100 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 
9 
JS' ! t T le T d r <A) a " d CopUla,lon “ U “”P ,S ®)*lek, throughout to Lesser Prairie- 
spnng lekktng seasou t„ the Texas Southern High Plains a, two Inks in 2008 and three leks in 200a 
Overall, morphological traits contained the lea 
variability (mean CV = 6.72%; Table 1). Pinm 
length exhibited the most variation of the morphi 
logical traits. Mean CV for behavioral trait 
excluding proportion of time fighting, was 44.05? 
We did not observe fighting often (/? = ) 
occurrences by 8 males) while recording locatic 
points, which could explain the greater variation i 
this variable. Other behavioral traits exhibited litt 
heterogeneity. Territory characteristics (distance i 
lek center and size of territory) was the most variabl 
parameter (mean CV = 70.35%, Table I). 
Males displayed strong fidelity to specifi 
territory position on the lek within a season afte 
initial territory establishment. The only marke 
male that was observed during both years of th 
study occupied the same territory. Males occupie. 
the same territories between morning and evenim 
A " Jocations Gained during th', 
locations in k ,v mg - Pen ° d Were consis, ent witl 
locations in morning periods. 
’ J.005) and successful (r = —05 
P = 0.04) copulations obtained by males (Fig. 2|. 
Percent of time engaged in face off behavior was 
also correlated (positive) with proportion ot all 
copulations obtained (/• = 0.5. P = 0.03). although 
marginally with proportion of successful copula¬ 
tions (r = 0.4. P = 0.06). The other characteristic, 
were not significantly correlated (P > 0.1) with 
proportion of copulations obtained 
The most parsimonious model, considering ail 
copulation attempts (n = 52 copulation attempts- 
19 males), included percent of time spent idle anct 
MCP, and obtained 39% of the Akaike weigh 1 
(Table 2). All models containing idle were tanked 
higher than those not containing idle, which also 
had the greatest model averaged slope coefficient 
(Fig. 3). The third and fourth best model 1 - 
appeared competitive but contained pretending 
variables, which contribute little to model lit as 
evidenced by similar deviance values. Relame 
importance values for idle. MCP. age, pinnae, 
display, distance, and mass were 1.00. 0.52.04 
0. J 3, 0.12. 0.00, and 0.00. respectively. 
T he most parsimonious model, considering only 
successful copulation attempts (n = 30 copulations. 
