320 
PACIFIC SCIENCE, Vol. XXII, July 1968 
tatus, 600-800 meq/1; L. marinus 700-900 
meq/1) is sufficient to allow them to exist on 
sea water (500-550 meq/1) even if their diet 
was restricted to isotonic marine invertebrates. 
Brewster (1883) even reported a captive Kitti- 
wake ( Rissa tridactyla ) which refused fresh 
water but drank salt water eagerly. 
Even if gulls did require a source of fresh or 
brackish water, this is available on most of the 
Hawaiian islands and on several of the Line 
islands. Many of the main Hawaiian islands 
have a high annual rainfall and permanent 
bodies of fresh or brackish water. Palmyra 
Island, although a low-lying atoll, has over 100 
inches of rain a year, and there are enough pools 
of fresh water to support a population of toads 
( Bufo marinus ). Washington Island also has 
over 100 inches of rain which maintains a large 
fresh water lake in the interior of the island. 
Thus we believe that the known physiological 
capabilities of gull salt glands are sufficient 
evidence to reject the theory that the salt glands 
are a limiting factor in the adaptation of gulls 
to these islands. 
The failure of gulls to colonize tropical 
islands remains a puzzle. We have presented 
evidence that a number of species may winter 
in the central Pacific, and we have questioned 
several theories concerning their failure to 
establish breeding colonies in this area. Part of 
the answer may lie in the ease with which gulls 
arrive at, and presumably depart from, the 
islands. With the exception of the Laughing 
Gull, all of the species recorded from these 
islands breed at more northerly latitudes. Rather 
than being accidental (i.e., lost) birds, they may 
represent irregular migrants. Perhaps the ques- 
tion is not why gulls have failed to colonize 
tropical islands, but why migrants rarely estab- 
lish breeding populations on their wintering 
grounds. 
SUMMARY 
Data collected from February 1963 to May 
1966 by the Pacific Ocean Biological Survey 
Program on gull distribution in the central 
Pacific are presented. The 41 specimens and 
over 50 sight records indicate that Larus argen- 
tatus and L. glaucescens are the most frequent 
visitors in the Hawaiian Islands, L. atricilla and 
L. pipixcan in the Line Islands. Wind drifting 
is presumed to be primarily responsible for the 
arrival of gulls on central Pacific islands. 
Gulls do not necessarily arrive at the islands 
in poor condition as previously believed, nor do 
they seem to remain on any one island for long. 
Garbage dumps are an important food source 
in the Hawaiian Islands but gulls in the Line 
Islands survive without access to any but local 
foods. Evidence is presented to show that a 
gull’s salt glands are not a limiting factor in 
its survival on central Pacific islands. 
It is proposed that gulls are irregular winter 
visitors to the Hawaiian Islands and that most 
of them return to their nesting grounds. 
REFERENCES 
Amadon, Dean. 1965. Distribution of gulls — 
some further considerations. Elepaio 25:95- 
96. 
Bent, Arthur Cleveland. 1921. Life His- 
tories of North American Gulls and Terns. 
U. S. Natl. Mus. Bull. 113, 337 pp. 
Bourne, W. R. P. 1965. Observations of sea 
birds. The Sea Swallow 7:10—39. 
Bryan, Edwin H., Jr. 1958. Check-list and 
Summary of Hawaiian Birds. Books about 
Hawaii, Honolulu, Hawaii. 22 pp. 
1962. Larus argentatus vagae collected 
on Midway Islands. Elepaio 23:28. 
1964. Gulls on Pacific islands. Elepaio 
24:53-54. 
Brewster, William. 1883. Notes on the birds 
observed during a summer cruise in the Gulf 
of St. Lawrence. Proc. Boston Soc. Natl. 
Hist. 23:364. 
Clapp, Roger B., and Fred C. Sibley. 1967. 
New distributional records for the Phoenix 
and Line islands. The Ibis 109:122-125. 
Clapp, Roger B., and Paul W. Woodward. 
In press. New records of birds from the 
Hawaiian Islands. Proc. U. S. Natl. Mus. 
Frings, Hubert M. 1965^. Absence of gulls 
on Pacific islands — some speculations. Elepaio 
25:51-54. 
1965&. Distribution of gulls — still fur- 
ther considerations. Elepaio 25:96-98. 
