92 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 
4 p wwof £?T en c y ( , PerCenIage) ° f f °° dS COnSUmed b y adu,t Plush-crested Jays in southern Brazil: Vila 
cilia State Park (VYP). Klahin Ecological Park (KEP), and Ribeirao do Tigre Farm (RTF). 
Food item 
VVP 
KEP 
RTF 
Total 
Insects/ invertebrates 
External resources* 
Fruit 
Vertebrates 
Totals 
964 (80) 
134 (11) 
31 (3) 
80 (7) 
1.209 (100) 
1.457 (86) 
103 (6) 
89 (5) 
51 (3) 
1.700 (100) 
559 (93) 
21 (4) 
11 (2) 
8 (1) 
599 (100) 
2,980 (84.9) 
258 (7.4) 
131 (3.7) 
139 (4) 
3.508 (100) 
* Garbage, maize seeds. 
and mol I asks), fruit, many kinds of foodstuffs 
given or discarded by humans (garbage, bread, 
meal, cookies, grain plants, and maize), and 
vertebrates (frogs, eggs, and nestlings, and small 
reptiles). Invertebrates were the most commonly 
used tood resource at all ihree areas, but the 
percentages of each item varied (G = 105.06; 
df = 6; P < 0.01; Table 4). Food resulting from 
human activities was used more at VVP and KEP 
than at RTF where the most commonly consumed 
items were insects and invertebrates. 
Plush-crested Jays fed on fruit from 24 plant 
species, but only the fruits of Syagrus roman- 
zofficma were commonly eaten in all three are,;' 
( fable 5). A greater variety of fruit was observed 
to be consumed at KEP (13 species). while 10and 
five Iruit species were recorded at VVP and RTF. 
respectively. The fruits most frequently recorded 
were: Eriobotrya japonica , Diospyros kaki. Phil¬ 
odendron spp., Casearia sylvestris. Rapanea 
ferruginea, Syagrus romanzoffianum, and Fine 
enormis. 
TAELE 5. Plant species used by Plush-crested Jays and 
relative frequencies of fruits consumed in southern Brazil- 
“ ,VVPX ** 
(KEP), and Ribeirao do Tigre Farm (RTF). 
Plant species 
VVP 
Urera haccifera 
Psidium cattleianum 
Paullinia carpopoda 
Lepismium cruciforme 
Ficus enormis 
F. insipida 
F. monckii 
Melia azedaracli 
Syagrus romanzoffiana 
Casearia sylvestris 
Rapanea ferruginea 
R. umbel lata 
Philodendron spp. 
Cestrum calycinum 
Miconia spp. 
Sty rax leprosus 
Citronella gongonha 
Morus nigra 
Trema micrantha 
Diospyrus kaki 
Citrus spp. 
Eriobotrya japonica 
Peschiera australis 
Celt is iguanaea 
Totals 
KEP 
RTF 
3 
4 
2 
1 
3 
0 
2 
1 
4 
7 
4 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
2 
12 
15 
6 
14 
4 
1 
I 
1 
6 
1 
9 
0 
13 
0 
0 
85 
0 
0 
0 
0 
0 
I 
0 
I 
7 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
1 
0 
1 
1 
12 
The foraging forest stratum and capture sub¬ 
strate were recorded at 2,980 foraging events (964 
VVP, 1,457 KEP, 559 RTF; Table 6). 'Hie pro¬ 
portion ol observations related to foraging stratum 
differed among the areas studied (G = 178.2; df 
=* 6; P < 0.01). The understory and the middle 
level were the most frequently used strata in all 
areas, but the ground and the subcanopy were 
proportionally used more at VVP and RTF 
respectively. 
The proportions of use of the capture substrate' 
also differed among the three areas (G - 80.84. 
df = 8; P < 0.01). Branches and leaves were 
generally the most frequently used substrates in 
the understory and the middle level in all art4> 
(C = 24.18; df = 6; P < 0.01): epiphytes were 
less used, particularly in the subcanopy, and the 
air was the least used. 
Food caching behav ior was also observed and 
recorded (n = 110). Each individual had its own 
cache up to 100 m from where the food was 
captured. Each cache contained only one unit ot a 
given item (except for maize, for which each cache 
could contain up to 12 seeds). Leaves or small 
sticks were laid over the food cache. We did not 
observe individuals defending their caches. Cach¬ 
es locations were in the ground (n — 40), in 
epiphytes (// = 47). and on branches (n = 23). The 
caches mainly contained seeds of .4. angustifolia, 
bones, maize, and seeds of 5. romanzoffianum). 
