SHORT COMMUNICATIONS 
809 
1 
Ot 
C 
Vi 
O 
o 
£ 
U 
VI u 
V 41 
5 * 
£ = 
>— ‘r 
= P 
o * 
0.9 
0.8 
0.7 
0.6 
0.5 
0.4 
0.3 
0.2 
0.1 
0 
X 2 = 10.96 
P< 0.001 
X 2 = 20.54 
P< 0.001 
X 2 = 12.24 
0.UU1 
X 2 = 1.92 
P= 0.165 
X 2 = 26.38 
P< 0.001 
Both Vacant 
Fem. Saf. 
Fern. Sun. 
Social context 
Male Saf. 
Male Sun. 
FIG. 1. Male Northern Cardinal feeder visitation in five social contexts: both feeders vacant (Both Vacant, n -^0h ® 
female on the safflower feeder (Fem. Saf.. n = 85). a female on the sunflower leeder (Fun. an .,n • _ 
safflower feeder (Male Saf.. n = 70), a male on the sunflower feeder (Male Sun., n - 104). All / tests wi ^ . 
Vacant’ compared to a null expectation of no preference (1:1). All other distribut.ons compared to a null expectant 
0,62:0.38 (Both Vacant). 
We recorded the number of visits by male and 
female cardinals to each feeder and noted the 
location and sex of other cardinals present. We 
defined a visit to a feeder as any time a cardinal 
landed on the feeder perch and remained >1 see. 
Cardinals are larger than and dominant to other 
common feeder birds (e.g., Carolina Chickadee 
\Poecile carolinensis], Tufted Titmouse \Baeolo- 
phits bicolor]) and we assumed the occasional 
presence of smaller species did not influence choices 
made by cardinals. We excluded from our analyses 
eases in which both feeders were occupied when a 
third (focal) cardinal made a choice about which 
feeder to visit. The cardinals we observed were not 
individually marked, and wc sought to minimize 
pseudoreplication by collecting data at five separate 
locations and limiting the study to 1 week. 
Statistical Analyses.- We used GraphPad 
QuickCalcs software (GraphPad Software Inc.. La 
Jolla, CA, USA; GraphPad.com) to perform /; tests. 
We compared the proportion of visits to sunflower 
versus safflower seed feeders (when both feeders 
were vacant) to an expected equal preiportion. We 
compared visitation rales to feeders with sunflower/ 
safflower seed to expected rates generated when both 
feeders were vacant for our subsequent analyses. 
RESULTS 
Male Northern Cardinals visited the feeder 
containing sunflower seeds significantly more 
often (n = 360, * 2 , = 20.54. P < 0.001; Fig. 1) 
than the feeder containing safflower seeds when 
both feeders were empty of conspecifics. Females 
exhibited a similar bias towards sunflower seeds 
(„ = 473, x 2 i = 35.18, P < 0.001; Fig. 2). 
Wc compared the feeder choices of males in the 
presence of conspecifics to those made by males 
when both feeders were vacant. Males visited the 
safflower seed feeder significantly more often 
when a female was present at the sail lower seed 
feeder and the sunflower seed feeder was vacant 
(y= 12.24. P < 0.001; Fig. 1). Males visited 
the sunflower seed feeder significantly more 
often than expected when a female occupied the 
sunflower seed feeder (jr i = 10.96. P < 0.001; 
Fig. I). Males visited the sunflower seed feeder 
as expected when another male occupied the 
safflower seed feeder (x’i = L92, P = 0.165; 
Fig. 1). Males visited the safflower seed feeder 
significantly more often than expected when 
another male was present at the sunflower seed 
feeder Or, = 26.38. P < 0.001; Fig. 1). 
Females visited the sunflower seed feeder as 
expected when another female occupied the 
safflower seed feeder (y : i — 2.95, P = 0.086; 
Fig. 2). Females visited the safflower seed feed- 
eAignificantly more often than expected when 
another female was present at the sunflower seed 
feeder (y\ = >9.04. P < 0.001; Fig. 2). Females 
visited the sunflower seed leeder significant¬ 
ly more often than expected when a male occupied 
the safflower seed feeder (/ 2 i = 53.77. P < 0.001; 
