SHORT COMMUNICATIONS 
789 
scale, and tarsus length to the nearest 0.01 mm 
with digital calipers) of adults and nestlings (when 
they were 5- and 11-days of age). However, we 
only sampled 16 individuals at II-days of age, 
and all came from later broods. We had 
hematocrit data for 12 of these nestlings (6 
broods) at both 5- and 11-days of age. We also 
collected a blood sample (25-100 ul) between 
0900 and 1800 hrs from the brachial vein in adults 
and the metatarsal vein in nestlings in a 50-ul 
heparinized microcapillary tube. These samples 
were kept cool at 4 C until processing, 2 to 4 hrs 
later. Blood in microcapillary tubes was spun in 
a centrifuge (Sero-fuge, Model # 0591. Clay 
Adams. Parsippany. NJ, USA) at 8.000 rpm lor 
6 min. Two different measurements of hematocrit 
percentages were taken to compare with total 
blood volume. The first method involved mea¬ 
suring the length of the hematocrit and the total 
length of the blood (hematocrit and plasma) with a 
ruler to the nearest 1.0 mm to ascertain the 
percentage of hematocrit (Cuervo ct al. 2007), and 
the other used the standardized percentages 
written on the centrifuge. The results for these 
two methods were highly correlated </\ = 0.91. ii 
= 124, P < 0.0001), and were averaged to 
calculate the percentage of hematocrit in the 
blood. This average was compared to another 
index of condition (residuals of mass vs, average 
tarsus length) in adult males and females, and 
nestlings. Residuals were calculated through 
regressing mass on average tarsus length ol all 
individuals captured in 201 1 (regardless ol 
whether we had hematocrit data for them). 
Regressions were calculated separately lor adult 
males and females; males tended to have larger 
tarsi than females (mean ± SE = 34.81 ± 0.22 
vs. 34.32 ± 0.19. respectively; unpaired / = 
1.707, df = 39. P = 0.10), and also tended to 
weigh more than females (84.26 ± 1.19 vs. 81.44 
~ 1.01 g, respectively: unpaired / = 1.804. dt = 
39. P = 0.08). Residuals were similarly calculated 
for nestlings with a different regression for day 
5 and day 11 nestlings. We calculated mean 
hematocrit and mean condition per brood lor the 
5- and 11-day old nestlings to avoid pseudorep¬ 
lication in the correlation analyses. Mean hemat¬ 
ocrit and mean condition per brood were calcu¬ 
lated using only those nestlings lor which we had 
both hematocrit and condition scores. 
Ambient temperature ( C) and relative humid¬ 
ity (%) data were obtained from the web site for 
Environment Canada’s meteorological station at 
the Halifax Stanfield International Airport (44 
53' 00.000" N. 63 31' 00.000" W). Nova Scotia, 
35 km from our study site. Both ambient 
temperature and relative humidity are reported 
on an hourly basis; we took these data at the time 
when blood was collected, rounding up or down to 
the nearest hour. Two adults were omitted from 
this analysis as (heir capture times had not been 
recorded. 
Data were tested for normality with a D'Agos¬ 
tino and Pearson omnibus normality test. We used 
Pearson correlations when both variables were 
normally distributed, and Spearman rank correla¬ 
tions when al least one variable had a non-normal 
distribution, to examine potential relationships 
between hematocrit and (I) condition, and (2) 
factors such as ambient temperature. We ran a 
linear-mixed effects model with age/sex as a fixed 
effect and subject as a random effect to test if 
differences in hematocrit existed between: (1) 
male and female adults. (2) 5-day old nestlings 
from early and later broods, (3) adults and 5-day 
old nestlings. (4) adults and 11-day old nestlings, 
and (5) nestlings at 5 days of age compared to 
when they were 11 days ol age (only 6 later 
broods were sampled at both 5- and 11-days ol 
age). We included subject as a random effect, as a 
subset of the later brood nestlings (// =12) was 
measured on both days' 5 and 11. We used a 
Tukey USD test to identify the location of the 
differences. Data were analyzed using GraphPad 
Prism Version 5.04 (GraphPad Software. San 
Diego. CA. USA) and JMP Version 10 (SAS 
Institute. Cary. NC. USA). All tests were two- 
tailed. Results were considered significant when P 
< 0.05. 
RESULTS 
The mass versus tarsus regression equation for 
adult males was y = 3.585x-40.54 while for adult 
females it was y = 3.152x-26.73. For 5-day old 
nestlings, it was y = 2.9l7x-31.67 while tor 11- 
day old nestlings it was y = 5.565x-l 16.2. The 
slope was significantly greater than zero in all 
cases (males: F = 11.28. dl = 1. 15, P = 0.0043; 
females: F = 11.98, df = 1. 22. P = 0.0022; 5- 
day old nestlings: F = 1114. dl = 1, 122, P < 
0.0001; 11-dav old nestlings: F = 204.9. df = 1. 
103. P < 0.0001). 
Hematocrit was not correlated with condition in 
adult males (r, = 0.25, n = 15, P = 0.37) or adult 
females (r = —0.0997, n — 17, P = 0.70). Simi¬ 
larly, mean hematocrit/brood was not correlated 
