SHORT COMMUNICATIONS 
601 
dietary changes that could possibly have contrib¬ 
uted to the development of this condition at 
Clipperton would have been either a change in the 
ratio of fish and squid consumed, or an overall 
food shortage. 
Another possible cause of angel wing could he 
adult boobies acting aggressively toward other 
chicks. For example, non-breeding adult Nazca 
Boobies (A grant!) have been documented to 
attack nestlings of Blue-footed (5. nebonxii). Red- 
footed (S. sula). and other, nonfumilial Nazca 
boobies, resulting in lacerations on the chicks 
bodies, broken wings at times, and occasionally 
death (Nelson 1978. Townsend el al. 2002. 
Anderson et al, 2004, Muller ei al. 2011). These 
interactions have purportedly resulted in twisted 
wings' among the chicks, hut for three reasons, we 
do not believe such aggression is the cause ol angel 
wing. First, nearly all documented cases ol angel 
wing have involved domesticated waterfowl and 
the explanations that have been offered for its 
occurrence have not included aggression towards 
nestlings by adults, which should be fairly evident 
among birds raised in captivity. Second, angel 
wing has also been documented among hand- 
reared birds in a variety of species, more evidence 
that it is not aggression-induced. Third, we have 
seen no evidence of other damage (e.g... lacerations 
to the head and body) to chicks, including those 
with angel wing, which is commonly evident when 
adult Nazca Boobies attack chicks (Anderson et al. 
2004). We conclude adult aggression is not the 
cause of angel wing on Clipperton, although we 
cannot rule it out entirely. 
Evidence .suggests there was a food shortage at 
Clipperton, which resulted in a major chick die-olt. 
and this clearly coincided with and may have 
contributed to the high incidence ol angel wing. 
We counted 167 dead chicks during our strip 
transect survey. These birds appeared to have died 
fairly recently < we estimated within the previous 3- 
4 wks) and were from the current cohort ol chicks 
still alive on the island. We divided the number ot 
dead chicks by the total number of live plus dead 
chicks on our transects, minus the AHA chicks 
(167/1.019 + 167 - 5) to estimate the number of 
chicks that had died recently. We extrapolated the 
resulting 14.1% mortality rate to the entire island 
and estimated a minimum of 1.703 Masked Booby 
chicks had died during the recent event. 
This was apparently part of an even greater die¬ 
off. Weimerskirch et al. (2009) counted 19.686 
active Masked Booby nests (36% with eggs, 64% 
with chicks) on the island in January 2005; thus, 
the 10,375 chicks we counted 2 months later 
represenled a loss ol 9,311 nests, a 47.3% 
reduction. The die-off occurred mainly during 
February because Weimerskirch et al. (2009) 
reported 'normal’ feeding during January and, 
when we were there in March, adult Masked 
Boobies were bringing heavy loads to lecd chicks. 
We infer the die-off was due to a food shortage 
because prc-fledging mortality among chick 
boobies is usually due to starvation ( Anderson 
1993). All previous large-scale nesting failures 
and chick die-offs documented at booby colonies 
have also been attributed to food shortages (e.g.. 
Dorward 1962. Schreihcr and Schreiber 1984. 
Anderson 1989). We know of no records of 
epizootic disease causing large scale mortalities in 
any sulid species (Nelson 2005:115). 
There are at least two reasons why boobies 
might have experienced reduced foraging success 
at Clipperton in early 2005. Masked Boobies, like 
many other seabirds in the eastern tropical Pacific 
(FTP), depend heavily on feeding schools of 
yellowTin tuna (Thuwms alhacares ) to drive prey 
to the surface and make it available (Au and Pitman 
1986, Ballance et al. 1997). Catch rates for 
yellowfin tuna in the FTP in 2005 were lower 
than average, and fish caught were of smaller size 
(1ATTC 2006); this alone could have resulted in 
reduced feeding opportunities at Clipperton. In 
addition, commercial fishermen in the ETP not 
only target the same tuna schools that boobies rely 
on. but often use feeding bird flocks to help them 
locate tuna schools (Perrin 1969, Au and Pitman 
1986). It may be significant that Weimerskirch et al. 
(2009) reported (hat up to seven luna purse seiners 
were present al Clipperton Island during their visit 
in January 2005. That a Beet of tuna purse seiners 
was operating in the waters around the island just 
prior to the time when the colony suffered a major, 
probably food-related, chick die-off. further em¬ 
phasizes the possible negative impact that tuna 
fishing can have on seabird foraging in the ETP 
(Ballance et al. 1997. Ballance and Pitman 1999, 
Weimerskirch et al. 2008). Further research at 
Clipperton Island could provide not only key 
information on the etiology of angel wing among 
bird populations, but could also shed some light on 
the possible impact of industrial-scale tuna fishing 
on a globally important tropical seabird population 
(Weimerskirch et al. 2008). 
Prepared skeletons of tw o Masked Boobies and 
the one Brown Booby with angel wing from 
