean Society,<br />

43: 281-296.<br />

FORSMAN A., 1992. Adaptive body and head size variation in<br />

populations of the adder Vipera berus.Acta Univ.Upsal.Compr.<br />

Summ. Uppsala Diss. Fac. Sciences, 401: 1-34.<br />

FORSMAN A., 1993.The advantage of a big head: swallowing performance<br />

in adders, Vipera berus. Functional Ecology, 7: 183-<br />

189.<br />

GIBSON R. & FALLS J.B., 1979. Thermal biology of the common<br />

garter snake Thamnophis sirtalis (L.). 1.Temporal variation,<br />

environmental effects and sex differences. Oecologia,<br />

43: 79-97.<br />

GOMILLE A., 2002. Die Äsculapnatter Elaphe longissima. Edition<br />

Chimaira, Frankfurt am Main, p. 158.<br />

HARVEY P.H. & PAGEL M.D., 1991. The Comparative Method in<br />

Evolutionary Biology. Oxford Univ. Press, Oxford.<br />

HOFER U., 2001a. Coluber (Hierophis) viridiflavus (Lacépède,<br />

1789). In: Hofer, U., Monney, J.-C. & Dusei, G. (eds.), Les<br />

reptiles de Suisse. Birkhauser Verlag, Basel: 91-96.<br />

HOFER U., 2001b. Elaphe longissima (Laurenti, 1768). In: Hofer,<br />

U., Monney, J.-C. & Dusei, G. (eds.), Les Reptiles de Suisse.<br />

Birkhauser Verlag, Basel: 97-104.<br />

HURLBERT S.H., 1984. Pseudoreplication and the design of ecological<br />

field experiments. Ecological Monographs, 54: 187-211.<br />

KAMMEL W., 1999. Zur Biologie der heimischen Elaphe longissima<br />

longissima.Unpublished Dissertation, University of Graz,<br />

p. 160.<br />

LENK P. & JOGER U., 1994. Genetic relationships between populations<br />

and intraspecific subdivision of Elaphe longissima<br />

(Laurenti, 1768) as suggested by plasma protein electrophoresis<br />

and DNA fingerprinting. Amphibia-Reptilia,15: 363-373.<br />

LENK P. & WÜSTER W., 1999. A multivariate approach to the systematics<br />

of Italian rat snakes of the Elaphe longissima complex<br />

(Reptilia, Colubridae): revalidation of Camerano’s<br />

Callopeltis longissimus var. lineata. Herpetological Journal, 9:<br />

153-162.<br />

LICHT P., DAWSON W.R., SHOEMAKER V.H. & MAIN A.R., 1966.<br />

Observations on the thermal relations of western Australian<br />

lizards. Copeia, 1966: 97-110.<br />

LUISELLI L. & AKANI G.C., 2002. Is really thermoregulation<br />

unimportant for tropical reptiles? Comparative study in four<br />

sympatric snake species from Africa. Acta Oecologica, 23: 59-<br />

68.<br />

LUISELLI L. & ANGELICI F.M., 1996. The prey spectrum of terrestrial<br />

snakes in the Tolfa Mountains (Latium, Central Italy).<br />

A synthesis from earlier analyses. Herpetozoa, 9:111-119.<br />

LUISELLI L. & ANGELICI F.M., 2000. Ecological relationships in<br />

two Afrotropical cobra species (Naja melanoleuca and Naja<br />

nigricollis). Canadian Journal of Zoology, 78: 191-198.<br />

LUISELLI L., ANGELICI F.M. & AKANI G.C., 2002. Comparative<br />

feeding strategies and dietary plasticity of the sympatric<br />

cobras Naja melanoleuca and Naja nigricollis, in three diverging<br />

Afrotropical habitats. Canadian Journal of Zoology, 80:<br />

55-63.<br />

LUISELLI L. & CAPIZZI D., 1997. Influences of area, isolation and<br />

habitat features on distribution of snakes in <strong>Mediterranea</strong>n<br />

fragmented woodlands. Biodiversity and Conservation,6:1339-<br />

1351.<br />

LUISELLI L. & RUGIERO L., 1993. Food habits of the Aesculapian<br />

snake, Elaphe longissima, in central Italy: Do arboreal snakes<br />

eat more birds than terrestrial ones? Journal of Herpetology,<br />

27: 116-117.<br />

LUISELLI L. & ZIMMERMANN P.,1997. Thermal ecology and<br />

reproductive cyclicity of the snake Natrix tessellata in southeastern<br />

Austria and central Italy: A comparative study.<br />

Amphibia-Reptilia, 18: 383-396.<br />

MAGNUSON J.J., 1990. Long-term ecological research and the<br />

invisible present. Bioscience, 40: 495-501.<br />

MATHUR D. & SILVER C.A., 1980. Statistical problems in studies<br />

of temperature preferences of fishes. Canadian Journal of<br />

Fisheries and Aquatic Sciences, 37: 733-737.<br />

MONTELUCCI G., 1977. Note preliminari sulla flora e la vegetazione<br />

delle cerrete di Manziana e di Canale-Monterano. In:<br />

(no eds.), Ricerche ecologiche, flogistiche e faunistiche nel comprensorio<br />

Tolfetano-Cerite-Manziate, Accademia Nazionale dei<br />

Lincei, Roma: Quaderno 227: 51-73.<br />

MUSHINSKY H.R., 1987. Foraging ecology. In: Seigel, R.A.,<br />

Collins, J.T. & Novak, S.S. (eds.), Snakes - Ecology and<br />

Evolutionary Biology, MacMillan, New York: 302-334.<br />

NAJBAR B., 1999a. Breeding biology of the Aesculapian snake<br />

Elaphe longissima longissima (Laurenti) in the Bieszczady<br />

Zachodnie Mountains (SE Poland). Chronmy Przyrode<br />

Ojczysta, 55 (2): 5-20.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


INTERACTIONS BETWEEN SYMPATRIC SNAKES, COLUBER VIRIDIFLAVUS AND ELAPHE LONGISSIMA… ◆<br />

NAJBAR B., 1999b.The diet of the Aesculapian snake Elaphe longissima<br />

longissima (Laurenti) in the Bieszczady Zachodnie<br />

Mountains (SE Poland). Chronmy Przyrode Ojczysta, 55 (2):<br />

21-33.<br />

NAJBAR B., 2000a. The Aesculapian snake Elaphe l. longissima<br />

(Laurenti, 1768) population in Bieszczady (Poland) between<br />

1990-1998. Bulletin of the Polish Academy of Biology,48: 41-51.<br />

NAJBAR B., 2000b. The state of the Aesculapian snake Elaphe l.<br />

longissima (Laurenti, 1768) population in Poland. Bulletin of<br />

the Polish Academy of Biology, 48: 53-62.<br />

NAULLEAU G., 1984. Les serpents de France. Revue Francaise d’<br />

Aquariologie, 11: 1-64.<br />

NAULLEAU G., 1987. Use of biotelemetry in the study of freeranging<br />

snakes: example of Elaphe longissima.In:Van Gelder,<br />

J.J., Strijbosch, H. & Bergers, P.J.M. (eds.), Proc. Fourth Ord.<br />

Gen. Meet. S.E.H., Nijmegen, 1987, Nijmegen Faculty of<br />

Sciences, Nijmegen: 289-292.<br />

NAULLEAU G., 1989. Etude biotélémètrique des déplacements<br />

et de la température chez la couleuvre d’Esculape Elaphe longissima<br />

(Squamata, Colubridae) en zone forestière. Bulletin<br />

de la Société Herpétologique de France, 52: 45-53.<br />

NAULLEAU G., 1992a. Reproduction de la couleuvre d’Esculape<br />

Elaphe longissima Laurenti (Reptilia, Colubridae) dans le<br />

Centre Ouest de la France. Bulletin de la Société Herpétologique<br />

de France, 62: 9-17.<br />

NAULLEAU G., 1992b. Activité et température corporelle automnales<br />

et hivernales chez la Couleuvre d’Esculape Elaphe longissima<br />

(Squamata, Colubridae) dans le Centre Ouest de la France.<br />

Bulletin de la Société Herpétologique de France, 64: 21-35.<br />

NAULLEAU G. & BONNET X., 1995. Reproductive ecology, body<br />

fat reserves and foraging mode in females of two contrasted<br />

snake species: Vipera aspis (terrestrial, viviparous) and Elaphe<br />

longissima (semi-arboreal, oviparous). Amphibia-Reptilia, 16:<br />

37-46.<br />

PAWVELS O.S.G., DAVID P. & LENGLET G., 1999. Pseudohaje goldii<br />

(Boulenger, 1895): Goldie’s tree cobra. Diet. African<br />

Herpetological News, 30: 32-33.<br />

PIANKA E.R., 1986. Ecology and Natural History of Desert Lizards.<br />

Princeton University Press, Princeton.<br />

PILLET J.-M., 1997. Reptiles du Valais. Unpublished Report to<br />

KARCH, Bern, 50 p.<br />

RUGIERO L. & LUISELLI L., 1995. Food habits of the snake<br />

Coluber viridiflavus in relation to prey availability. Amphibia-<br />

Reptilia, 16: 407-411.<br />

SAINT GIRONS H., 1975. Coexistence de Vipera aspis et Vipera<br />

berus en Loire-Atlantique : un problème de compétition interspécifique.<br />

Revue d’Écologie (Terre et Vie), 29: 590-613.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 75-91<br />

SCALI S. & MONTONATI S., 2000. Analisi multivariata del<br />

dimorfismo sessuale in due specie di colubrini italiani<br />

(Coluber viridiflavus ed Elaphe longissima: Reptilia,<br />

Serpentes, Colubridae) su basi biometriche. In: Giacoma,<br />

C. (ed.), Atti del I Congresso Nazionale della Societas<br />

Herpetologica Italica, Torino, Museo regionale di Scienze<br />

Naturali, p. 429-434.<br />

SCALI S. & ZUFFI M., 1994. Report on a reptile community ecology<br />

in a suburban habitat of northern Italy (Reptilia, Sauria,<br />

Serpentes). Bollettino di Zoologia, 61: 73-76.<br />

SCHOENER T.W., 1983. Field experiments on interspecific competition.<br />

American Naturalist, 122: 240-285.<br />

SCHOENER T.W., 1986. Mechanistic approaches to community<br />

ecology: a new reductionism? American Naturalist, 126: 633-<br />

641.<br />

SCHWARZKOPF L. & SHINE R., 1991. Thermal biology of reproduction<br />

in viviparous skinks, Eulamprus tympanum: why do<br />

gravid females bask more? Oecologia, 88: 562-569.<br />

SHINE R. & LAMBECK R., 1990. Seasonal shifts in the thermoregulatory<br />

behaviour of Australian blacksnakes, Pseudechis<br />

porphyriacus (Serpentes: Elapidae). Journal of Thermal Biology,<br />

15: 301-305.<br />

SIMPSON E.H., 1949. Measurement of diversity. Nature,163: 688.<br />

SPADA F., 1977. Primi lineamenti della vegetazione del comprensorio<br />

Tolfetano-Cerite-Manziate. In: (no eds.), Ricerche<br />

ecologiche, flogistiche e faunistiche nel comprensorio Tolfetano-<br />

Cerite-Manziate, Accademia Nazionale dei Lincei, Roma:<br />

Quaderno 227: 37-49.<br />

SPRINGOLO M. & SCALI S., 1999. Sexual dimorphism and ontogenetic<br />

changes in Coluber viridiflavus: a head morphometric<br />

analysis. In: Miaud, C. & Guyetant, R. (eds.), Current<br />

Studies in Herpetology, S.E.H., Le Bourget du Lac, p. 413-<br />

417.<br />

STRONG D.R., LAWTON J.H. & SOUTHWOOD R., 1984. Insects on<br />

Plants. Community Patterns and Mechanisms. Blackwell<br />

Scientific Publications, Oxford.<br />

VUAGNIAUX J.-P., 1977. Amphibiens et reptiles du Bassin Genevois.<br />

I. Ophidiens. WWF, Genève, 102 pp.<br />

WERNER Y.L., 1994. Head size variation in Cerastes (Ophidia:<br />

Viperidae) parallels body size variation in potential prey<br />

(Rodentia: Gerbillinae). The Snake, 26: 57-60.<br />

WIENS J.A., 1977. On competition and variable environments.<br />

American Scientist, 65: 590-597.<br />

ZUFFI M.A.L., 2001. Diet and morphometrics of Coluber (=<br />

Hierophis) viridiflavus on the island of Montecristo (Tyrrhenian<br />

Sea, Italy). Herpetological Journal, 11: 123-125.<br />

91


Conséquences phytoécologiques<br />

de la restauration d’un écosystème dunaire fixé<br />

avec le tamaris (Tamarix gallica L.), en Algérie<br />

Phytoecological consequences of a sand dune restauration<br />

with tamaris (Tamarix gallica L.), in Algeria<br />

Khéloufi Benabdeli 1, Halima Mohammedi 1, Noreddine Letereuch-Belarouci 2<br />

1. Faculté des sciences, département des sciences de l’environnement, Université Djilali-Liabes, BP89 Sidi Bel-Abbes, 22000 Algérie<br />

2. Faculté des sciences, département de foresterie, Université de Tlemcen, Algérie<br />

Résumé<br />

La lutte biologique contre l’ensablement grâce à l’utilisation du<br />

Tamarix gallica Wahl. donne des résultats intéressants, tant du<br />

point de vue écologique qu’économique. L’évolution phytoécologique<br />

de la dune fixée et mise en défens est nettement perceptible à travers<br />

l’analyse des résultats dendrométriques et phytoécologiques obtenus.<br />

Les résultats attestent d’une remontée biologique où le Tamarix<br />

gallica joue un rôle déterminant dans la lutte contre l’ensablement<br />

en favorisant (par une protection et un ombrage) l’installation d’une<br />

strate herbacée.<br />

Les paramètres dendrométriques permettant d’évaluer l’évolution<br />

de la production ligneuse sont le nombre de brins par cépée, la hauteur<br />

moyenne, le diamètre et la surface terrière. Leur quantification<br />

aux différents âges laisse apparaître une augmentation du<br />

nombre de brins par hectare (densité), un accroissement annuel<br />

moyen de la hauteur de 56 cm. Le diamètre à 1,30 m connaît lui<br />

aussi un accroissement moyen annuel de 0,57 cm, soit une augmentation<br />

annuelle moyenne de la surface terrière de 0,00709 m 2.<br />

Une plantation à une densité moyenne de 1 600 plants par hectare,<br />

Tamarix gallica permet une production de matière ligneuse,<br />

très sollicitée dans la région, de l’ordre de 0,28 m 3/ha et par an,<br />

soit un accroissement moyen annuel moyen de 0,31 m 3/ha. Cette<br />

production peut répondre aux besoins de deux familles et contribuera<br />

à assurer une protection de cet écosystème, utile pour ses utilisateurs.<br />

Mots-clés<br />

Tamarix gallica, paramètres dendrométriques, restauration écologique,<br />

fixation des dunes, Algérie.<br />

Short version<br />

The use of Tamarix gallica as biological sand binder provides interesting<br />

results from anecological and economical point of view. These<br />

species induces also a valuable biomass. Increased contribution of<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 93-99<br />

