176 
PACIFIC SCIENCE, Vol. IX, April, 1955 
measurements of his specimens appear to be 
larger than those taken on Upolu and Manono 
but of similar length in maxillary toothrow. 
The Samoan rats seem to lie somewhat on the 
small side of his total series. 
While the rats on Manono appear to be 
smaller than those on Upolu, comparison 
with R. hawaiiensis measurements as given by 
Miller (1924) shows the different proportions 
of the Hawaiian rat which has a larger head 
in proportion to the size of the body. Rattus 
maorium , a synonym for R. exulans , is found 
in New Zealand and forms the southernmost 
example of the group. The type skull (Mar- 
pies, 1954) is rather larger than the Samoan 
forms with a longer toothrow, and slightly 
different proportions. 
R. exulans is distributed rather evenly over 
Polynesia, a fact which indicates distribution 
by man. It was already widespread before the 
coming of Europeans and R. rattus and R. 
norvegicus . Since the time of distribution there 
has probably been very little mixing, for 
ocean-going canoes do not appear to have 
persisted in their extreme form, and thus in 
fairly recent times many isolated populations 
of rats have been established. Variations in 
colour, size and proportions would be ex- 
pected to arise among the constituent popu- 
lations of such an assemblage finally giving 
rise to separable species. However, this has 
not yet occurred in the Pacific rats. The dis- 
covery of differing populations at two close 
areas, as at Manono and Upolu, show how 
complicated the picture may become, making 
general trends difficult to assess. The effect 
may be exaggerated in this case for Manono 
is a small islet lying three miles from the 
Upolu coast and within the reef and therefore 
only small canoes are used for transport. This 
may have had the effect of a wider geograph- 
ical separation as the chance of introduction 
would be less. 
The taxonomic position of the many popu- 
lations of exulans type rats remains in doubt. 
Mayr tends to place bird populations across 
the Pacific in the same species. This could 
also be done with R. exulans , perhaps leaving 
the Hawaiian and Hebridean rats as sub- 
specific forms. The geographical races occur- 
ring on each island are not sufficiently im- 
portant for separate nomenclature. But it is 
possible that with long series to work from, 
factors useful for comparison of larger groups 
of rats may be found. The present series 
constitutes a starting point for this work. 
SUMMARY 
Specimens of R. exulans were collected in 
Western Samoa, 12 from the island of Upolu 
and 32 from the islet of Manono. Statistics 
of the populations are given and significant 
differences in several characters were found, 
though the islands are separated by only 3 
miles of sea. 
REFERENCES 
Miller, G. S. 1924. The characters and prob- 
able history of the Hawaiian rat. Bernice P. 
Bishop Mus ., Bui. 14: 3-6. 
Tate, G. H. H. 1935. Rattus and Mus from 
Pacific Islands. Amer. Mus. Nat. Hist., Bui. 
68: 145-178. 
Stone, W. 1917. The Hawaiian rat. Bernice 
P. Bishop Mus., Occas. Papers 13(4): 253— 
260, pi. 8. 
Stokes, J. F. G. 1917. Notes on the Hawaiian 
rat. Bernice P. Bishop Mus., Occas. Papers 
3(4): 261-271, pis. 9-15. 
Turbet, C. R. 1925. Rats and coconuts. Fiji 
Dept. Agr. Cir. 5(2): 97-99. 
Marples, R. R. 1954. Note on the type 
specimen of the Maori rat. Roy. Soc. New 
Zeal. Trans, and Proc. 82: 703-704. 
