20.03.2013 Views

Intraspecific Variation in Viola suavis in Europe ... - Annals of Botany

Intraspecific Variation in Viola suavis in Europe ... - Annals of Botany

Intraspecific Variation in Viola suavis in Europe ... - Annals of Botany

SHOW MORE
SHOW LESS

Create successful ePaper yourself

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

448<br />

TABLE 1. List <strong>of</strong> characters used <strong>in</strong> the morphometric analyses <strong>of</strong> <strong>Viola</strong> <strong>suavis</strong> populations, <strong>in</strong>clud<strong>in</strong>g eigenvectors express<strong>in</strong>g<br />

correlations <strong>of</strong> characters with the first and second pr<strong>in</strong>cipal components <strong>of</strong> PCA (Axis 1 and Axis 2; see Fig. 5)<br />

Character Character scor<strong>in</strong>g/units Axis 1 Axis 2<br />

Stolons<br />

StA* Above-ground stolons 0 absent; 1 present 0.044 0.129<br />

StU* Underground stolons 0 absent; 1 present 0.230 0.090<br />

Lam<strong>in</strong>as and petioles<br />

LHL*<br />

LSL*<br />

LSA*<br />

LCN*<br />

Maximum hair length (on petiole)<br />

Lam<strong>in</strong>a s<strong>in</strong>us depth<br />

Lam<strong>in</strong>a s<strong>in</strong>us angle<br />

Number <strong>of</strong> crenulae along both<br />

lam<strong>in</strong>a marg<strong>in</strong>s (¼ lam<strong>in</strong>a<br />

dentations)<br />

(mm)<br />

(cm)<br />

(degree)<br />

–<br />

–0.070<br />

0.293<br />

–0.312<br />

–0.054<br />

–0.127<br />

0.262<br />

–0.296<br />

0.063<br />

LAA*<br />

LP*<br />

Lam<strong>in</strong>a apex angle<br />

Violet pigmentation <strong>of</strong> lam<strong>in</strong>a<br />

(degree)<br />

0 absent; 1 present<br />

0.130<br />

0.279<br />

–0.011<br />

–0.145<br />

Stipules (outer stipules <strong>of</strong> ma<strong>in</strong> rosette-leaves)<br />

SW*<br />

SFN*<br />

Stipule width<br />

Number <strong>of</strong> fimbriae (¼ glandular<br />

fimbriae, non-glandular fimbriae and<br />

sessile glandule) along both stipule<br />

marg<strong>in</strong>s<br />

(mm)<br />

–<br />

0.258<br />

0.234<br />

–0.001<br />

0.204<br />

SFL* Maximum fimbriae length on stipule (mm) –0.212 0.286<br />

Peduncules<br />

PL Peduncle length (cm)<br />

PL1<br />

PP*<br />

PL1/PL*<br />

Peduncle length below bracteoles<br />

Peduncle pigmentation<br />

Peduncle length below bracteoles/<br />

peduncle length<br />

(cm)<br />

0 absence; 1 presence<br />

–<br />

0.314<br />

0.341<br />

–0.253<br />

0.091<br />

Calyx (sepals)<br />

KAL* Anterior sepal length (mm) –0.075 0.146<br />

KAW* Anterior sepal width (mm) 0.220 0.254<br />

KP* Violet pigmentation <strong>of</strong> sepals 0 absence; 1 presence 0.217 –0.201<br />

Corolla (petals)<br />

CPL*<br />

CPW*<br />

CP*<br />

CPSP*<br />

Posterior petal length<br />

Posterior petal width<br />

Corolla colour (exclud<strong>in</strong>g spur)<br />

Pigmentation <strong>of</strong> corolla <strong>in</strong> contrast<br />

(mm)<br />

(mm)<br />

0 white; 1 blue to violet<br />

0 spur paler than corolla; 1 spur the<br />

0.109<br />

0.110<br />

–0.278<br />

–0.278<br />

0.368<br />

0.345<br />

0.338<br />

–0.298<br />

to pigmentation <strong>of</strong> spur<br />

same colour as corolla; 2 spur<br />

darker than corolla<br />

* Characters used <strong>in</strong> multivariate analyses; for characters illustrations see Hodálová et al. (2008).<br />

characters were quantitative, six b<strong>in</strong>ary, and one semiquantitative.<br />

