SHORT COMMUNICATIONS 
617 
2.5 
.§ 1.5 
0.5 
0 
Fertile period Nestling period 
FIG. 1. Female Dark-eyed Junco home-range size during the fertile and nestling periods. There was a significant 
difference in female home-range size between periods (P = 0.011: fertile period, n = 8: nestling penod. n - 7). Error 
represent ±1 SE. Fertile period data from Neudorf ct al. (2002). 
Virginia base station (37 12' N. 80 25' W). The 
corrected coordinates were projected in shapefile 
format into Universal Transverse Mercator (UTM) 
zone 17. NAD 83. using the Geographic Informa¬ 
tion System (CIS) Program ArcGIS 9.2 (ESR1 
2007). A home-range area (in ha) lor each female 
was estimated using minimum convex polygons 
(MCP) in Hawths Tools 3.26 (Beyer 2004). 
The MCP method (Harris et al. 1990 provides a 
comparison of common methods of home-range 
estimation) was chosen to allow lor an equivalent 
comparison with previously published data on 
female home-range si/e during the. lertile period 
(Neudorf et al. 2002). We used a nonparametric 
Mann-Whitney LMest to compare home-range 
sizes, and a Spearman’s Rho Correlation to test 
tor relationships between home-range size and 
methodological effects. A Kruskal-Wallis test was 
used to examine differences between implant types. 
All tests were performed in SPSS 11.5 (SPSS 2002). 
RESULTS 
We had an average of 62 (range = 53-65) 
observation points per female, and the total 
number of observation points did not correlate 
with home-range size (r — 0.493. P - 0.261; 
Table 1). We found substantial variation (mean ± 
SDi in female home-range size during the nestling 
period with females maintaining a mean home 
range of 0.833 ± 0.788 ha (Table 1). Home-range 
size was not significantly correlated with duration 
of tracking (r = 0.493. P = 0.261). Julian date 
(r = -0.214. P = 0.645). female age (r - 
-0.045, P = 0.924). or nesting attempt (r = 
-O. 543 ' P = 0.208) (Table l).We were unable to 
make strong statistical comparisons about the 
effect of elevated testosterone on female home- 
range size due to our small sample size (testos¬ 
terone-implant - 2, control-implant = 4) and a 
low effect size (Observed Effect Size [hi from 
Retrospective Power Analysis - 0 050). The two 
females that received testosterone implants had 
intermediate size home ranges, and did not ditfei 
detectably from females receiving control im¬ 
plants (Table 1; = 2.89. P = 0.235). 
DISCUSSION 
Female home range was significantly smaller 
during the nestling period than during the fertile 
period when comparing our data to Neudorf et 
al.’s (2002) fertile period home-range estimate 
collected at the same study site from different 
individuals (Fig. 1; Z = 0.008; P = 0.011). We 
collected a mean of 62 (range = 53-65) 
observation points per female, similar to the 
number collected by Neudorf et al. (2002) (mean 
= 71.5 points/individual, range = 54-77) during 
the fertile period. Thus, variation in the number of 
observation points between studies likely had 
minimal impact on our comparison. 
The difference in home-range size between the 
fertile and nestling periods may also be confounded 
