340 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 2. June 2012 
TABLE 1. Number of recorded visits per bird species, visited infructescences (mean ± SE). consumed fruits per boui, 
and per infructescence in the Dry Zone of Santa Cruz (Galapagos Islands) during the 2009 cool-dry season. 
Species 
No. visits (%) 
Visited infructescences 
per bout 
Consumed fruits per 
infructescence 
Consumed frails per fens 
Geospiza fortis 
G. magnirostris 
83(92) 
7(8) 
6 ± 1 
22 ± 9 
5 ± 0 
3 ± I 
26 ±4 
60 ±22 
RESULTS 
/-. comma plants produced an average (± SE) 
of 1,075 ± 158 infructescences, each containing 
12 ± I fruits per infructescence. Adult L. camara 
plants had an urea of • 6.0 i 0.8 m*, and were 
2.5 ± 0.2 m in height on average; thus, each 
plant produced a mean of 2,820 ± 229 fruits/nr. 
Only four (2%) of a total of 252 fruits in the 
bagged infructescences fell into the bags alter 
1 month. 
Ninety records of birds feeding on L. camara 
fruits were obtained during 24 hrs of observation: 
8% corresponded to the Large Ground Finch 
(Gcospiza magnirostris) and 92% to the Medium 
Ground Finch (G. fortis) (Table I). Neither finch 
species swallowed entire fruits of /.. camara. but 
crushed them, eating the embryo and the endo¬ 
sperm and discarding the empty fruit coal. G 
fortis removed 2.138 fruits and G. magnirostris 
149 fruits. G. magnirostris ate fruits front an 
average of 22 ± 9 infructescences on each visit to 
a L. camara plant, consuming 3 ± I fruits per 
infructescence. G. fortis visited 6 ± 1 infructes- 
cences. consuming 5 ± 0 fruits per infructescence 
(Table 1). The number of fruits removed per 
foraging bout was variable: 1-157 fruits for G 
magnirostris and 1-207 fruits for G. fortis. On 
average. G. magnirostris ate 60 ± 22 fruits pet- 
bout and G. fortis 26 ± 4 fruits per bout (Table I). 
Some fruits were observed falling intact to the 
ground as a consequence of infructescence 
manipulation by both species of finches. 
Most G. fortis (59%), after feeding on L 
camara , flew from the area (out of sight), while 
25% flew to a nearby tree. 12% continued feeding 
on another L. camara plant, and 4% flew to the 
ground and continued foraging. Twenty-nine 
percent ol the G. magnirostris flew from the area 
after feeding on L. camara. 29% flew to a nearby 
tree. 29% continued eating on another L. camara 
plant, and 13% Hew to the ground (Fig. |) The 
“ a r™L finchcs f after ,cedins l ~ 
™ independent of ,l le number of fruits eaten 
caison correlation coefficient. P > 0.05). 
Fruit traps under L. camara plants collected an 
average of 684 ± 105 fruits/nr during a month, 
which represented 24% of the fruit production ot a 
plant (Fig. 2). Of the trapped fruit. 311 r 59 were 
entire fruits and 373 ± 67 were empty fruit coats, 
indicating seed predation on the plant. Thus. 13T 
of the fruit production was consumed on the plant, 
w hile I I % was dislodged, mostly as a consequence 
ol infructescence manipulation (Fig. 2). 
After I month, 66% of the fruits placed in plots 
on the ground had been removed by unknown 
agents, 15% were crushed and eaten in situ (fruit 
remains were found, similar to those dropped by 
finches), and 19% remained intact on the soil 
surface (Fig. 2). 
The number of seedlings under adult L. camara 
plants varied from 0 to 12, resulting in an average 
°f I 1 0 seedlings/m 2 . A significant positive 
correlation was found between seedling density 
and percentage of soil covered by rock (r - 0,67. 
P < 0.05, n = 9), discarding as outlier an adult 
plant under which was a high density of seedlings 
(3 seedlings/m 2 ) and 50% rock cover. 
DISCUSSION 
Our results demonstrate that seeds of invasive L 
camara are part of the diet of endemic ground 
finches (G. magnirostris and G. fortis ) during die 
cool-dry season in the Dry Zone of Santa Cmz 
Island (Galapagos). Guerrero and Tve (2009 1 and 
Buddenhagen and Jewell (2006), reported ground 
finches most commonly crush dry seeds. We ab 
found that both finch species behaved mostly * 
seed predators of L. camara rather than disperers. 
crushing the dry fruits and feeding on seed enibO'' 
Ground finches, despite their predominant 
predatory role in this study, may also act as shori- 
distance dispersers as their foraging activity 
caused fruit drop, which may be important given 
the low percentage of fruit drop from bagge J 
infructescences. G. magnirostris visited n ,n 
infructescences and ate many fruits per bout-b|‘ 
this species was only occasionally observed- 6 
fortis visited more plants, ate at fewer intrude- 
cences per plant and ate less fruits per bout, hm 