Saharo-sindian species balance the decrease role of ubiquitous species,<br />

but Mediterraneen species stay relatively constant. An appreciable<br />

increase of species number can be seen as composed to the<br />

control plot; increase from 6 to 12, for genera, and from 9 to 15<br />

for species. A dynamic process and fluctuation for species in each<br />

biogeographic element are then documented. The number of genera<br />

greatly depends on species fluctuations witch are under the effect of<br />

many parameters not always under control. In such a situation, it<br />

is obvious that the fencing system play a determinant role in the<br />

restoration process. Dendrometric parameters witch permit the evaluation<br />

of woody biomass production and itsevolutionare number<br />

of shoot by plant, the height mean, the diameter and ground cover.<br />

Quantification of these parameters at differents stages shows an<br />

increase of shoot number per hectare (density), and an increase of<br />

the annual mean height by 56 centimeter. The diameter at 1.30<br />

meter show also an increase of 0.57 cm witch correspond to an<br />

annual mean increase of its cover of 0.00709 m 2.<br />

With a mean plantation density of 1600 plants per hectare, Tamarix<br />

gallica induces a annual production of ligneous matter of 0.28 m 3<br />

per hectare, i.e. an annual increase of 0.31 m 3 per hectare. This<br />

production would contribute to solve some needs of fuel wood, and<br />

should benefit the ecosystem mainly because of its use by local population.<br />

The combinaison of fencing system and plantation of woody<br />

species, as Tamarix, permit to yield appreciable results. This ligneous<br />

plant with its rapid growth allow the establishment of a favorable<br />

microclimate, and the establishment of a herbaceous stratum could<br />

play an important role in the dune stability and the evolution of<br />

this ecosystem. Local population takes care of fencing audits maintenance,<br />

because of wood and animal feed produced in return of<br />

their participation of the protection process.<br />

Keywords<br />

Tamarix gallica, dendrometric parameters, ecological restoration,<br />

dune fixation, Algeria.<br />

93


94<br />

◆ KHÉLOUFI BENABDEL, HALIMA MOHAM<strong>MED</strong>I, NOREDDINE LETEURECH-BELAROUCI<br />

INTRODUCTION<br />

Par leur instabilité et leurs mouvements continuels sous<br />

l’influence des vents dominants, les ensembles dunaires sont<br />

source de préjudices économiques et écologiques pour l’espace<br />

qu’ils envahissent. La résultante de ces mouvements<br />

est un ensablement qui constitue un processus important<br />

de la désertification si l’on retient la définition qui en a été<br />

donnée par Le Houérou (1969) : « La désertification est considérée<br />

comme étant un ensemble d’actions qui se traduisent par<br />

une réduction du couvert végétal, aboutissant à l’extension de<br />

paysage désertique caractérisés par des regs, des hamadas et des<br />

ensembles dunaires. »<br />

Un grand nombre d’espèces végétales sont utilisées à<br />

travers le monde pour lutter contre l’ensablement et elles<br />

montrent toutes, à un certain moment, leur limite. Ainsi,<br />

pratiquement 6 millions d’hectares sont transformés en<br />

désert chaque année et plus de 20 millions d’hectares<br />

montrent une productivité nulle, suite à ce phénomène<br />

(Oldache, 1988). En Algérie, cet ensablement est bien<br />

présent, et le rempart naturel que constitue la steppe n’arrive<br />

pas à lutter efficacement contre l’avancé des dunes<br />

car la steppe dégradée ne permet plus le piégeage du sable.<br />

Dans cette zone de transition toutes les biocénoses suivent<br />

une dynamique régressive (Pouget, 1979).<br />

La lutte contre l’ensablement dans une commune de<br />

l’ouest de l’Algérie a été menée en utilisant du Tamarix<br />

mis en défens pendant plusieurs années. Cette espèce permet<br />

la stabilisation des dunes et l’installation d’une strate<br />

herbacée à l’origine d’une remontée biologique perceptible.<br />

L’étude porte sur le suivi du comportement de cette<br />

espèce dans son utilisation comme moyen de lutte biologique<br />

contre l’ensablement qui menace cette localité et<br />

les terres agricoles environnantes. Des plantations d’âges<br />

différents ont été étudiées afin de comparer la dynamique<br />

de la végétation dans le temps et dans l’espace.<br />

Caractéristiques stationnelles<br />

de la zone étudiée<br />

Les paramètres climatiques retenus sont ceux du poste<br />

météorologique de Mécheria (Algérie occidentale), situé<br />

à une quinzaine de kilomètres de la zone d’étude, dans la<br />

même zone géographique et dans l’étage aride, d’où sa<br />

représentativité à défaut de poste météorologique in situ.<br />

La carte pluviométrique dressée par l’Office national de<br />

la Météorologie, sur la base des données de la période<br />

1950-1980, situe la localité de Ain Benkhélil et ses environs<br />

dans la tranche pluviométrique des 200 mm de pluviométrie<br />

moyenne annuelle. Le régime pluviométrique<br />

est du type APHE, favorable à une activité végétative malgré<br />

la longueur de la période de sécheresse qui s’étale<br />

d’avril à octobre.<br />

Les températures moyennes minimales du mois le plus<br />

froid m sont de 1,7 °C et celles du mois le plus chaud M<br />

de 35,1 °C ; le quotient pluviothermique d’Emberger est<br />

de 20, ce qui permet de classer la zone d’étude dans l’étage<br />

bioclimatique aride inférieur frais selon la classification<br />

d’Alcaraz (1962). D’après les fluctuations des paramètres<br />

climatiques interannuels, le climat varie de l’aride au semiaride<br />

inférieur (Mederbal, 1992 ; Djelouli, 1981).<br />

Les sols de la zone sont peu évolués, d’apport d’origine<br />

éolienne, et caractérisés généralement par trois horizons<br />

difficilement discernables. Le taux de sable est toujours<br />

supérieur à 78 % (78 à 95 %), celui de l’argile est<br />

inférieur à 12 % (4 à 12 %) et les limons ne dépassent<br />

jamais 10 % (1 à 10 %). Le taux de matière organique<br />

varie essentiellement dans les horizons superficiels et diminue<br />

selon la profondeur de 1,5 à 0,5 %. Le pH est compris<br />

entre 7,2 et 8,5 (Bachir, 1993).<br />

Contexte floristique et phytoécologique<br />

La carte de végétation réalisée par le Centre de<br />

recherche en biologie terrestre de l’Université d’Alger, en<br />

1977, signale la présence des espèces dominantes suivantes<br />

: Lygeum spartum L., Retama retamWebb., Aristida<br />

pungens L., Stipa tortilis Desf., Artemisia herba-alba Asso.,<br />

Arthrophytum scoparium Schrank., Thymelaea microphylla<br />

Coss. En fait, les trois dernières espèces ne sont que rarement<br />

présentes, alors que d’autres espèces caractéristiques<br />

de biotopes salés et d’espèces psammophiles de terrains<br />

humides telles que Tamarix gallica L., Phragmites communis<br />

Trin., Stipa lagascae R. et S., Helianthemum lippii<br />

Ball sont constamment présentes (Djelouli, 1981).<br />

La strate herbacée, dominante, est représentée par des<br />

espèces appartenant à une dizaine de familles (tableau 1) où<br />

dominent les Graminées avec 10 genres et 11 espèces, les<br />

Composées avec 7 genres et 7 espèces et les Chénopodiacées<br />

avec 6 genres et 6 espèces, les Légumineuses avec 5 genres<br />

et 6 espèces et les Crucifères avec 3 genres et 5 espèces.<br />

D’autres familles sont également présentes, mais ne sont pas<br />

signalées, car très faiblement représentées, avec un nombre<br />

de genres et d’espèces généralement inférieur à 2, comme<br />

les Convolvulacées, les Liliacées, les Géraniacées, les<br />

Malvacées, les Plombaginacées, les Polygonacées, les<br />

Saxifragacées, les Aizoacées, les Capparidées, les<br />

Frankéniacées, les Inulées et les Onogracées.<br />

Les Graminées, les Composées, les Chénopodiacées<br />

et les Légumineuses jouent donc un rôle déterminant dans<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


CONSÉQUENCES PHYTOÉCOLOGIQUES DE LA RESTAURATION D’UN ÉCOSYSTÈME DUNAIRE FIXE AVEC LE TAMARIS…<br />

Éléments géographiques 1990 1994 1996 Témoin<br />

la végétation de la zone. Les espèces végétales les plus<br />

constantes se répartissent comme suit (tableau 1) : 1 à 2<br />

plurirégionales ; 8 à 9 méditerranéennes ; 5 à 7 saharosindiennes<br />

; et 3 à 8 endémiques. En comparant à la placette<br />

témoin, une évolution est nettement perceptible, car<br />

cette dernière ne regroupe que 3 plurirégionales, 4 méditerranéennes,<br />

2 saharo-sindiennes et 3 endémiques.<br />

Les travaux de Djebaili (1978, 1990) mettent en évidence<br />

les groupements à Stipa tenacissima L., Artemisia<br />

herba-alba Asso. et Lygeum spartum L. qui sont rattachés<br />

au Lygeo-Stipetalia (Braun-Blanquet et De Bolos, 1957)<br />

enrichi de nombreuses thérophytes et psammophytes<br />

endémiques du Maghreb. Les autres groupements à<br />

Lygeum spartum et Stipa tenacissima et ceux à Lygeum spartum<br />

et Thymelaea microphylla Coss. peuvent également<br />

être rattachés au Lygeo-Stipetalia, tout comme les groupements<br />

vivaces steppiques maghrébins de l’aride (Rivas-<br />

Martinez, 1977).<br />

MÉTHODOLOGIE<br />

Espèces % Espèces % Espèces % Espèces %<br />

Plurirégionales 1 5 2 9 2 9 3 25<br />

Méditerranéennes 9 43 9 39 8 36 4 33<br />

Saharo-sindiennes 7 33 5 28 6 27 2 17<br />

Endémiques 3 14 7 31 9 38 3 25<br />

Tableau 1. Évolution du spectre des éléments biogéographiques dans la zone de mise en défens entre 1990 et 1996.<br />

Des placettes d’observation et de mesures, avec des<br />

unités d’observations de 20 pieds en moyenne (Dagnélie,<br />

1957), soit une surface de 10 m 2,ont été retenues. Douze<br />

placettes ont été installées dans le périmètre protégé à raison<br />

de 4 par catégorie d’âge (2 ans, 5 ans et 10 ans). Une<br />

placette n’ayant bénéficié ni de plantation ni de mise en<br />

défens a été choisie à l’intérieur de la zone pour servir<br />

d’espace témoin.<br />

Dans un but de cerner la dynamique phytoécologique,<br />

deux relevés phytoécologiques ont été effectués par placette<br />

et par catégorie d’âge, soit 24 relevés ; seule la notion<br />

d’absence et de présence a été retenue pour évaluer l’évolution<br />

quantitative des espèces en comparaison avec la<br />

parcelle témoin.<br />

Les travaux de fixation de dunes entrepris dans la zone<br />

d’Ain Benkhélil (Naama) durant la campagne de planta-<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 93-99<br />

tion 1988/89 se sont faits selon les données techniques<br />

suivantes : la densité de plantation est de 1 600 plants par<br />

hectare ; les plants de Tamarix gallica L. sont issus de boutures<br />

élevées en pépinière pendant 9 mois en moyenne et<br />

ont une hauteur oscillant entre 0,60 et 1 m et un diamètre<br />

de l’ordre de 1,5 à 3 cm au collet. La plantation se fait en<br />

ouvrant un trou de 30 cm de profondeur suivie d’un tassement<br />

du sol et d’un arrosage de 20 litres d’eau. La superficie<br />

plantée est mise en défens pendant neuf années<br />

(1988/89 à 1996/97).<br />

Une évaluation permanente et annuelle de la diversité<br />

végétale a été faite au mois de mai des années 1990, 1994,<br />

1996 et les résultats obtenus comparés à ceux de la placette<br />

témoin qui représente une certaine stabilité phytoécologique<br />

de la zone étudiée soumise à une utilisation.<br />

RÉSULTATS ET DISCUSSIONS<br />

Dynamique de la flore spontanée<br />

L’exploitation des résultats du comportement des<br />

espèces dans les parcelles protégées et plantées de Tamarix<br />

permet d’apprécier le rôle de chaque espèce dans la dynamique<br />

évolutive de l’écosystème dunaire fixé. La fréquence,<br />

la phytomasse et la productivité en unités fourragères<br />

sont des paramètres déterminants pour classer les<br />

espèces et comprendre l’importance écologique de chacune<br />

d’elles dans les conditions précitées.<br />

La phytomasse totale obtenue dans l’écosystème<br />

dunaire fixé par le Tamarix, estimée par Bergheul en 1990<br />

avec une pluviométrie moyenne de l’ordre de 269 mm<br />

étalés sur 61 jours, est très proche de nos résultats. Elle<br />

est évaluée à 533 kilogrammes de matière sèche par hectare<br />

(kg MS/ha) pour la zone témoin. Deux ans après une<br />

mise en défens, la phytomasse atteint 1 032 kg MS/ha, et<br />

après une autre période de deux ans de protection, elle<br />

diminue à 384 kg MS/ha. La pluviométrie moyenne de<br />

95


96<br />

◆ KHÉLOUFI BENABDEL, HALIMA MOHAM<strong>MED</strong>I, NOREDDINE LETEURECH-BELAROUCI<br />

1990 1994 1996<br />

Types biologiques Espèces % Espèces % Espèces %<br />

Thérophytes 4 19,04 7 30,43 9 35<br />

Chaméphytes 7 33,03 7 30,43 6 23<br />

Hémicryptophytes 6 28,57 5 21,47 5 19<br />

Phanérophytes 1 4,76 1 4,33 1 3<br />

Géophytes – – 1 4,33 2 6<br />

Tableau 2. Évolution du spectre biologique dans la zone de mise en défens entre 1990 et 1996.<br />

1990 1994 1996<br />

Types biologiques Espèces % Espèces % Espèces %<br />

Thérophytes 4 40 3 22 6 0<br />

Chaméphytes 2 20 4 28 3 25<br />

Hémicryptophytes 2 20 3 22 1 8<br />

Cryptophytes 1 10 2 14 1 8<br />

Phanérophytes 1 10 1 7 1 8<br />

Géophytes – – 1 7 – –<br />

Tableau 3. Évolution du spectre biologique de la placette témoin entre 1990 et 1996.<br />

