790 
THE WILSON JOURNAL OF ORNITHOLOGY . Vol 124. No. 4. December 2012 
TABLE 1 Mean hematocrit (%) (SE). range, and sample size (/,) for European Starlings of different sexes and ages. 
Breeding males 
Breeding females 
5-dav old nestlings (early brood) 
5-day old nestlings (later brood) 
11-day old nestlings 
Brood means: 5-day old (early) 
Brood means: 5-day old (later) 
Brood means: 11-day old 
Mean ± SE 
52.76 ± 1.58 
52.37 ± 1.62 
37.03 ± 0.86 
33.64 ± 1.37 
43.06 ± 0.73 
36.44 ± 1.16 
34.30 ± 1.48 
42.78 ± 1.31 
Low 
39.17 
42.24 
23.00 
24.45 
38.53 
25.20 
28.80 
38.53 
High 
59.92 
69.63 
56.15 
48.19 
48.59 
47.55 
43.89 
47.88 
15 
17 
51 
24 
16 
17 
9 
6 
with mean brood condition for 5-day old nestlings 
(/• = -0.152, n = 26 broods, P = 0.46) or 11 -dav 
old nestlings (r s = -0.143, n = 6 broods, P = 
0.80). although sample size was small for the 
latter. 
We found significant variation in hematocrit 
among adults, 5- and 11-day old nestlings (r = 
0.63; P < 0.0001); fixed effect tests indicated that 
hematocrit was significantly related to age/sex 
(/r 4, 90.16 ^ 43.96, P < 0.0001). Individual 
explained 3.0% of the variation in hematocrit. 
There was, however, no significant difference in 
the hematocrit levels of adult males and females 
(Table 1). AdulLs (both male and female) had 
significantly higher hematocrit levels than 5 - and 
ll-dav old nestlings, while 11 -day old nestlings 
(from later broods) had intermediate hematocrit 
levels that were significantly higher than 5 -day 
old nestlings from early and later broods. 
Hematocrit was not significantly correlated 
with Julian date (sampling date) for adults (r 
- -0.129. „ = 32, P = 0.48) or 11-day old 
nestlings (r v - 0.177. n = 6 broods, / J = 0 71) 
However, mean hematocrit/brood tended to be 
negatively correlated with Julian date for 5-day 
old nestlings (r, = -0.341, „ = 26 broods P ~ 
0.09). Hematocrit was not correlated with 
ambient temperature (adults; r v = -0.093, „ = 
30, P = 0.62: 5-day old nestlings: r x = 0.032. n 
- 26, P — 0.88; 11-day old nestlings; r v = 0.177 
- = °' 7I) ’ ° rreIative humidity (adults; r 
- -0.096, « = 30, P = 0.61; 5-day old 
nestlings: r = -0.180. it = 26, P = 0 38- 1 |- 
day old nestlings: r s = 0.177, „ = 6. P = [)j\ ) 
Hematocrit was also not correlated with sampling 
time of day tor adults (/• = 0.117, n = 30 p = 
0-54) or 11-day old nestlings (r x = 0.143 n = 6 
for °5 d- ^ Th 0 80 ;- bl " Was P° si ‘^ correlated 
for 5-day old nestlings (/• = 0.407, n = 7 6 P 
0.04; Fig. 1). ’ 
DISCUSSION 
We delected no significant correlation between 
hematocrit and condition in adults or nestlings ( 5 - 
or II-days of age). We do note, however, that 
sample size was small for broods of 11 -day old 
nestlings. Our results agree with previous studies 
where no association was found between hemat¬ 
ocrit and condition (e.g., Dawson and Bortolotti 
1997, Aniat et al. 2009). Other studies have found 
(hat supplementally-fed birds had higher hemat- 
oeiits than non-fed birds (e.g., Merino and Potti 
1998; but see Cuervo et al. 2011), or that 
nun itionally-stressed birds had lower hematocrit 
lhan non-stressed birds (e.g., Piersma et al. 2UU0). 
( uervo et al. ( 2011 ) in their study of the closely 
tclated Spotless Starling (Stunms unicolor) found 
hematocrit was positively correlated with condi- 
1,0,1 rinass controlled for body size) in a control 
group ol nestlings. This relationship became non¬ 
significant when nestlings were provided with 
supplemental feedings, but there was no signifi¬ 
cant reduction in the variation of nestling 
condition. Cuervo et al. (2011) concluded that 
■t£ 
50 
40 
30-1 
20H 
10i 
10 
—i— 
12 
14 
16 
18 20 
Time of Day 
fig. I. Mean hematocrit versus time of day when 
oo was sampled for 5-day old ncstlina European 
Starlings (/• = 0.407, n = 26 broods. P = 0.04). 
