waders and their estuarine food supplies - Vlaams Instituut voor de ...

0.6 0.8 1.0 1.2 1.4 1.6
jaw lengih (mm)
Kig. 3. Nereis diversii olor. Worm length (cm ± SE) as a function
ol jaw length: September 1981. The regression is highly signilkanl
(R J =812%, p< 0.001, n = 288).
estimated by 2 persons as having exactly the same
length and 48% differed 1 cm only (Fig. I). Size estimates
were adjusted to take into account the small differences
between observers. The field estimates were
calibrated with the measurements in the laboratory by
holding living worms near the bill of stuffed Curlews
during sessions comparable to the situation in the field
(al the same distance and using the same telescopes).
The field estimates arc on average 0.5 cm below the
measurements in ihe laboratory (Fig. 2).
Pellet analysis gave another check on the estimated
prey sizes. Four times we collected a pellet which was
regurgitated by individuals observed during an entire
low water period in late summer 1981. The jaws of
Nereis give a good prediction of worm length (Fig. 3).
There is a close agreement between the frequency distribution
of worm sizes as estimated iii the field and as
derived from the pellets (Fig. 4).
Prey
The methods used to determine length, weight, burrow
depth and feeding activity of Nereis are described in
Esselink and Zwarts (1989). A core sampler (15 cm 0.
40 cm deep) was used to determine worm density. The
core was sieved through a I-mm sieve in plenty of sea
water. Although the smaller worms were not always
VERSATILITY OF CURLEWS FEEDING ON NEREIS
320
35
30
251-
•20-
I 15
10-
5 . FIELD ESTIMATES
n=367 intact worms
8 10 12 14
worm lengih (cm)
Fig. 4. Numerous urqiialu feeding on Sereis div, rsi, olor. Cheek un
field (Statutes ol worm length bj analysis of pellets (Fig. 3l. Four
pelleis were produced by individual Curlews which had been observed
during entire low water periods in September 1981.
retained on the I-mm sieve, no worms > 6 cm were
missed (Table 2). which means thai ihe size classes selected
by Curlews were sampled accurately (Table 2
and Fig. 8A).
Analysis
SPSS (Nie etal.
yses.
Results
975) was used for all statistical anal-
Prey handling
Two types of prey capture were distinguished: (1)
faille 2. \n-eis diversicoloi I taction ol worms passing ihe 1-nim
sieve, but retained on a ll.s-iinn sieve; Irom Blomen I unpubl.). ll is
likely thai small worms pass the 0.5-mm sieve, so that an even larger
number ol the smaller worms are missed when the l-inni sieve is
used.
Size (cm) 1 2 3 4 5 6 7 >7
n 2302 1217 37S OS 60 57 44 325
Passing i'', i S4 75 59 31 22 4 2 0