Espèces Zone témoin Mise en défens Mise en défens Mise en défens<br />

(1990) (1994) (1996)<br />

Freq. C. Freq. C. Freq. C. Freq. C.<br />

Cynodon dactylon Pers. 67 4 – – 19 2 20 2<br />

Aristida pungens L. 13 2 – – 19 2 18 2<br />

Peganum harmala L. 10 2 13 2 – – 8 1<br />

Malva parviflora L. 6 2 – – – – – –<br />

Echinops spinosus 5 2 21 2 10 2 8 2<br />

Helianthemum lippii Ball. 19 2 20 2 20 2 22 3<br />

Zygophyllum album 23 2 5 2 7 2 9 1<br />

Scirpus holoschoenus L. 6 2 8 2 17 2 21 3<br />

Phragmites communis Adams. 48 3 28 3 9 2 6 1<br />

Tamarix gallica L. 5 2 24 2 46 4 52 4<br />

Onopordon acaulis 4 2 10 2 4 2 5 1<br />

Schismus barbatus P.B. 4 2 3 2 6 2 8 2<br />

Stipa lagascae L. – – 37 3 23 2 19 2<br />

Iris sisyrinchium L. – – 5 2 – – 4 1<br />

Coronilla juncea L. – – 2 2 – – 2 1<br />

Thymelaea microphylla Tourn. – – 6 2 3 2 – –<br />

Lygeum spartum L. – – 5 2 31 3 43 4<br />

Muricaria prostata Desv. – – 2 2 – – 2 1<br />

Malcolmia africana R. Br. – – 9 2 – – – –<br />

Astragalus gombo L. – – 4 2 10 2 18 3<br />

Launaea resedifolia L. – – 8 2 6 2 4 1<br />

Phytomasse kg MS/ha 336,51 – 474,27 – 976,30 – 1 124,60 –<br />

Equivalent UF/ha 35 – 45 – 105 – 115 –<br />

Recouvrement global 20 % – 10 % – 40 % – 50 % –<br />

Freq. : fréquence ; C. : coefficient d’abondance dominance ; UF : unité fourragère ; kg MS/ha : kilogramme de matière sèche par hectare<br />

Tableau 4. Dynamique phytoécologique exprimée par la fréquence et l’abondance des principales espèces végétales de la zone d’étude.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


CONSÉQUENCES PHYTOÉCOLOGIQUES DE LA RESTAURATION D’UN ÉCOSYSTÈME DUNAIRE FIXE AVEC LE TAMARIS…<br />

cette période est légèrement plus faible que celle de l’année<br />

1990, seulement 256 mm sur 57 jours.<br />

Cette chute nette de phytomasse peut avoir trois justifications<br />

possibles : l’impact de la différence de précipitations<br />

; l’écosystème a atteint son maximum de production<br />

et régresse ; les parcelles ont été utilisées illicitement<br />

la nuit par les troupeaux.<br />

Les résultats obtenus après une mise en défens et une<br />

plantation en Tamarix gallica L. sont assez remarquables<br />

par rapport à l’état de dégradation des autres espaces de<br />

la région. Les formations végétales issues de cette mise en<br />

valeur sont un taillis de Tamarix sur une strate buissonnante<br />

permanente dont le taux de recouvrement moyen<br />

est de 10 à 20 % et une pelouse à espèces annuelles à l’origine<br />

de la phytomasse relativement importante quantifiée.<br />

Par espèces, la phytomasse est très fluctuante (tableau 4),<br />

avec une prédominance d’Aristida pungens, Lygeum spartum,<br />

Helianthemum lippii, Peganum harmala, Stipa lagascae,<br />

Astragalus gombo, Launaea residifolia, Helianthemum hirtum,<br />

Thymelaea microphylla et Ononis natrix. Respectivement entre<br />

1990 et 1994, puis 1994 et 1996, la phytomasse a connu des<br />

augmentations de 474, 976 et 1124 kilogrammes de matière<br />

sèche par hectare comparée à la placette témoin qui n’offre<br />

qu’une biomasse fluctuant entre 456 et 588 kg MS/ha, soit<br />

une moyenne de référence plus ou moins stable de 522 kg<br />

MS/ha.<br />

La placette témoin montre une abondance des espèces<br />

éphémères (tableau 3), où les espèces indicatrices de<br />

dégradation par les pressions anthropozoogènes dominent.<br />

Dès la première mise en défens apparaissent des<br />

espèces vivaces recherchées par les troupeaux pour leur<br />

apétabilité telles que : Aristida pungens, Lygeum spartum et<br />

Stipa lagascae. Dans la steppe à alfa, Le Houérou (1968)<br />

signale que ces trois espèces peuvent se réinstaller facilement<br />

sous l’effet d’une amélioration des conditions climatiques,<br />

édaphiques (notamment par une diminution<br />

du piétinement) et de la conduite des troupeaux.<br />

Ce sont surtout les espèces saharo-sindiennes qui<br />

ont connu un accroissement de leur rôle alors que celui<br />

des pluri-régionales a diminué. Les méditerranéennes,<br />

quant à elles, sont restées relativement constantes. Le<br />

nombre de genres présents connaît, par rapport à la<br />

placette témoin, une progression appréciable en 1990,<br />

1994 et 1996, avec respectivement des gains de 6, 11<br />

et 12, alors que pour les espèces les gains sont de 9, 11<br />

et 15. Le tableau 2 donne un aperçu sur la dynamique<br />

et la fluctuation du nombre de genres et d’espèces par<br />

élément biogéographique. La mise en défens joue un<br />

rôle déterminant dans cette évolution de la composition<br />

spécifique.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 93-99<br />

Les résultats obtenus en nombre d’espèces et en pourcentage<br />

par type biologique récapitulé dans le tableau 3<br />

laisse apparaître une certaine instabilité des végétaux temporaires<br />

(thérophytes et géophytes) alors que les végétaux<br />

ligneux (phanérophytes et chaméphytes) et les hémicryptophytes<br />

présentent une certaine stabilité.<br />

Comparaison entre relevés phytoécologiques<br />

L’évolution phytoécologique a pu être appréciée après<br />

une mise en défens et introduction de Tamarix gallica.<br />

Les espèces qui ont le plus progressé sont Echinops spinosus,<br />

Tamarix gallica, Schismus barbatus, Stipa lagascae,<br />

Lygeum spartum et Astragalus gombo. Par contre, il faut<br />

noter une diminution ou disparition d’espèces telles que<br />

Malva parviflora, Zygophyllum album, Peganum harmala<br />

et Phragmites communis. La composition floristique<br />

moyenne est proche de celle des steppes de la région,<br />

telle que définie par Djebaili (1982), avec une dominance<br />

de Chaméphytes et d’Hémicryptophytes. La mise<br />

en défens, ou l’utilisation équilibrée de cet écosystème,<br />

conduit donc généralement à une remontée biologique<br />

appréciable.<br />

Le recouvrement global a connu une diminution lors<br />

des deux premières années de mise en défens, essentiellement<br />

à cause des perturbations causées par les travaux<br />

de préparation du sol et de plantation, le taux de couvert<br />

est alors passé de 20 à 10 %. Après six ans de protection,<br />

le recouvrement est passé de 10 à 40 %, atteignant même<br />

50 % après huit ans de mise en défens.<br />

La phytomasse aérienne a également connu une augmentation<br />

récapitulée comme suit :<br />

Année Pluviométrie Phytomasse en Unités<br />

en mm kg MS/ha fourragères<br />

1988 247 336.51 35<br />

1990 269 474.27 45<br />

1994 256 976.30 105<br />

1996 273 1124.68 115<br />

Les résultats obtenus sont assez proches de ceux de<br />

Bergheul (1990) et Bachir (1993) qui ont travaillé dans<br />

la même zone, mais avec des objectifs et approches différents.<br />

Les légères différences enregistrées se localisent<br />

essentiellement dans le nombre d’espèces recensées, le<br />

coefficient d’abondance-dominance et la biomasse avec<br />

une marge d’erreur oscillant entre 15 et 20 %.<br />

La phytomasse, après une diminution pendant les<br />

premières années, augmente considérablement entre 2 et<br />

4 ans de mise en défens, puis se stabilise au-delà de 5 ans<br />

97


98<br />

◆ KHÉLOUFI BENABDEL, HALIMA MOHAM<strong>MED</strong>I, NOREDDINE LETEURECH-BELAROUCI<br />

Paramètres quantifiés Phytomasse sèche en kg MS/ha<br />

Espèces principales Pt 1990 1994 1996<br />

Aristida pungens L. 205 286 248 373<br />

Peganum harmala L. 57 38 45 71<br />

Helianthemum lippii Ball. 157 75 121 137<br />

Tamarix gallica L. 103 131 221 295<br />

Stipa lagascae L. – 83 28 19<br />

Lygeum spartum L. – 14 304 366<br />

Thymelaea microphylla Tourn. – 25 12 11<br />

Ononis natrix L. – 5 5 7<br />

Astragalus gombo L. – 11 8 18<br />

Launaea resedifolia L. – 6 13 14<br />

Helianthemum hirtum Ball. – – 16 13<br />

Total 522 474 976 1 124<br />

Pt : parcelle témoin (check plot).<br />

qui semble être le stade d’équilibre (tableau 5). La phytomasse<br />

des steppes à Lygeum spartum est estimée à<br />

350 kg MS/ha et par an, selon le Centre de recherche<br />

en biologie terrestre (CRBT, 1978) ; à 500 kg MS/ha,<br />

selon Aidoud (1983). En matière de productivité, Le<br />

Houérou (1971) l’évalue, pour les steppes arides où la<br />

pluviométrie oscille entre 200 et 300 mm/an, entre 80<br />

et 150 kg MS/ha.<br />

Comportement du Tamarix gallica<br />

Tableau 5. Évolution de la biomasse végétale chez les principales espèces de la zone d’étude.<br />

En plus de la restauration de l’écosystème, constatée<br />

par la bonne “remontée biologique”, l’espèce introduite<br />

procure une biomasse ligneuse intéressante. Dans ces<br />

zones, les besoins en bois comme source d’énergie domestique<br />

sont très élevés et ont été estimés, dans des conditions<br />

écologiques similaires, dans le sud du massif forestier<br />

de Télagh (Oranie-Algérie), à plus de 0,6 stères par<br />

famille et par jour, soit 60 kilogrammes (Benabdeli, 1983).<br />

Les paramètres dendrométriques permettant d’évaluer<br />

l’évolution de la production ligneuse sont le nombre de<br />

brins par cépée, la hauteur moyenne, le diamètre et la surface<br />

terrière. La quantification de ces paramètres aux différents<br />

âges laisse apparaître une augmentation du nombre<br />

de brins par hectare (densité), un accroissement annuel<br />

moyen de la hauteur de 56 cm. Le diamètre à 1,30 m<br />

connaît lui aussi un accroissement moyen annuel de<br />

0,57 cm, soit une augmentation annuelle moyenne de la<br />

surface terrière de 0,00709 m2 (voir les résultats détaillés<br />

dans le tableau 6).<br />

Avec une plantation à une densité moyenne de 1600<br />

plants par hectare, le Tamarix gallica permet une production<br />

de matière ligneuse de l’ordre de 0,28 m3/ha et<br />

par an, soit un accroissement moyen annuel moyen de<br />

0,31 m3/ha. Selon les besoins moyens par famille en bois<br />

domestique, la productivité d’un hectare de Tamarix gallica<br />

permet de faire face aux besoins de deux familles.<br />

Paramètres 1990 1994 1996 Moyenne<br />

Nombre de cépées par hectare 1350 1325 1295 1323<br />

Nombre de brins par hectare 2525 6375 7025 5308<br />

Hauteur moyenne en cm 3,61 4,62 5,21 4,48<br />

Diamètre moyen en cm 3,99 4,63 5,17 4,59<br />

Surface terrière par brin en cm 2 0,0017 0,0027 0,0034 0,0026<br />

Surface terrière par hectare en m 2 0,0425 0,0675 0,0814 0,0638<br />

Tableau 6. Evolution des paramètres dendrométriques du Tamarix gallica dans la zone mise en défens.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


CONSÉQUENCES PHYTOÉCOLOGIQUES DE LA RESTAURATION D’UN ÉCOSYSTÈME DUNAIRE FIXE AVEC LE TAMARIS…<br />

CONCLUSION<br />

De toutes les espèces végétales utilisées pour la lutte<br />

contre l’ensablement, le Tamarix gallica, dans les conditions<br />

écologiques particulières de Mécheria (Ain Ben-khélil)<br />

semble donner des résultats de stabilisation des dunes<br />

appréciables. La colonisation de l’écosystème dunaire par<br />

une strate herbacée et arbustive est réelle ; ainsi il y a une<br />

bonne “remontée biologique”, garante d’une stabilisation<br />

du sol particulaire et de l’installation d’une dynamique<br />

phytoécologique progressive. Les résultats obtenus ne sont<br />

pas tous dus à l’introduction du Tamarix gallica, puisque<br />

la mise en défens joue aussi un rôle important.<br />

La combinaison entre la mise en défens et la plantation<br />

d’espèces ligneuses comme le Tamarix gallica permet<br />

d’aboutir à une restauration appréciable. Ce ligneux à<br />

croissance rapide permet de créer un microclimat favorable<br />

à l’installation d’une strate herbacée, jouant un rôle<br />

important dans la stabilisation des dunes et l’évolution de<br />

l’écosystème. Les riverains respectent la mise en défens,<br />

car elle leur procure des unités fourragères et du bois de<br />

feu en contrepartie de leur participation à la sauvegarde<br />

de cet espace.<br />

Bibliographie<br />

AIDOUD A., 1983. Contribution à l’étude des écosystèmes steppiques<br />

du sud oranais : phytomasse, productivité énergétique,<br />

productivité primaire et application pastorale.Thèse doct. 3e cycle, USTHB, Alger, 254 p.<br />

ALCARAZ C., 1969. Étude géobotanique du pin d’Alep dans le<br />

Tell oranais. Thèse de spécialité, Univ. Montpellier, 183 p.<br />

BACHIR D., 1993. Contribution à l’étude du comportement de<br />

Tamarix articulata dans la lutte contre l’ensablement dans la<br />

région de Ain Benkhélil (Naama, Algérie). Mémoire Ing.<br />

Forestier. Univ. Tlemcen, 121 p.<br />

BENABDELI K., 1983. Mise au point d’une méthodologie d’appréciation<br />

de la pression anthropozoogène sur la végétation<br />

dans la région de Télagh (Algérie).Thèse doct. 3e cycle, Univ.<br />

Aix-Marseille III, 183 p.<br />

BERGHEUL N., 1990. Contribution à l’étude des dunes fixées au<br />

niveau des mises en défens d’âges différents de Ain Benkhélil<br />

(Naama). DES, USTHB, Alger, 63 p.<br />

BRAUN-BLANQUET J. & BOLÒS DE O., 1957. Les groupements<br />

végétaux du bassin moyen de l’Èbre et leur dynamique. Ann.<br />

Estac., 5:1-266.<br />

CRBT, 1978. Rapport phytoécologique et pastoral sur les hautes<br />

plaines steppiques de la wilaya de Saida (Algérie). Polycopié,<br />

Alger, 256 p.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002, p. 93-99<br />

