Uejimaet al. • BREEDING AND FORAGING OF THE PLUSH-CRESTED JAY 
89 
and home range of the jays using the minimum 
convex polygon method (MCP) (Mohr 1947, 
Odum and Kuenzler 1955). This consists of 
joining the outermost observation points for each 
flock with a straight line. The largest polygon 
obtained was taken as the size of the territory of 
a flock. This procedure was chosen due to its 
simplicity and wide use in ornithology (e.g.. 
Jullien and Thiollay 1998. Wiktander el al. 2001. 
Ribeiro et al, 2002. Duca et al. 2006). 
Reproduction.— Reproductive activities were 
observed dunng -720 hrs of field observations 
(240 hrs/area). We recorded the number of eggs 
hid and hatching data, and monitored the 
development of nestlings. Nestlings were mea¬ 
sured the second week post-hatching. A nest was 
considered depredated when the entire brood 
disappeared or when there was no evidence inside 
or underneath the nest of any other kind of loss. 
Activities related to incubation, provisioning of 
the brooding bird or nestlings, and cleaning of the 
nesf ' verc also recorded. Two fledglings were 
color banded. The nests were measured (height, 
diameter, brood chamber diameter and diameter 
' ivv 'r s used in its construction) alter nesting 
activities had ceased, collected, and deposited at 
the Capuo da Inibuia Museum ot Natural History 
tC.uritiba. Brazil) and the Klabin Ecological Park 
Museu, n (Telemaeo Borba, Brazil). 
paging .-We made field observations of 
'paging behavior during 1.080 hrs (360 hrs/area). 
'• used focal animal sampling following Altmann 
with 5-min breaks between observ ation of 
? erem individuals or flocks (scan-sampling). 
procedure should have increased data inde- 
penT^nce. The food item, capture substrate, and 
"“lum were recorded for each foraging event of 
1,1 individual jay. We considered three levels ol 
strata, besides the ground: undergrowth (up 
Jj 1 - m above the ground), middle level (from 2 to 
ll: and suheanopy (from 7 m above the ground 
1 'ii below the canopy). The substrates were: 
c! Jun{ T branches, leaves or epiphytes, and air lor 
foraging. 
. Statistical Analyses.— The R X C test for 
dependence (G-test) was used to evaluate the 
M gnificance (P = 0.01) of proportions of events 
dated to foraging behaviors among the three sites 
'.Fowler and Cohen 1986). 
RESULTS 
Rock Sizes, Home Range, and Territory — 
lerritorial behavior was first observed in early 
October (spring). The size (x ± SD) of flocks 
during this period was smaller at VVP (6 ± 0.8 
individuals) and at RTF (5 ± 0.2 individuals) than 
at KF.P (II ± 0.6 individuals). Territories were 
smaller at VVP (5 ± 0.3 ha) and KEP (5 ± 0.7 ha) 
than at RTF (10 ± 0.4 ha: Fig. 2). Territories at 
KF.P and RET did not overlap and were separated 
by 350 and 500 m. respectively. However, over¬ 
lapping (~ 20 %) of territories was detected at 
VVP. The area of greatest anthropogenic food 
supplementation had the greatest overlap at VVP, 
and aggressive territorial behavior was frequently 
observed at this site. Aggressive territorial 
behavior between individuals started with “bob¬ 
bing", (described by Hardy 1961). followed by 
alarm calls. Direct contact between individuals 
was recorded when the dominant individual 
touched the back of an opponent after sweeping 
flights. At times, the opponent persisted and the 
two individuals attacked each other with their feet 
while the remainder of the individuals in the (lock 
stayed close, emitting alarm calls. Territorial 
behavior was relaxed throughout the end of 
summer (Mar). Flock sizes were largest in the 
home-range period (Apr-Sep, mostly fall/wintei) 
at KEP (15 ± 1.92 individuals), followed by RTF 
(11 ± 1.48 individuals), and VVP (8 ± 1.63 
individuals). Home ranges were smaller at VVP 
(20 ± 1.34 ha) than at KEP (30 ± 0.89 ha) and 
RTF' (30 ± 1.04 ha; Fig. 2). Overlapping of home 
ranges (where aggressive behavior was not 
recorded among individuals ol difterent flocks) 
was —25% in all areas. 
Reproduction.— Five active nests were found, 
three at VVP and two at KEP. The first active nest 
was found in mid-October at VVP. Nests were 
placed closer to the animal breeding center at 
KEP and to the restaurant at VVP. Only one nest 
was successful at VVP. The other two nests found 
in this area were depredated several times (n - 4) 
after hatching. Nests were reconstructed (each 18 
to 21 days after depredation) into February, but 
none was successful. Two new nests were 50 m 
from the previous nests and the other two were 
in trees that were closer, but no new nests were 
found in the same tree in which a nest was pre¬ 
viously depredated. 
Nests </7 = 5) were placed in a fork between 4 
and 7 m above the ground in the middle stratum of 
the forest, and were rounded and consisted of a 
platform with a brood chamber in the center. They 
were built with sticks up to 3 mm in diameter, and 
the brood chamber was lined with twigs <1 mm 