Characters express<strong>in</strong>g colouration and the<br />

presence/absence <strong>of</strong> stolons (StA, StU, LP, PP, KP, CP,<br />

and CPSP; for character explanation, see Table 1) were<br />

scored immediately <strong>in</strong> the field on fresh plants. Otherwise<br />

the vegetative (leaves and stipules) and floral (peduncles,<br />

calyx and corolla) organs were attached by adhesive<br />

tape to paper, dried, and then used for measurements.<br />

Whenever possible, three measurements were made for<br />

each vegetative character, and two measurements were<br />

made for each floral character; average values were then<br />

entered <strong>in</strong>to the data matrix. Twenty characters were used<br />

<strong>in</strong> morphometric analyses; two characters were used solely<br />

for calculat<strong>in</strong>g ratios (Table 1). Corolla colour (CP) was<br />

excluded from canonical discrim<strong>in</strong>ant analyses because it<br />

was constant with<strong>in</strong> the groups.<br />

Diagnostic characters <strong>in</strong> the subsection <strong>Viola</strong> <strong>of</strong>ten change<br />

after the flower<strong>in</strong>g dur<strong>in</strong>g late spr<strong>in</strong>g and summer (e.g. leaf<br />

shape, the length <strong>of</strong> hairs on petioles) or disappear (characters<br />

Mered’a et al. — <strong>Intraspecific</strong> <strong>Variation</strong> <strong>in</strong> <strong>Viola</strong> <strong>suavis</strong><br />

on stipules and flowers). Therefore, all morphometric<br />

analyses were performed on material collected <strong>in</strong> spr<strong>in</strong>g.<br />

The <strong>in</strong>dument <strong>of</strong> leaves was observed on petioles <strong>of</strong> young,<br />

still-develop<strong>in</strong>g leaves <strong>in</strong> spr<strong>in</strong>g or, <strong>in</strong> some cases, on petioles<br />

<strong>of</strong> over-w<strong>in</strong>ter<strong>in</strong>g leaves that had developed <strong>in</strong> summer or<br />

autumn <strong>of</strong> the previous year. Only open (chasmogamous)<br />

vernal flowers were considered.<br />

Two datasets (matrices) were assembled. Matrix A<br />

<strong>in</strong>cluded all <strong>in</strong>dividuals <strong>of</strong> V. <strong>suavis</strong> (both morphotypes;<br />

340 <strong>in</strong>dividuals from 35 populations; Appendix) sampled.<br />

This matrix was used to ga<strong>in</strong> <strong>in</strong>sight <strong>in</strong>to the overall variation<br />

pattern <strong>in</strong> V. <strong>suavis</strong>, especially with regard to the position<br />

<strong>of</strong> the white-flowered populations. Matrix B represented a<br />

selection <strong>of</strong> samples <strong>of</strong> V. <strong>suavis</strong> correspond<strong>in</strong>g to the<br />

material analysed for AFLPs: from spatially separated<br />

populations all <strong>in</strong>dividuals sampled and measured were<br />

<strong>in</strong>cluded; <strong>in</strong> cases where blue- and white-flowered <strong>in</strong>dividuals<br />

co-occurred, only the <strong>in</strong>dividuals analysed for AFLPs were<br />

<strong>in</strong>cluded (245 <strong>in</strong>dividuals from 29 populations). Matrix B<br />

Downloaded from<br />

http://aob.oxfordjournals.org/ by guest on March 20, 2013

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!