CHAUMONT M. & PAQUIN C., 1971. Notice explicative de la carte<br />

pluviométrique de l’Algérie + carte annexe. Bull. Soc. Hist.<br />

Nat. Afr. Nord, Alger, 24 p.<br />

DAGNÉLIE P., 1957. Recherche sur la productivité des hêtraies<br />

de l’Ardenne en relation avec les types phytosociologiques et<br />

les facteurs écologiques. Bull. Inst. Agro. Gembloux, 25:44-<br />

94.<br />

DJEBAILI S., 1978. Recherches phytoécologiques et phytosociologiques<br />

sur la végétation des hautes plaines steppiques et de<br />

l’Atlas Saharien.Thèse doct. Univ. Languedoc, Montpellier,<br />

229 p.<br />

DJEBAILI S., 1990. Syntaxonomie des groupements préforestiers<br />

et steppiques de l’Algérie aride. Ecol. Medit., 16:231-244.<br />

DJELOULI Y., 1981. Étude climatique et bioclimatique des hauts<br />

plateaux oranais (wilaya de Saida). Comportement des<br />

espèces vis-à-vis d’éléments du climat. Thèse doct. 3 e cycle,<br />

USTHB, Alger, 178 p.<br />

GODRON M., 1976. Les échantillonnages phytoécologiques. Note<br />

n° 8 CNRS/CEPE, Montpellier, 23 p.<br />

GODRON M., 1984. Abrégé d’écologie de la végétation terrestre.<br />

Masson, Paris, 197 p.<br />

LE HOUÉROU H.N., 1967. Problèmes et potentialités des zones<br />

arides de l’Afrique du Nord. Options Médit., 26:17-35.<br />

LE HOUÉROU H.-N., 1968. La désertification du Sahara septentrional<br />

et des steppes limitrophes (Libye, Tunisie, Algérie).<br />

Prog. Bio. Intern. : 33 p.<br />

LE HOUÉROU H.-N., 1969. La végétation de la Tunisie steppique<br />

(avec références aux végétations analogues d’Algérie, de Libye<br />

et du Maroc). Ann. Inst. Nation. Agro., 42:624 p.<br />

LE HOUÉROU H.-N., 1971. Les bases écologiques de la production<br />

pastorale et fourragères en Algérie. FAO, Rome : 60 p.<br />

LE HOUÉROU H.-N., 1979. La désertification des régions arides.<br />

La Recherche, 99:21-27.<br />

<strong>MED</strong>ERBAL K., 1992. Compréhension des mécanismes de transformation<br />

du tapis végétal : approches phytoécologique par<br />

télédétection aérospatiale et dendroécologie de Pinus halepensis<br />

Mill. dans l’ouest algérien.Thèse doctorat d’État, Univ.<br />

Aix-Marseille III : 229 p.<br />

OLDACHE E., 1988. Contribution à l’étude de la fixation des<br />

dunes dans la région d’El Mesrane (Djelfa) et Bousaada<br />

(M’sila). Thèse de magister, INA, Alger : 98 p.<br />

POUGET M., 1979. Les relations sol-végétation dans les steppes<br />

sud-algéroises (Algérie). Thèse doctorat d’État, Univ. Aix-<br />

Marseille III, 555 p.<br />

RIVAS-MARTINEZ S., 1977. Sur la syntaxonomie des pelouses<br />

thérophytiques de l’Europe occidentale. In : La végétation<br />

des pelouses sèches à thérophytes. Coll.Phytosociol.,6:55-71.<br />

99


A new colony of critically endangered Northern<br />

Bald Ibises has been discovered in an Al Badia<br />

(desertic steppe) area of central Syria<br />

A new colony of critically endangered Northern Bald Ibises<br />

(Geronticus eremita) has been discovered in an Al Badia (desertic<br />

steppe) area of central Syria.The small colony contained three pairs<br />

which were discovered incubating eggs, and a seventh adult.This is<br />

the first evidence of the continued breeding of Northern Bald Ibises<br />

in the Middle East since a colony at Birecek in Turkey became extinct<br />

in 1989.The new birds were found in spring 2002 by a team carrying<br />

out wildlife surveys on behalf of the Syrian Government’s<br />

Ministry of Agriculture and Agrarian Reform (MAAR).<br />

Survey Team leader, Associate Professional Officer, Wildlife<br />

Expert, Gianluca Serra said, « Discovering this bird was like finding<br />

the Arabian Phoenix regenerated from the ashes.The survey work<br />

through remote and rough terrain was some of the most exciting and<br />

challenging fieldwork we had ever experienced.Throughout it all, my<br />

Syrian colleagues from MAAR, Ghazy Al-Qaim and Mahmoud<br />

Abdallah, were optimistic that Northern Bald Ibises still existed in<br />

the Al Badia or desertic steppe of central Syria because we had received<br />

reports of their presence from Bedouin nomads and local hunters.<br />

» Dr Michael Rands, Director of BirdLife International, said<br />

« This fascinating species, once common throughout much of the<br />

Middle East and southern Europe, is now on the brink of global<br />

extinction, despite much conservation effort in Morocco and Turkey.<br />

This fantastic discovery gives new hope that the Northern Bald Ibis<br />

can be saved, and the BirdLife Partnership will do all it can to assist<br />

the Syrian authorities to conserve this amazing threatened species for<br />

future generations to enjoy. »<br />

The Northern Bald Ibis was formerly widespread across the<br />

<strong>Mediterranea</strong>n region, but has suffered a long-term decline and<br />

now has an extremely small population. The reasons behind<br />

decline include human persecution, loss of steppe or unintensive<br />

agricultural areas, pesticide poisoning, human disturbance<br />

and dam construction. Until this recent discovery, the total world<br />

population was put at 220 individual birds, confined to two colonies<br />

in north-west Morocco.The Northern Bald Ibis is classified<br />

as Critically Endangered according to the World Conservation<br />

Union (IUCN) Red List criteria. This means it faces an extremely<br />

high risk of extinction in the wild in the immediate future.<br />

The discovery was made at the end of an extensive two-year<br />

programme of wildlife surveys aimed at producing a biodiversity<br />

inventory for Al Talila Reserve, the first Syrian protected area.<br />

Contact: Birdlife International, Wellbrook Court,<br />

Girton Road, Cambridge, CB3 0NA, United Kingdom.<br />

(http://www.birdlife.net/news).<br />

ecologia mediterranea, tome 28, fascicule 2, p. 101-105<br />

Faits de conservation en Méditerranée<br />

<strong>Mediterranea</strong>n Conservation News<br />

The Iberian lynx needs networks of large reserves<br />

The Iberian lynx, an endemic species of southern Iberia, is<br />

considered by the IUCN as the most endangered felid in the<br />

world. Recent research made at Estación Biológica de Doñana<br />

(Sevilla, Spain) has focused on the factors responsible of lynx<br />

decline (Rodríguez & Delibes, 2002-2003). In 1953 myxomatosis,<br />

a viral disease, quickly spread across the Iberian peninsula<br />

resulting in a strong drop of rabbit numbers (virtually the only<br />

lynx prey species) which lasted for decades. An analysis of the<br />

subsequent range contraction during a 35-year period shows that<br />

only lynx populations inhabiting areas larger than 500 km 2 were<br />

able to persist. The extinction of these small lynx populations<br />

has been attributed to demographic stochasticity related with<br />

metapopulation disequilibrium.The refill of gaps, i.e. vacant territories,<br />

may have accounted for reduced fragmentation of larger<br />

populations.There was also strong indication that migration<br />

from neighbouring sources was essential for the survival of vulnerable<br />

small populations.<br />

Simple regression models could be used to choose where<br />

conservation efforts should be located in order to optimise the<br />

persistence of lynx populations. The quality of lynx habitat in<br />

terms of scrubland structure and rabbit availability was probably<br />

higher around 1950 than it is nowadays. Keeping human<br />

pressure low, and assuming that habitat quality was restored to<br />

the former levels, several nature reserves larger than 50,000 ha<br />

should be established and saturated with populations of about<br />

100 adults to guarantee lynx survival in the short to medium<br />

term (30 to 50 years). Reserves of this size can be labelled “large”<br />

for European standards. Moreover, reserves should be arranged<br />

in networks where inter-population distance should be minimised<br />

within a maximum of 30 km, a figure close to the upper threshold<br />

for lynx dispersal distance. Intervening habitats between<br />

reserves may be not as good as within reserves but their structure<br />

should facilitate migration between lynx populations, and<br />

sources of disperser mortality must be removed.<br />

For more information read:<br />

— RODRÍGUEZ A. & DELIBES M., 2002.<br />

Internal structure and patterns of contraction in the geographic<br />

range of the Iberian lynx. Ecography, 25: 314-328.<br />

— RODRÍGUEZ A. & DELIBES M., 2003.<br />

Population fragmentation and extinction in the Iberian lynx.<br />

Biological Conservation, 109: 321-331.<br />

ALEJANDRO RODRÍGUEZ<br />

Dept. of Applied Biology, Estación Biológica de Doñana, CSIC,<br />

Avda. María Luisa s/n, 41013 Sevilla, Spain.<br />

email address: alrodri@ebd.csic.es<br />

101


102<br />

◆ FAITS DE CONSERVATION EN MÉDITERRANÉE / <strong>MED</strong>ITERRANEAN CONSERVATION NEWS<br />

Quelles sont les chances de maintien d’un ligneux<br />

endémique relictuel ? Le cas de la bourdaine bétique<br />

(Frangula alnus subsp. baetica) en Andalousie<br />

La bourdaine bétique, Frangula alnus subsp. baetica (Rhamnaceae),<br />

est un ligneux endémique, éminemment résiduel et vulnérable,<br />

présent dans quelques populations du sud de l’Espagne<br />

et du nord-ouest du Maroc (région biogéographique béticorifaine).<br />

Au Maroc, il se localise dans quelques ravins du revers<br />

septentrional du Tidighin (Rif), entre 1 400 et 1 750 m d’altitude,<br />

en bordure de sources et suintements permanents parmi<br />

la cédraie, en compagnie de Betula pubescens subsp.celtiberica.En<br />

Andalousie, une vingtaine de populations de bourdaine existent<br />

au fond de certains vallons humides de basse altitude (entre 125<br />

et 700 m) des provinces de Huelva, Cadiz et Malaga, en particulier<br />

dans les barrancos de la Sierra del Aljibe.<br />

Une récente étude conduite par Hampe et Arroyo (2002) ont<br />

étudié en détail les processus de recrutement et de régénération de<br />

diverses populations andalouses de cette relique tertiaire. Les individus<br />

de ces petits peuplements de bourdaine semblent souffrir de<br />

dépression de consanguinité et d’une baisse de fitness ; la reproduction<br />

femelle s’avère anormalement basse puisque seulement<br />

1,4 % d’ovules sont produits. Les arbres ne sont reproducteurs<br />

qu’au bout d’un vingtaine d’années (DBH = 6 cm). Au sein d’un<br />

secteur de ripisylve long de 600 m, Hampe et Arroyo (2002) ont<br />

calculé que 82 % des adultes produisaient des drupes, soit un total<br />

d’environ 748 000 graines viables. Ces fruits charnus sont dispersées<br />

par endozoochorie par divers oiseaux sédentaires (Erithacus<br />

rubecula, Sylvia atricapilla et Turdus merula) ;cependant ce processus<br />

est relativement limité et inefficace comparé à d’autres ligneux<br />

méditerranéens, puisque seulement 36 % des fruits matures sont<br />

consommés par les oiseaux. En fait, la dispersion secondaire lors<br />

des épisodes de crues hivernales joue un rôle majeur en entraînant<br />

les graines à plus ou moins longue distance des peuplements, le<br />

plus souvent dans des banquettes d’alluvions récentes. La prédation<br />

post-dispersion des fruits s’effectue presque exclusivement par<br />

le mulot qui engendre une perte très importante de graines, de<br />

l’ordre de 73 +/- 14 %. L’augmentation depuis les années 1970 du<br />

gibier introduit pour la chasse (Capreolus capreolus,Cervus elaphus)<br />

accroît aussi notablement les pressions d’herbivorie. De plus, les<br />

débroussaillements réalisés en périphérie de ripisylve favorisent les<br />

phénomènes érosifs et les événements de rapide montée des eaux,<br />

d’où un recrutement plus réduit de la bourdaine. Mais les conditions<br />

de sécheresse des banquettes alluvionnaires représentent le<br />

facteur de mortalité des plantules le plus important.Au total, aucune<br />

des 1 124 plantules recensées en juin 1997 n’a survécu en septembre<br />

1999. Dans ce contexte, le nombre de juvéniles et de jeunes<br />

arbres est réduit, et le pic de distribution culmine pour les classes<br />

d’âge comprises entre 25 et 30 ans (DBH = 8-10 cm), c’est-à-dire<br />

celles correspondant aux années favorables sur le plan pluviométrique.<br />

Sous l’influence des changements globaux, l’exacerbation<br />

prévue des périodes de sécheresse et des épisodes extrêmes et soudains<br />

d’inondation pourrait encore plus menacer la régénération<br />

des peuplements et même entraîner l’extinction de cette relique tertiaire<br />

en situation déjà fort précaire.<br />

Pour en savoir plus lire :<br />

— HAMPE A. & ARROYO J., 2002. Recruitment and regeneration<br />

in populations of an endangered South Iberian Tertiary relict tree.<br />

Biological Conservation, 107 : 263-271.<br />

FRÉDÉRIC MÉDAIL<br />

Institut méditerranéen d’écologie et de paléoécologie (IMEP, CNRS-<br />

UMR 6116), Université d’Aix-Marseille III,<br />

Faculté des sciences et techniques de Saint-Jérôme.<br />

Case 461. F-13397 Marseille Cedex 20.<br />

A <strong>Mediterranea</strong>n pest attacks a relict pine forest<br />

under climatic warming (Sierra Nevada, Southern Spain)<br />

The National Park of Sierra Nevada (southeastern Spain) is<br />

inhabited by a highly diversified flora, including species that are<br />

common in northern latitudes and remained relict in the southern<br />

high mountains due to the glaciations.Thus, Sierra Nevada harbours<br />

the southernmost populations of the Scots pine (Pinus sylvestris<br />

L.), a pine species widely distributed in Europe. These<br />

populations belong to the endangered, endemic subspecies nevadensis,<br />

which forms the treeline at about 2 100 m a.s.l. in these<br />

mountains. This altitude protect these trees from the attack of a<br />

severe defoliating <strong>Mediterranea</strong>n pest, the pine processionary<br />

moth Thaumetopoea pityocampa, strongly limited by low winter<br />

temperatures. However, as a consequence of climatic change, the<br />

pine processionary caterpillar has increasingly attacked these<br />

populations of Scots pine in recent years.The decade 1991-2000<br />

was the warmest one of the 20th century. Coinciding with several<br />

consecutive winters of mild temperatures and strong rainfall,<br />

a severe outbreak of the pine processionary caterpillar spread to<br />

higher altitudes than usual. Scots pine populations in Sierra<br />

Nevada, both natural and afforested, repeatedly underwent severe<br />

defoliations during the winters of 1997 and 1998.The detrimental<br />

effects of the caterpillar defoliation on the pine varied.The growth<br />

rate was severely reduced, seed production was unusually low and<br />

seeds were of smaller size in defoliated pines. Complete and repeated<br />

defoliation for more than two consecutive winters exhausted<br />

energy reserves, inflicting some pine deaths. In a healthy population<br />

of pines with a good regeneration capacity, the negative<br />

impact of these outbreaks would be negligible. However, the Scots<br />

pine in Sierra Nevada has limited regeneration capacity, owing<br />

to high pre- and postdispersal seed predation, scant success in<br />

seedling establishment due mostly to summer drought, and<br />

delayed growth in saplings and juveniles because of the high herbivore<br />

pressure. If winter temperatures continue to rise as a consequence<br />

of the climatic warming, uphill invasion of the processionary<br />

will be faster than pine displacement. Outbreaks will be<br />

more frequent and more severe, crippling the regeneration of<br />

these forests.The wide altitudinal gradient of Sierra Nevada offers<br />

the possibility of studying the detrimental effects of climate change<br />

in these and other plant species, and of experimenting with possible<br />

measures to mitigate or avoid such negative impact.<br />

For more informations read:<br />

— HÓDAR, J.A., CASTRO, J.& ZAMORA, R.2003.<br />

Pine processionary caterpillar Thaumetopoea pityocampa<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


as a new threat for relict <strong>Mediterranea</strong>n Scots pine forests under<br />

climatic warming. Biological Conservation, in press.<br />

JOSÉ A. HÓDAR<br />

Grupo de Ecología Terrestre, Departamento de Biología Animal y Ecología,<br />

Facultad de Ciencias, Universidad de Granada, E-18071 Granada, Spain<br />

email address: jhodar@ugr.es – Website : www.ugr.es/~rnm220<br />

État de conservation du cyprès de Duprez<br />

(Cupressus dupreziana) au Sahara central ;<br />

une espèce en voie d’extinction ?<br />

Le cyprès de Duprez (Cupressus dupreziana A. Camus) est<br />

un conifère rarissime et remarquable par sa présence au centre<br />

du Sahara, à 60 km du Tropique du Cancer. Vivant sous climat<br />

hyper-aride où la moyenne annuelle des précipitations est estimée<br />

à 30 mm, il peut dépasser 20 m de hauteur et atteindre 12 m<br />

de circonférence. Dès sa découverte en 1924 sur les hauteurs du<br />

plateau du Tassili n’Ajjer (sud-est algérien), il a été déclaré espèce<br />

en voie de disparition. Les vieux arbres aperçus dans les lits<br />

d’oueds laissaient croire qu’il s’agissait d’individus reliques d’une<br />

forêt constituées sous un climat humide et survivant actuellement<br />

grâce à la présence de nappes phréatiques. En 1978, l’Union<br />

internationale pour la conservation de la nature (UICN) l’a inscrit<br />

sur la liste rouge des espèces les plus menacées d’extinction<br />

au monde.<br />

Durant 47 ans, le recensement des arbres vivants s’est fait au<br />

gré de missions diverses : géologiques, militaires, archéologiques,<br />

zoologiques et botaniques. La plus fructueuse fut celle d’un forestier<br />

algérien Saïd Grim (inédit), qui entre 1971 et 1972 a décrit<br />

230 cyprès vivants et 153 morts sur pied répartis sur une bande<br />

de 120 km de long et 6 km en moyenne de large, longeant la bordure<br />

sud-ouest du plateau du Tassili. L’exhaustivité de ce dénombrement,<br />

la précision des descriptions et des localisations des<br />

arbres, a permis d’estimer l’évolution démographique de ce peuplement<br />

grâce à un recensement précis effectué entre 1997 et<br />

2001.<br />

Vingt arbres parmi les 230 recensés par Grim ont été retrouvés<br />

morts, alors que 23 autres sujets ont été découverts, ce qui<br />

le fixe l’ensemble de la population mondiale connue à 233 cyprès<br />

vivants. Parmi les arbres nouvellement inventoriés, deux très<br />

jeunes individus sont remarquables car ils se situent sur l’itinéraire<br />

de la mission Grim. Le premier, de 3,4 m de hauteur et<br />

32 cm de circonférence est situé à quelques mètres d’un vieux<br />

sujet décrit par Grim à l’oued In Gharohane (24° 27’ N,<br />

9° 47’ E). Le second, dans l’oued In Ellidj (24° 57’ N, 9° 21’ E),<br />

atteint 37 cm de circonférence et 3 m de hauteur. Il se situe à<br />

proximité de 6 arbres décrits en 1971. Ces deux sujets correspondent<br />

vraisemblablement de régénérations naturelles par<br />

graines. Des observations de plantules et semis près des vieux<br />

arbres furent signalées dans les années 1940 et 1960. Cependant,<br />

l’idée d’une inéluctable disparition de ces plantules n’ayant pas<br />

assez développé leur système racinaire qui leur permettrait de<br />

surmonter les sécheresses successives demeurait incontestable.<br />

Or, l’existence de ces jeunes arbres indique pourtant la persistance<br />

d’une régénération, certes rare, mais qui peut échapper en<br />

ecologia mediterranea, tome 28, fascicule 2, p. 101-105<br />

FAITS DE CONSERVATION EN MÉDITERRANÉE / <strong>MED</strong>ITERRANEAN CONSERVATION NEWS ◆<br />

partie à la fois à la sécheresse et aux perturbations humaines et<br />

pastorales. Cependant, la perte de près de 8 % en 29 ans du<br />

nombre total des cyprès vivants est considérable et d’autant plus<br />

inquiétante qu’elle est dûe à l’exploitation humaine. L’émondage<br />

pastoral, les prélèvements de bois pour le chauffage et la cuisson<br />

par les nomades et les touristes n’épargnent pas les cyprès<br />

vivants. Dans une zone où le bois est rare et où il n’existe aucune<br />

mesure effective de protection pour cette espèce, le cyprès de<br />

Duprez demeure, malgré sa capacité naturelle de régénération<br />

in situ, en danger d’extinction au Tassili n’Ajjer.<br />

Pour en savoir plus, lire :<br />

— ABDOUN F. & BEDDIAF M., 2002. Cupressus dupreziana<br />

A. Camus : répartition, dépérissement et régénération au Tassili<br />

n’Ajjer, Sahara central. Comptes Rendus Biologies, 325 : 617-627<br />

FATIHA ABDOUN<br />

Institut méditerranéen d’écologie et de paléoécologie<br />

(IMEP, CNRS-UMR 6116), Université d’Aix-Marseille III,<br />

Faculté des sciences et techniques de Saint-Jérôme.<br />

Case 461. F-13397 Marseille Cedex 20.<br />

email : f.abdoun@caramail.com<br />

From pets to pest: the endemic <strong>Mediterranea</strong>n<br />

shearwater threatened by feral cats<br />

The <strong>Mediterranea</strong>n shearwater Puffinus yelkouan is, along<br />

with the Balearic shearwater Puffinus mauretanicus, one of the<br />

only two pelagic seabirds strictly restricted to the confined<br />

<strong>Mediterranea</strong>n basin. Until recently, these two species were considered<br />

as distinct subspecies of Yelkouan shearwater P. yelkouan<br />

but the two taxa are now distinguished by their specific morphological,<br />

ecological, behavioural and genetic features. The<br />

<strong>Mediterranea</strong>n shearwater is a medium-size species that breeds<br />

exclusively on islands or islets within rocky habitats. The distribution<br />

and the size of the colonies within the <strong>Mediterranea</strong>n<br />

basin remain largely unknown. The total world breeding population<br />

is considered to amount to only 15,000 to 20,000 pairs<br />

but this number is suspected of being largely overestimated.<br />

However, virtually nothing is known about what threatens the<br />

<strong>Mediterranea</strong>n shearwater either at sea or ashore. A recent study<br />

evidenced the possible impact of feral cat Felis catus on this species,<br />

a phenomenon that until now, has generally been considered<br />

limited. The analysis of 650 feral cat scats from Port-Cros<br />

island (south-east France), showed that this introduced predator<br />

preys regularly on <strong>Mediterranea</strong>n shearwaters (remains were<br />

found in 2,8 % of scats analysed). Even though it was impossible<br />

to quantify the total amount of shearwaters killed yearly by<br />

feral cats on the small island of Port-Cros, this study, the first on<br />

this topic, undoubtedly evidenced that the feral cat predation on<br />

the <strong>Mediterranea</strong>n shearwater may be more important than previously<br />

expected. All the recorded predation cases corresponded<br />

to adult shearwaters, killed during the first half of their breeding<br />

cycle, probably when their vocal activity was intense (display<br />

activities). There was no predation case corresponding to chick<br />

or fledging young. It is quite surprising that <strong>Mediterranea</strong>n shearwater<br />

remains were only found in scats collected far from the<br />

103


104<br />

◆ FAITS DE CONSERVATION EN MÉDITERRANÉE / <strong>MED</strong>ITERRANEAN CONSERVATION NEWS<br />

colonies and not in those collected close to the colonies. This<br />

showed that feral cats from the inland of Port-Cros can travel<br />

long-distances to specifically prospect <strong>Mediterranea</strong>n shearwaters<br />

colonies.This preliminary study stressed the urgent need for<br />

further investigation on the conservation status of the <strong>Mediterranea</strong>n<br />

shearwater and on what threatens this species, in order<br />

to set up accurate conservation action plans.<br />

Contact : Eric Vidal, Institut méditerranéen d’écologie<br />

et de paléoécologie (IMEP, CNRS-UMR 6116),<br />

Faculté des sciences et techniques de Saint-Jérôme,<br />

case 461, 13397 Marseille cedex 20, France.<br />

email : eric.vidal@univ.u-3mrs.fr<br />

The effect of nature conservation on flowering<br />

of Paeonia mascula in Israel<br />

Mount Meiron Nature Reserve is the largest <strong>Mediterranea</strong>n<br />

reserve in Israel, dominated by Quercus calliprinos an evergreen<br />

sclerophylous tree. Paeonia mascula (L.) Mill. reaches here its southern<br />

border of distribution and is a main visitor’s attraction. In<br />

1965 it was declared a nature reserve and drastic decrease in<br />

woodcutting and goat grazing affected the landscape. The area<br />

turned from a mosaic of dens and open maquis into homogeneous<br />

thick evergreen oak woodland. Under such shaded conditions,<br />

Paeonia plants survive but do not flower which is a threat<br />

to the local population.The aim was to explore the optimal light<br />

regime and to examine the effect of wood thinning on Paeonia<br />

growth and flowering. In autumn 1997, 15 experimental plots<br />

(5 x 5 m) were established. All trees were removed from 5 plots.<br />

Relative PAR radiation and Paeonia plant-height were measured<br />

and the plants and flowers were counted in May for four years.<br />

Flowering plants were 47 cm high while non-flowering plants<br />

were only 27 cm high. Relative radiation of 30-40 % in plots seems<br />

to be optimal for Paeonia growth and flowering.We conclude that<br />

the Paeonia population is endangered by shading, and active management<br />

actions to open the dense maquis are recommended.<br />

For more informations read:<br />

— NE’EMAN G. 2003. To be or not to be – the effect of nature<br />

conservation management on flowering of Paeonia mascula (L.)<br />

Miller in Israel. Biological Conservation, 109: 103-109.<br />

GIDI NE’EMAN<br />

Department of Biology, Faculty of Science and Science Education,<br />

University of Haifa at Oranim, Tivon 36006, Israel.<br />

email address: gneeman@research.haifa.ac.il<br />

Dehesa’s lizards are worth to be conserved too<br />

In the iberian peninsula, the structure and configuration of<br />

holm oak (Quercus ilex) dehesa woodland is the result of human<br />

intervention, through clearance, thinning and pruning of the once<br />

natural vegetation.Traditional management techniques have created<br />

a patchwork of sectors that differ in their structure and composition<br />

of understorey vegetation.To understand how the fauna<br />

responds to these gradients, we can use models that relate the<br />

abundance of a species to variables describing the habitat structure.<br />

Previous studies have shown that dehesas maintain more<br />

diverse bird communities than adjacent treeless pasture, areas of<br />

cultivation and even <strong>Mediterranea</strong>n forest. Thus, dehesa management<br />

seemed to favour biodiversity. However, there has been<br />

very little research on other representative animal groups.There<br />

is a danger that focusing conservation activities on a few charismatic<br />

species may overlook important features for other species.Thus,<br />

one should take account of the ecology of a wide spectrum<br />

of species to develop maximum conservation of the biological<br />

diversity.<br />

Lizards are one of the more prominent groups of animals in<br />

<strong>Mediterranea</strong>n climates, whith many endemic species that reach<br />

high densities, and that are important keys in the trophic chains.<br />

Researchers from the Department of Evolutionary Ecology of<br />

the Madrid Natural History Museum (CSIC) have analysed<br />

whether different management techniques of a dehesa area in<br />

Central Spain affect the populations of several species of lizards.<br />

The results showed that lizards did not use microhabitats at random.<br />

They preferentially used forested areas with scrub and<br />

island rocky outcrops, while they avoided open herbaceous areas.<br />

Furthermore, characteristics of a dehesa patch influenced both<br />

population densities and species richness of lizards. Lizards were<br />

more abundant when the understorey bushy vegetation increased.<br />

In contrast, grasslands or cereal fields were scarcely occupied<br />

even if holm oak trees were present, because low shrubs<br />

were scarce here. Shrub cover is primarily important in providing<br />

cover against predators, but it is insufficient for lizards in<br />

most dehesa plots because of regular clearance management.<br />

Therefore, some of the traditional management practices of dehesas<br />

may negatively affect lizard populations. These conclusions<br />

could have wider implications for the design of wildlife reserves,<br />

which has been previously based on the umbrella species concept<br />

(e.g., “conserving imperial eagles habitat will provide conservation<br />

for all other species”). An integrated design and management<br />

of protected areas would require a detailed analysis of habitat<br />

requirements of other representative animal groups.<br />

For more informations read:<br />

— Martín, J. & López, P. 2002. The effect of <strong>Mediterranea</strong>n<br />

dehesa management on lizard distribution and conservation.<br />

Biological Conservation, 108: 213-219.<br />

JOSÉ MARTÍN<br />

Departamento de Ecología Evolutiva<br />

Museo Nacional de Ciencias Naturales<br />

José Gutiérrez Abascal, 2. E-28006 Madrid. Spain.<br />

email address: Jose.Martin@mncn.csic.es<br />

When gulls drive beetle communities on islands<br />

Island communities are very sensitive to disturbances, and<br />

pressure exerted by colonial seabird in particular. Seabirds are<br />

known to be a major ecological factor driving both species assemblage<br />

patterns and community dynamics of plants on islands.<br />

However, very little was known about the response of insect<br />

assemblages to seabirds, although the important ecological and<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


functional role of insects within all ecosystems. A recent study<br />

aimed to assess the response of beetle communities to the<br />

influence of a Yellow-legged gull Larus cachinnans colony on a<br />

small <strong>Mediterranea</strong>n island.This study was conducted on Bagaud<br />

island, which is a fully protected reserve within Port-Cros<br />

National Park (south-east France), where the Yellow-legged gulls<br />

are the most densely clumped. Beetle communities were sampled<br />

three times during spring and autumn 2001 to increase the<br />

probability of sampling uncommon and seasonal species and to<br />

meet the successive stages of gull behaviour and abundance<br />

during their breeding cycle. A total of 255 beetle specimens from<br />

28 species and 15 families was collected. Surprisingly, species<br />

richness and abundance of beetles were not significantly affected<br />

by gulls. However, ordination of beetle assemblages showed<br />

that there was a marked change in species composition along the<br />

gradient of gull influence. Gull colony also affected distribution<br />

of feeding-types of beetles, as the Tenebrionidae family in general,<br />

and Bioplanes meridionalis in particular, seemed to draw strong<br />

benefit from such a disturbance. Tenebrionidae species, as they<br />

are polyphagous, markedly increased at the expense of phytophagous<br />

species that decreased. Collected species were likely to<br />

be found in other gull-free environments and were not strictly<br />

endemic to <strong>Mediterranea</strong>n islands. Some of them presented<br />

however some particular conservation interests: Asida dejeani,<br />

Otiorhynchus vitellus, Thorictus grandicollis, Dichillus minutus,<br />

Trachyphloeus laticollis and Ocypus fortunatarum that were stated<br />

either uncommon or restricted to <strong>Mediterranea</strong>n regions.<br />

For more informations read:<br />

— ORGEAS J., VIDAL E. & PONEL P. 2003. Colonial seabirds change<br />

beetle assemblages on a <strong>Mediterranea</strong>n Island. Ecoscience, in press.<br />

JÉRÔME ORGEAS<br />

Institut méditerranéen d’écologie et de paléoécologie<br />

(IMEP, CNRS-UMR 6116), Faculté des sciences et techniques<br />

de Saint-Jérôme, case 461, 13397 Marseille cedex 20, France.<br />

email address: jerome.orgeas@univ.u-3mrs.fr<br />

Immature eagles recruited as breeder alert us about future<br />

changes in population trends: the case of Bonelli’s Eagle in<br />

Andalusia.<br />

The most common techniques used to evaluate the status<br />

and trend of bird populations generally detect changes in population<br />

tendency once they have already occurred, therefore lac-<br />

ecologia mediterranea, tome 28, fascicule 2, p. 101-105<br />

FAITS DE CONSERVATION EN MÉDITERRANÉE / <strong>MED</strong>ITERRANEAN CONSERVATION NEWS ◆<br />

king of predictive capability. The Bonelli’s Eagle Hieraaetus fasciatus<br />

is an endangered bird of prey that has suffered a rapid<br />

decline in its whole distribution range.The checking of the breeding<br />

population is easy by the fact that breeding individuals<br />

occupy year-round territories, and because by adult and immature<br />

eagles have different plumage which allow to age them accurately.<br />

The Department of Applied Biology of the Estación Biológica<br />

de Doñana disposed of a more than 20 years data set on the evolution<br />

of the breeding territories of this eagle in Andalusia, as<br />

well as the breeding performance of pairs.We analysed the trend<br />

of the Andalusia population during a long-term period and discover<br />

that, in spite of the number of occupied territories have<br />

remained stable, both productivity (measured as number of<br />

young fledged per occupied territory) and age at fist breeding<br />

(measured as number of immature eagles on breeding pairs)<br />

decreased in the last decade.<br />

In our case, it was possible to attribute the detected decrease<br />

in the age at first breeding to a decrease in adult survival rate. A<br />

lack of adults within the population allows the immature eagles<br />

to occupy empty breeding places, resulting in a decrease on the<br />

breeding performance at the population level, probably due to<br />

the lower breeding experience of young. An increase in immature<br />

within the breeding population therefore represents an earlywarning<br />

signal of a change in two extremely important demographic<br />

parameters, as productivity and adult survival rate. For<br />

this reason, monitoring the age at first breeding can be considered<br />

as a new tool in conservation biology, useful to detect changes<br />

in population trends of long-lived species with deferred maturation<br />

age. Therefore, it will allow researcher to plan conservation<br />

strategies well before an endangered population could reach<br />

a low threshold of productivity and/or number of adult breeding<br />

individuals, ensuring more effectiveness in population management<br />

and lower costs of intervention.<br />

For more informations read:<br />

BALBONTÍN J., PENTERIANI,V. & FERRER,M.2003.Variations in the age<br />

of mates as an early-warning signal of changes in population trends?:<br />

the case of Bonelli’s eagle in Andalusia. Biological Conservation, 109:<br />

417-423.<br />

J. BALBONTÍN,V.PENTERIANI ET M. FERRER<br />

Estación Biológica de Doñana (CSIC), Department of Applied Biology.<br />

Avda. Maria Luisa s/n. Sevilla 41013. Spain<br />

e-mail: balbonja@ebd.csic.es<br />

105


106<br />

◆ FAITS DE CONSERVATION EN MÉDITERRANÉE / <strong>MED</strong>ITERRANEAN CONSERVATION NEWS<br />

ASINARA DECLARATION on OIL POLLUTION AND CONSERVATION OF BIODIVERSITY<br />

INTRODUCTION<br />

An international conference was held at Porto Torres (Sardinia) from 17th to 20th October 2002 on Oil Pollution and Conservation of Biodiversity.<br />

80 participants coming from 17 different countries, among whom the world experts on oil polluted wildlife, were attending this conference.They<br />

decided to formulate the Asinara Declaration pleading for various priority actions to be taken to safeguard marine fauna and flora and their habitats<br />

from oil pollution.<br />

Isola dell’Asinara (Sardinia), 20th October 2002<br />

the Delegates of the Conference on Oil Pollution and Conservation of Biodiversity, organized at Porto Torres, Sardinia (17-20 October<br />

2002) by the Asinara National Park, IFAW (International Fund for Animal Welfare) and <strong>MED</strong>MARAVIS (<strong>Mediterranea</strong>n Marine Bird<br />

Association):<br />

Item 1. Control of Tanker Traffic in Biological Sensitive Areas<br />

While considering the marine ecosystem as vital for the survival of mankind and harbouring unique biota evolved in diverse<br />

geo-morphological, climatic and oceanographical conditions,<br />

Realizing that numerous species of marine biota, particularly seabirds with a high oil vulnerability index, have population levels<br />

that are critically low and are directly threatened by oil pollution,<br />

Considering that marine wildlife is a natural resource not only to be exploited for economic interests but is also of great scientific,<br />

ecological, educational, aesthetic and recreational importance for the general public,<br />

Being aware that many marine, island and coastal habitats are directly threatened by oil pollution,<br />

Noting that these habitats are essential for the survival of such biota,<br />

Urge<br />

— the regional and national governments, with the co-operation of the competent international institutions to establish marine<br />

Nature Reserves and Zones of Tanker Traffic Control,<br />

— the International Maritime Organization (I.M.O.) to encourage third parties to respect the designated Zones of Tanker<br />

Traffic Control.<br />

Item 2. Separation Lanes for Shipping<br />

Considering that coastal wildlife and its habitats are extremely vulnerable in narrow straits,<br />

Taking note that traffic of oil tankers and of vessels carrying hazardous cargo is often very dense in the narrow straits of the<br />

<strong>Mediterranea</strong>n and Black Seas,<br />

Invite all national governments concerned,<br />

— to install radar controlled separation lanes for shipping in narrow straits of high biological interest,<br />

— to have in place in these straits an oil contingency plan that deals explicitly with oiled wildlife and provides for effective<br />

wildlife response.<br />

Item 3. Stricter Implementation of Maritime Regulations<br />

Realising that the current maritime regulations as promulgated by the International Maritime Organization (IMO) are not fully<br />

implemented and do not guarantee an environmentally correct management of vessels,<br />

Being aware that certain countries, members of the Paris Memorandum of Understanding on Port State Control, do not meet<br />

the target of inspecting 25 % of vessels visiting their ports,<br />

Observing that many vessels do not use port reception facilities for processing oil residues despite the fact that some ports<br />

have installed adequate disposal facilities in order to prevent oil pollution from bilge, cargo and ballast,<br />

Considering that the European Directive 2000/59 regarding port reception facilities for ship-generated waste will enter into<br />

force on 28th December 2002,<br />

Recommend to all national governments and concerned international institutions to:<br />

— adopt and fully implement the existing agreements under MARPOL 73/78 and its Annexes, the Regional Memoranda of<br />

Understanding, the Barcelona and Bucharest Conventions and relevant Protocols,<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


ecologia mediterranea, tome 28, fascicule 2<br />

FAITS DE CONSERVATION EN MÉDITERRANÉE / <strong>MED</strong>ITERRANEAN CONSERVATION NEWS ◆<br />

— to provide adequate shore reception facilities for vessels’ generated wastes,<br />

— to refuse all uncertified vessels in ports unless they are sailing to a repair yard for careening or dismantling,<br />

— to improve the inspection of vessels’ structure,<br />

— to develop a data base of vessels to be consulted by charterers.<br />

Item 4. Prosecution<br />

Being fully aware of the severe impact on seabirds and their habitats by the successive groundings and break-ups of vessels at<br />

sea as well as by illegal discharges and dumping at sea of fuel oils and their residues,<br />

Considering that the owner(s) of the transported cargo are directly responsible for vessel control during port operations and<br />

on the high seas, and that IMO is responsible for the safety of maritime traffic,<br />

Realizing the importance of cumulative costs by oil pollution for marine wildlife, fisheries, tourism and health,<br />

Urge governmental and juridical authorities:<br />

— to vigorously enforce MARPOL regulations on oil discharges from vessels,<br />

— to review the safety regulations of maritime traffic in straits of ecological importance,<br />

— to enforce the practice of the “polluter pays” principle over and above the ceiling<br />

level of the oil company’s insurance cover,<br />

— to earmark mitigation payments for oiled wildlife to the benefit of its rehabilitation and the protection of its habitats.<br />

Item 5. Research<br />

Observing that not enough data exist on the distribution at sea of marine wildlife during its life cycle,<br />

Realizing that marine wildlife can be extremely vulnerable to mineral oil pollution at sea,<br />

Considering the high risk of contamination for marine wildlife from oil pollution, toxic chemicals and all vessels’ generated<br />

waste,<br />

Invite the national governments, the competent international institutions as well as the private oil industry and shipping companies:<br />

— to finance annual surveys and monitoring programmes of marine wildlife and its distribution at sea,<br />

— to sponsor the publication of regional guides and atlasses related to the distribution of coastal and marine biota,<br />

— to launch research programmes on population dynamics and age structure of mortality of oil-prone species especially seabirds,<br />

— to define the best methods for the effective treatment and cleaning of ballast waters.<br />

Item 6. Effective Oiled Wildlife Response<br />

Being aware that the immediate rehabilitation and release of rare oiled seabird species are important for maintaining their populations<br />

at a level which guarantees their long term survival,<br />

Knowing that biological “hot-spots” of coastal and marine biota are generally found in isolated areas as well as in wildlife reserves<br />

that do not have the necessary facilities and equipment to protect marine wildlife,<br />

Urge local, regional and national authorities, as well as the specialized nongovernmental<br />

organizations:<br />

— to establish wildlife contingency plans for areas of major importance for vulnerable<br />

marine wildlife in the <strong>Mediterranea</strong>n and Black Sea regions,<br />

— to identify and acquire the necessary equipment for their implementation,<br />

— to organize training courses in the treatment of oiled wildlife and rehabilitation of their habitats,<br />

— to disseminate technical information on the biology, conservation and rehabilitation of marine and coastal biota threatened<br />

by oil pollution.<br />

Cala Reale, Isola dell’Asinara, 20th October 2002<br />

Agreement validated unanimously by the conference participants and signed by the three conference organizers:<br />

Barbara CALLAHAN Eugenio COSSU Xaver MONBAILLIU<br />

International Fund President Asinara Secretary-General<br />

for Animal Welfare National Park <strong>MED</strong>MARAVIS<br />

Anchorage, USA Porto Torres, Italy Saint-Maximin, France<br />

107


Écologie : de l’écosystème à la biosphère<br />

Christian Lévêque<br />

Dunod, Masson Sciences, Paris : 502 p. (2001)<br />

Le livre de Christian Lévêque est une contribution originale<br />

et moderne dans la série des ouvrages consacrés à l’écologie générale.<br />

Originale surtout dans son écriture ; en effet, plutôt que de<br />

fournir une somme de connaissances dans les divers domaines<br />

de l’écologie avec de nombreux exemples à l’appui, il propose<br />

une vision bien plus synthétique où sont privilégiées les grandes<br />

théories, leurs controverses et leurs domaines d’applications. Y<br />

sont présentés les paradoxes et paradigmes les plus modernes en<br />

écologie, le tout dans une écriture accessible aux non-spécialistes,<br />

avec de multiples encarts pour les lecteurs désirant approfondir<br />

tel ou tel domaine. Enfin, l’auteur a délibérément choisi<br />

comme objet d’étude “l’écosystème” et il montre combien cette<br />

unité de travail en écologie est pertinente. Enfin, en rapport avec<br />

la spécialité de recherches de l’auteur, les exemples seront surtout<br />

pris dans le domaine de l’écologie des eaux courantes, ce<br />

qui est relativement rare en comparaison des autres ouvrages qui<br />

s’intéressent surtout aux écosystèmes continentaux terrestres.<br />

Ce sont surtout les premiers chapitres (1-4) qui font l’originalité<br />

du travail de Christian Lévêque ; ils sont consacrés aux<br />

natures de l’écologie et à la construction de la pensée scientifique<br />

en écologie des écosystèmes. La lecture de ces chapitres<br />

est à conseiller à tout écologue, qu’il soit étudiant ou chercheur<br />

confirmé, car ils permettent de bien identifier l’écologie en tant<br />

que science à travers l’originalité de ses approches, méthodes,<br />

finalités fondamentales et appliquées. C’est donc une “saine”<br />

lecture qui permet au lecteur de se resituer par rapport aux<br />

contextes fondamentaux, courants de pensée et applications en<br />

écologie.<br />

Les autres chapitres de la deuxième partie, consacrée à la<br />

structure et l’organisation des écosystèmes, traitent plus classiquement<br />

de l’application des principes de physique en écologie<br />

(5) et de l’écologie factorielle (6), pour se terminer par trois chapitres<br />

consacrés aux problèmes d’échelles, de hiérarchies et de<br />

l’hétérogénéité spatiale (7-9). Inutile de signaler l’importance de<br />

ces notions vis-à-vis des recherches menées en écologie. Elles<br />

sont particulièrement bien présentées ici et intègrent les apports<br />

les plus récents des recherches en cours.<br />

La troisième partie est consacrée au fonctionnement des écosystèmes,<br />

avec, notamment, deux chapitres présentant deux problématiques<br />

très actuelles, d’une part la dynamique des peuplements<br />

et l’équilibre de la nature (10), d’autre part les relations<br />

ecologia mediterranea, tome 28, fascicule 2, p. 109-111<br />

entre diversité biologique et fonctionnement des écosystèmes<br />

(12). Les problèmes de disparition d’espèces, d’espèces envahissantes<br />

et d’espèces clés sont particulièrement bien présentés.<br />

Ces chapitres sont toujours illustrés par des exemples pris dans<br />

les domaines de l’écologie aquatique.<br />

Enfin, la quatrième partie concerne l’écologie globale, avec,<br />

notamment, un apport important sur les interactions entre dynamique<br />

des écosystèmes et changements globaux. Ici aussi,<br />

Christian Lévêque apporte une contribution très importante à<br />

un débat d’actualité.<br />

Nous recommandons donc la lecture de l’ouvrage de<br />

Christian Lévêque aux étudiants et aux plus confirmés d’entre<br />

nous, car il se lit presque comme un roman et permet de resituer<br />

le lecteur et ses problématiques de recherches dans les débats<br />

les plus modernes en écologie fondamentale.<br />

THIERRY DUTOIT<br />

Institut méditerranéen d'’écologie et de paléoécologie<br />

(IMEP, CNRS-UMR 6116), Université d’Aix-Marseille I,<br />

Faculté des sciences et techniques de Saint-Jérôme.<br />

Case 461. F-13397 Marseille Cedex 20.<br />

Endemic plants of Greece<br />

The Peloponnese<br />

Analyses d’ouvrages<br />

Kit Tan avec la collaboration de G. Iatrou,<br />

dessins de B. Johnsen<br />

Gad Forlag, Copenhagen, 480 p., 111 pl. couleur (2001)<br />

email: sekr@gads-forlag.dk<br />

Après des décennies de travaux floristiques et phytoécologiques<br />

qui ont fait nettement progresser la connaissance du patrimoine<br />

floristique grec, il était temps de synthétiser tout ce corpus<br />

de connaissances dans des flores modernes. Une équipe internationale,<br />

essentiellement composée de Scandi-naves et de Grecs<br />

s’est attelée à l’ouvrage, et, après la Mountain Flora of Greece<br />

(1986-1991), est sorti en 1997 le premier volume de l’excellente<br />

Flora Hellenica. Dans cette lignée, une série de trois volumes est<br />

plus spécialement destinée à détailler la flore endémique de<br />

Grèce, et le présent ouvrage Endemic plant of Greece. The<br />

Peloponnese est le premier publié. D’après les estimations fournies,<br />

le Péloponnèse comporte ca. 2 960 espèces et sous-espèces<br />

dont 355 sont des endémiques greques et 4,6 % des endémiques<br />

strictes de la dition. Cette haute richesse en endémiques s’avère<br />

109


110<br />

◆ ANALYSES D’OUVRAGES<br />

particulièrement bien mise en exergue par cet ouvrage qui allie<br />

une qualité iconographique exceptionnelle à une rigueur et une<br />

documentation scientifique de tout premier plan.<br />

L’introduction botanique, riche de 65 pages, constitue une<br />

précieuse entrée en matière sur le plan géographique, géologique<br />

et phytoécologique. Des données numériques et biogéographiques<br />

sur l’ensemble de la flore endémique grecque, notamment<br />

celle des montagnes, permettent de bien resituer l’originalité<br />

de cette flore au sein du monde méditerranéen. Puis, sont<br />

détaillées les diverses régions d’intérêt botanique, avec mention<br />

des principales caractéristiques abiotiques, et bien entendu, des<br />

particularités floristiques de chacune d’entre elles. Ceci constitue<br />

un précieux guide pour le botaniste qui voudra parcourir les<br />

monts du Chelmos, Killini, Parnon, Taygète…<br />

L’ensemble des endémiques du Péloponnèse est ensuite passé<br />

en revue, avec, curieusement, l’inclusion d’une dizaine de plantes<br />

non-endémiques, ce qui peut induire une certaine confusion,<br />

même si elles sont signalées dans le texte. Chaque taxon fait l’objet<br />

d’une description botanique précise suivie de données sur<br />

l’habitat et l’écologie, la distribution, le nombre chromosomique,<br />

et d’un court texte précisant les aspects taxonomiques et systématiques<br />

; la carte de répartition de l’espèce est précise et issue<br />

de la base de données mise au point pour le projet Flora Hellenica.<br />

Il faut encore insister sur la qualité des planches iconographiques<br />

couleur qui font ressortir les moindres détails morphologiques<br />

diagnostiques et qui accompagnent la quasi-totalité des espèces<br />

décrites. Trois taxons nouveaux pour la science, dont une nouvelle<br />

espèce (Carum depressum Hartvig & Kit Tan), sont décrits,<br />

et sept combinaisons nouvelles sont proposées ; parmi ces dernières,<br />

ériger Cyclamen peloponnesiacum en tant qu’espèce semble<br />

très discutable, et le récent article de Debussche et Thompson<br />

(in Bot. J. Linn. Soc., 2002) montre qu’il vaut mieux continuer à<br />

considérer ce taxon comme une sous-espèce de C. repandum car<br />

les différences morphologiques et génétiques ne sont pas suffisamment<br />

tranchées. Mais, par ailleurs, cet ouvrage permet de<br />

mettre au jour des données chorologiques sur des taxons présumés<br />

disparus comme Adonis cyllenea Boiss. qui se rencontre en<br />

fait dans plusieurs localités du nord de la péninsule !<br />

En attendant la parution des volumes consacrés à la Grèce<br />

continentale et aux îles égéennes, il faut souhaiter qu’un travail<br />

de cette qualité fasse des émules et que d’autres territoires du<br />

bassin méditerranéen puissent bénéficier de cette approche qui<br />

contribue à faire mieux connaître et donc – espérons-le – protéger<br />

ce patrimoine naturel méditerranéen si riche mais si menacé.<br />

FRÉDÉRIC MÉDAIL<br />

Institut méditerranéen d'’écologie et de paléoécologie<br />

(IMEP, CNRS-UMR 6116), Université d’Aix-Marseille III,<br />

Faculté des sciences et techniques de Saint-Jérôme.<br />

Case 461. F-13397 Marseille Cedex 20.<br />

Écologie des forêts naturelles d’Europe.<br />

Biodiversité, sylvigenèse,<br />

valeur patrimoniale des forêts primaires<br />

Annick Schnitzler-Lenoble<br />

Editions TEC & DOC, Paris : 271 p. (2002)<br />

Pour tous les écologues qui auraient pu penser qu’il n’existe<br />

plus de forêts primaires en Europe, l’ouvrage d’Annick<br />

Schnitzler-Lenoble sera une révélation. En effet, il existe encore<br />

en Europe, et surtout en Europe de l’Est, des massifs forestiers<br />

dont le fonctionnement écologique peut être considéré comme<br />

proche de celui d’écosystèmes forestiers naturels ou primaires.<br />

C’est dans les domaines de la sylvigenèse, du fonctionnement<br />

(rôle des perturbations, des stress) et de l’analyse hiérarchique<br />

de l’écosystème forestier (ch. 3 et 4) que l’ouvrage d’Annick<br />

Schnitzler-Lenoble apporte le plus d’informations originales en<br />

liaison avec les problématiques de recherches de l’auteur. Les<br />

autres chapitres sont beaucoup plus conventionnels et présentent<br />

la place des forêts au niveau de la biogéographie européenne<br />

(ch. 1) et les composants de l’écosystème forestier (ch. 2) illustrés<br />

avec des planches photos présentant un joli bestiaire de la<br />

faune des forêts européennes. Ces deux derniers chapitres sont<br />

purement descriptifs et rapidement traités. Il est regrettable qu’ils<br />

ne soient pas précédés d’un chapitre ou d’une introduction définissant<br />

le plus clairement possible les écosystèmes forestiers avant<br />

de présenter leur place en Europe. Cette définition n’arrivant<br />

qu’au début du sixième chapitre.<br />

Les chapitres consacrés à la biodiversité des forêts d’Europe<br />

depuis la fin du Tertiaire (ch. 5 et 6) sont également en déséquilibre<br />

avec les chapitres consacrés au fonctionnement de l’écosystème<br />

forestier. En effet, ici, plus qu’à la présentation de résultats<br />

scientifiques originaux, nous assistons à la présentation d’une<br />

liste de sites relevant davantage d’un guide naturaliste. Cependant,<br />

grâce au militantisme de l’auteur et à sa passion pour la conservation<br />

des forêts primaires, le chapitre 7 constitue un très bon<br />

argumentaire sur la nécessité de protéger les dernières reliques<br />

de forêts primaires en Europe.<br />

En conclusion, le livre d’Annick Schnitzler-Lenoble est une<br />

contribution originale et importante à l’écologie et à la gestion<br />

des forêts primaires d’Europe. La rédaction de cet ouvrage est<br />

cependant inégale et oscille entre la présentation de résultats<br />

scientifiques originaux relativement pointus et une approche plus<br />

naturaliste. Ce compromis devrait cependant permettre à chaque<br />

lecteur de trouver des éléments pour ses propres recherches ou<br />

augmenter ses connaissances en matière d’écologie forestière.<br />

THIERRY DUTOIT<br />

Institut méditerranéen d'’écologie et de paléoécologie<br />

(IMEP, CNRS-UMR 6116), Université d’Aix-Marseille I,<br />

Faculté des sciences et techniques de Saint-Jérôme.<br />

Case 461. F-13397 Marseille Cedex 20.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


Important bird areas in Africa<br />

and associated islands: priority sites<br />

for conservation<br />

Fishpool L.D.C. & Evans M.I. (eds)<br />

Newbury and Cambridge, UK: Pisces.<br />

Publications and Birdlife International<br />

(Birdlife Conservation Series N° 11), 1 144 p. (2001)<br />

Cet ouvrage monumental représente la dernière-née des<br />

publications de Birdlife International, consortium mondial d’organisations<br />

consacrées à la conservation durable des oiseaux et<br />

de leurs habitats. Ce volume correspond à la traduction écrite<br />

de l’exécution de l’ambitieux programme IBA (Important Bird<br />

Areas) conduit sur le continent africain et les îles adjacentes (y<br />

compris certaines terres australes de l’Océan indien ou de<br />

l’Atlantique sud). Ce document résulte des contributions d’un<br />

très grand nombre d’ornithologues amateurs ou professionnels,<br />

d’experts dans le domaine de la conservation et d’une centaine<br />

de rédacteurs. Sur la base du croisement de critères complexes<br />

de vulnérabilité et de rareté d’espèces et d’inventaires et cartographies<br />

de terrain, cette compilation identifie un total de 1 228<br />

IBA couvrant une superficie de plus de deux millions de km 2 ,<br />

soit environ 7 % de la région considérée (58 pays). L’essentiel<br />

de l’ouvrage est constitué par 58 sous-chapitres (un par pays ou<br />

territoire) comprenant une description générale de la situation<br />

écologique et ornithologique de la zone concernée, une cartographie<br />

des IBA identifiées, un tableau synthétique des critères<br />

de classification retenus, ainsi qu’autant de fiches descriptives<br />

ecologia mediterranea, tome 28, fascicule 2, p. 109-111<br />

ANALYSES D’OUVRAGES ◆<br />

complètes que de zones identifiées. Enfin, une bibliographie souvent<br />

abondante est également présentée.<br />

Les territoires sous bioclimat méditerranéen, bien que spatialement<br />

minoritaires à l’échelle du continent africain comptent<br />

une densité particulièrement élevée d’IBA identifiées. Ainsi, le<br />

Maroc, l’Algérie, l’Egypte, la Tunisie et la Province du Cap en<br />

Afrique du Sud, figurent parmi les territoires africains présentant<br />

la plus grande valeur en termes d’abondance de zones à fort<br />

intérêt pour la conservation du patrimoine avifaunistique. Les<br />

critères les plus fréquemment retenus correspondent à la présence<br />

d’espèces d’oiseaux fortement menacées à l’échelle mondiale<br />

(species of global conservation concern) ou à la concentration<br />

d’une part notable des effectifs mondiaux de certains taxons.<br />

Incontestablement, un tel ouvrage de synthèse faisait défaut<br />

à la recherche ornithologique africaine et nul doute que sa publication<br />

et sa large diffusion (le prix demeure modeste au regard<br />

de la masse de données présentées) constitueront une aide précieuse<br />

à la planification et la dynamisation de la politique de<br />

conservation des oiseaux sur ce continent incroyablement varié<br />

et contrasté. En outre, ce livre devrait grandement participer à<br />

éclairer un plus vaste public y compris celui des décideurs, sur<br />

la formidable diversité de l’avifaune africaine et sur sa situation<br />

de conservation souvent critique, tant dans la partie méditerranéenne<br />

que dans le reste du territoire.<br />

ERIC VIDAL<br />

Institut méditerranéen d'’écologie et de paléoécologie<br />

(IMEP, CNRS-UMR 6116), Université d’Aix-Marseille I,<br />

Faculté des sciences et techniques de Saint-Jérôme.<br />

Case 461. F-13397 Marseille Cedex 20.<br />

111


Trivelly Elyse<br />

Boisement spontané des pelouses sèches :<br />

trajectoires historiques, représentations sociales<br />

et perspectives de gestion<br />

Thèse de doctorat en sciences soutenue le 03 juillet 2002 à l’Université<br />

d’Aix-Marseille III, Faculté des sciences de Saint-Jérôme, IMEP<br />

(CNRS-UMR 6116), case 461, 13397 Marseille Cedex 20, France.<br />

Jury<br />

Raphaël Larrere (directeur de recherche, INRA-Ivry), Bernard Brun (MCF<br />

HDR, Univ. Provence), rapporteurs ; Thierry Tatoni (professeur, Univ.<br />

Aix-Marseille III), Jacques Daligaux (MCF, Univ. Provence), Pierre Frapa<br />

(directeur d’étude, Agence Paysages, Avignon), examinateurs ; Thierry<br />

Dutoit (MCF-HDR, Univ. Provence) et Marcel Barbero (professeur, Univ.<br />

Aix-Marseille III), co-directeurs.<br />

Les pelouses sèches connaissent aujourd’hui un regain d’intérêt<br />

au sein de la communauté scientifique, en raison de leur biodiversité.<br />

Or, face à la baisse d’une pression agropastorale pluriséculaire,<br />

elles sont soumises à un phénomène de boisement spontané provoquant<br />

leur régression généralisée. Ce travail vise à apporter une contribution<br />

au débat sur l’écologie de la restauration par une analyse pluridisciplinaire.<br />

Nous avons ainsi recours à l’histoire, la géographie et<br />

l’écologie d’une part, pour comprendre ce phénomène sur nos deux<br />

sites d’étude situés dans le sud-est de la France, les crêtes du Grand<br />

Luberon (Vaucluse) et le plateau de Caussols (Alpes-Maritimes), et<br />

d’autre part à la sociologie afin d’appréhender les représentations<br />

sociales de ces espaces, à l’heure de leur mise en gestion conservatoire.<br />

Il s’agit ainsi d’obtenir un double éclairage, territorial et social,<br />

sur un phénomène qui met en jeu les relations d’une société à son<br />

espace et d’en tirer des enseignements en terme d’aide à la décision<br />

de gestion.<br />

Mots-clés : pelouses, pression agropastorale, restauration, écologie,<br />

Luberon, sociologie, gestion.<br />

ecologia mediterranea, tome 28, fascicule 2, p. 113-115<br />

Résumés de thèses de doctorat en sciences<br />

Rathgeber Cyrille<br />

Impact des changements climatiques et de l’augmentation<br />

du taux de CO2 atmosphérique sur la productivité<br />

des écosystèmes forestiers : exemple du pin d’Alep<br />

(Pinus halepensis Mill.) en Provence calcaire (France)<br />

Thèse de doctorat en sciences soutenue le 08 juillet 2002 à l’Université<br />

d’Aix-Marseille III, Faculté des sciences de Saint-Jérôme, IMEP<br />

(CNRS-UMR 6116), case 461, 13397 Marseille Cedex 20, France.<br />

Jury<br />

Joël Guiot, directeur de recherche (CEREGE, Aix-en-Provence), co-directeur<br />

; Philip Roche, maître de conférence (IMEP, Marseille), co-directeur<br />

; Jean-Luc Dupouey, directeur de recherche (INRA, Champenoux),<br />

rapporteur ; Félix Kienast, directeur de recherche (WSL, Birmensdorf),<br />

rapporteur ; Serge Rambal, ingénieur de recherche (CEFE, Montpellier),<br />

examinateur ; Carlo Urbinati, professeur (Université de Padoue), examinateur.<br />

La dendroécologie, qui a pour support la mesure de la croissance<br />

des arbres, dispose de données bien adaptées à l’étude de l’impact<br />

des changements climatiques et atmosphériques sur la productivité<br />

des peuplements forestiers. Les méthodes statistiques classiques de<br />

quantification de la relation cerne/climat ne sont pas capables de<br />

prendre en compte l’effet d’une dérive probable des facteurs environnementaux,<br />

pas plus que l’effet de l’augmentation du taux de CO2 atmosphérique. C’est pourquoi un modèle bioclimatique et un modèle<br />

biogéochimique ont été développés.<br />

Les données provenant de 21 peuplements de pin d’Alep ont permis<br />

de calculer un indice synthétique de croissance qui exprime les<br />

variations inter-annuelles de productivité. Ainsi, pour chaque peuplement,<br />

les trois types de modèle utilisés ont pu être confrontés aux<br />

indices de croissance et validés. Les variations de productivité des<br />

peuplements étudiés sont principalement contrôlées par les fluctuations<br />

du bilan hydrique.<br />

Le modèle biogéochimique n’est pas sensible aux changements<br />

climatiques, mais simule une très forte augmentation de la productivité<br />

liée à l’augmentation du taux de CO2. Le modèle climatique de<br />

croissance simule, en réponse aux changements climatiques, une forte<br />

augmentation de productivité liée à l’augmentation des températures<br />

printanières. Le modèle bioclimatique, en revanche, simule une diminution<br />

significative de la productivité liée à une augmentation de la<br />

sécheresse estivale. La réaction d’un peuplement dépend des conditions<br />

stationnelles. L’exposition, en effet, détermine le bilan thermique<br />

alors que la pente, la capacité hydrique du sol et la perméabilité du<br />

sous-sol contraignent le bilan hydrique.<br />

Mots-clés : dendroécologie, changements climatiques, augmentation<br />

du taux de CO2 atmosphérique, productivité forestière, conditions<br />

stationnelles, modèles de croissance, région méditerranéenne,<br />

pin d’Alep.<br />

113


114<br />

◆ RÉSUMÉS DE THÈSES<br />

Gachet Sophie<br />

Organisation de la biodiversité forestière :<br />

vers une modélisation de la dynamique du sous-bois<br />

en fonction des pratiques sylvicoles<br />

Thèse de doctorat en sciences soutenue le 13 novembre 2002<br />

à l’Université d’Aix-Marseille III, Faculté des sciences<br />

de Saint-Jérôme, IMEP (CNRS-UMR 6116), case 461,<br />

13397 Marseille Cedex 20, France.<br />

Jury<br />

Thierry Tatoni (professeur, Univ. Aix-Marseille III), directeur de thèse ;<br />

Richard Michalet (professeur, Univ. Bordeaux I), François Romane (IR,<br />

CEFE-CNRS, Montpellier), rapporteurs ; Philippe Dreyfus (CR, INRA-<br />

Avignon), Marcel Barbero (professeur, Univ. Aix-Marseille III), Renzo<br />

Motta (professeur, Univ. Turin, Italie), examinateurs.<br />

Pour des raisons historiques et dans un contexte écologique à<br />

fortes contraintes, certains types de formation forestière sont en évolution<br />

notable en région méditerranéenne française. Connaître l’impact<br />

des interventions sylvicoles sur la biodiversité forestière constitue<br />

donc une préoccupation majeure pour les gestionnaires. Dans le<br />

mont Ventoux (Vaucluse), une remontée biologique des feuillus est<br />

observée dans des forêts de pins noirs d’Autriche plantées vers 1900<br />

(restauration des terrains de montagne). L’inventaire floristique<br />

montre que le cortège spécifique du sous-bois est sensible aux changements<br />

générés par cette maturation, aux niveaux taxonomique et<br />

fonctionnel (via les attributs vitaux des espèces). Quatre modèles<br />

liant les caractéristiques dendrométriques du peuplement aux proportions<br />

des différents attributs (spectres biologiques) sont réalisés<br />

et comparés, et des simulations de l’évolution conjointe du peuplement<br />

et du sous-bois sont effectuées sur un siècle.<br />

Mot-clés : écologie forestière, sylviculture, diversité floristique, remontée<br />

biologique, pin noir, hêtre<br />

Véla Errol<br />

Biodiversité des milieux ouverts<br />

en région méditerranéenne : le cas de la végétation<br />

des pelouses sèches du Luberon (Provence calcaire)<br />

Thèse de doctorat en sciences soutenue le 06 décembre 2002<br />

à l’Université d’Aix-Marseille III, Faculté des sciences<br />

de Saint-Jérôme, IMEP (CNRS-UMR 6116), case 461,<br />

13397 Marseille Cedex 20, France.<br />

Jury<br />

Thierry Tatoni, professeur (Univ. Aix-Marseille 3), directeur de thèse ;<br />

Thierry Dutoit, maître de conférence (Univ. Aix-Marseille 1), rapporteur<br />

; Pierre Pech, professeur (Univ. Paris 1), rapporteur ; Marcel Barbero,<br />

Professeur (Univ. Aix-Marseille 3), examinateur ; Gérard De Belair, maître<br />

de conférence (Univ. Annaba, Algérie), examinateur ; Edouard Le Floc’h,<br />

ingénieur de recherche (CNRS, Montpellier), examinateur.<br />

L’érosion de la biodiversité mondiale rend urgente l’étude des<br />

points chauds d’endémisme et de richesse spécifique, afin d’élaborer<br />

une stratégie globale de conservation. Parmi ces régions sensibles,<br />

le bassin méditerranéen abrite notamment une flore vasculaire remarquable<br />

(30 000 taxons dont 50 % sont endémiques). Suite à une<br />

régression généralisée en Europe, les pelouses sèches et autres milieux<br />

ouverts sont devenus des habitats d’intérêt communautaire jugés prioritaires,<br />

tant la responsabilité patrimoniale qui incombe aux gestionnaires<br />

des espaces naturels paraît grande. Bien étudiés en contexte<br />

médio-européen, ces milieux, bien que plus riches et plus répandus,<br />

restent méconnus en contexte méditerranéen.<br />

En Provence calcaire (SE France), le site du Luberon a été inventorié<br />

et un réseau de placettes permanentes a été mis en place et suivi<br />

pendant cinq ans (1995-1999). Plusieurs approches complémentaires<br />

ont été utilisées, la méthodologie principale se basant notamment<br />

sur différentes échelles de perception (du mètre carré à l’écocomplexe).<br />

À partir de relevés floristiques, nous avons pu décrire<br />

l’organisation de la végétation face aux conditions stationnelles et aux<br />

régimes de gestion (stress et perturbations), et en dégager les principaux<br />

bio-indicateurs. À travers les éléments endémiques autochtones<br />

qu’ils hébergent, ces écosystèmes semi-naturels possèdent une<br />

richesse patrimoniale qui leur est propre. Mais cette richesse originelle<br />

est renforcée par l’histoire de la pression anthropique (coupes<br />

de bois, feux, pâturage ovin et caprin) grâce à laquelle ils ont hérité<br />

d’un patrimoine associé plus récent. Cette histoire et ses variations<br />

impliquent aujourd’hui une dynamique spatio-temporelle de la végétation<br />

et une réponse spécifique des groupes fonctionnels de plantes.<br />

Dans le cadre des mesures agri-environnementales (aide financière<br />

aux éleveurs et accompagnement scientifique et technique aux gestionnaires),<br />

il est possible d’améliorer le protocole de suivi, de proposer<br />

des protocoles expérimentaux, et de réorienter le choix des<br />

actions à mettre en œuvre pour une gestion conservatoire.<br />

Si le contexte socio-économique de l’Europe de l’Ouest au lendemain<br />

du XXe siècle est le même quel que soit le cadre biogéographique<br />

étudié (médio-européen ou méditerranéen), il est cependant<br />

très différent en Afrique du Nord bien que le cadre biogéographique<br />

soit similaire. Un travail de recherche serait à envisager dans les pays<br />

du sud de la Méditerranée afin d’y étudier la problématique des<br />

milieux ouverts, d’identifier les enjeux patrimoniaux et de hiérarchiser<br />

les priorités de conservation.<br />

Mots-clés : bases de données floristiques, biogéographie, biologie<br />

de la conservation, espèces menacées, gestion des milieux naturels,<br />

pastoralisme, phyto-écologie, secteurs d’endémisme.<br />

ecologia mediterranea, tome 28, fascicule 2, 2002


Vila Bruno<br />

Influence d’un grand mammifère phytophage<br />

sur la dynamique forestière – Étude dendrochronologique<br />

des conséquences actuelles et passées de l’introduction<br />

du cerf à queue noire sur la végétation ligneuse<br />

de l’archipel de Haida Gwaii (Canada)<br />

Thèse de Doctorat en sciences soutenue le 16 décembre 2002<br />

à l’Université d’Aix-Marseille III, Faculté des Sciences<br />

de Saint-Jérôme, IMEP (CNRS-UMR 6116), case 451,<br />

13397 Marseille Cedex 20, France.<br />

Jury<br />

Pekka Niemelä, professeur, Université de Joensuu, Finlande, rapporteur ;<br />

Regino Zamora, professeur, Université de Grenade, Espagne, rapporteur ;<br />

Michel Denis, directeur d’Unité, CEMAGREF, examinateur, Jean-Louis<br />

Martin, directeur de recherche, CNRS, examinateur ; Jean-Louis Edouard,<br />

chargé de recherche, CNRS, directeur de thèse ; Frédéric Guibal, chargé<br />

de recherche, CNRS, directeur de thèse.<br />

Le but de ce travail est d’identifier et d’évaluer les conséquences<br />

de la phytophagie exercée par un grand mammifère sur la végétation<br />

ligneuse du sous-bois pour comprendre comment le phytophage<br />

modifie la dynamique forestière. Nous avons utilisé la situation quasiexpérimentale<br />

de l’introduction du cerf à queue noire (Odocoileus<br />

hemionus sitkensis Merriam) sur l’archipel de Haida Gwaii (Colombie<br />

Britannique, Canada), recouvert d’une forêt tempérée pluviale jusqu’alors<br />

dépourvue de grand phytophage. Par l’étude des âges, l’approche<br />

dendrochronologique a mis en évidence le blocage du processus<br />

de régénération des deux principales espèces arbustives, le<br />

salal (Gaultheria shallon Pursh) et l’airelle (Vaccinium parvifolium<br />

Smith in Rees) conduisant à la disparition progressive mais totale du<br />

sous-bois. Associée à des analyses chimiques, l’approche dendrochronologique<br />

a mis en évidence le ralentissement de la régénération<br />

modulé par les défenses physiques chez l’épicéa (Picea sitchensis<br />

(Bong.) Carrière) et le tsuga (Tsuga heterophylla (Raf.) Sarg.) et une<br />

sélection des individus basée sur l’expression de défenses chimiques<br />

chez le thuya (Thuja plicata D. Donn ex Lamb.). Les changements<br />

disproportionnés induits par le phytophage entre les différentes<br />

espèces entraînent une modification de la dynamique forestière qui<br />

se traduira à long terme par un changement en composition de la<br />

forêt. L’approche historique rendue possible par l’étude des structures<br />

d’âge des espèces arbustives et des cicatrices laissées par le cerf<br />

sur les espèces arborescentes montre que si le cerf a rapidement colonisé<br />

tout l’archipel, des hétérogénéités spatiales et temporelles apparaissent,<br />

en liaison avec des facteurs géographiques (l’isolement, la<br />

présence/absence de montagne), des facteurs humains (présence/<br />

absence d’exploitation forestière et de chasse), l’habitat et le climat.<br />

Mots-clés : phytophage, pression d’abroutissement, dynamique forestière,<br />

insularité, cervidé, dendrochronologie, Picea sitchensis, Tsuga<br />

heterophylla, Thuja plicata, Vaccinium parvifolium, Gaultheria shallon.<br />

ecologia mediterranea, tome 28, fascicule 2, p. 113-115<br />

Kiss Laurence<br />

RÉSUMÉS DE THÈSES ◆<br />

Impact des incendies sur les communautés<br />

de gastéropodes terrestres en basse Provence calcaire<br />

Thèse de doctorat en sciences soutenue le 18 décembre 2002<br />

à l’Université d’Aix-Marseille III, Faculté des sciences<br />

de Saint-Jérôme, IMEP (CNRS-UMR 6116), case 451,<br />

13397 Marseille Cedex 20, France.<br />

Jury<br />

Roger Prodon (directeur d’études, EPHE, Montpellier II) rapporteur ;<br />

Markus Pfenninger (maître de conférence, Université de Francfort) rapporteur<br />

; Christopher Carcaillet (directeur d’études, EPHE) Montpellier II)<br />

rapporteur ; Robert Cameron (professeur, Université de Sheffield) examinateur<br />

; Marcel Barbero, (professeur, Université d’Aix-Marseille III)<br />

co-directeur ; Frédéric Magnin (chargé de recherche, CNRS, Aix-<br />

Marseille III) directeur.<br />

Le feu est une perturbation majeure des écosystèmes méditerranéens.<br />

Malgré un nombre important d’études traitant de l’impact du<br />

fe