428 
PACIFIC SCIENCE, Vol. V, October, 1951 
Vegetation: For most species vegetation was 
a minor factor since it was entirely absent 
in 41 per cent of the collections. This is not 
surprising in view of the temporary nature of 
so many of the habitats. There was a striking 
correlation between the type of vegetation 
and a particular species in only one case: C. 
squamosus was found with vegetation in 85 
per cent of the collections and, of these, 61 
per cent were with green algae. 
Nearly every species was found within 
a wide range. Alkaline readings were relative- 
ly few, and for no species was the mode or 
average alkaline. In general, there was little 
or no correlation between any one species 
and certain narrow values. 
Species Association 
In Table 4 the species represented by less 
than 10 collections are excluded and all asso- 
ciation values less than four are added and 
lumped under "others.” 
Conclusions from this table substantiate 
those of the habitat table. That is, surface 
water species have high association values 
for one another, and, likewise, container 
species have high association values for one 
another. The separation of surface water 
species and container species is quite apparent. 
However, as with the habitat groups, there 
is overlapping. The following, which were 
found both in surface and container habitats, 
are likewise found with both surface and con- 
tainer associates: C. papuensis, A. novalbitarsis, 
and A. papuensis. These three are clearly in an 
intermediate position. 
For almost every species the number of 
pure collections exceeded the number of as- 
sociations with any other particular species. 
A. kochi. Arm. milnensis, and Arm. hreinli had 
the highest values for pure collections. 
It is usually stated that predatory species 
are found more commonly in pure collections. 
My figures were not unusually high for the 
predatory species, Toxorhynchites spp. and 
Culex halifaxi. Indeed, in the case of halifaxi, 
the value 18 was surprisingly low. 
The association of a particular species with 
any one other exceeded 20 in very few cases. 
It is believed that the concept of indicator 
species is in no case substantiated by these 
data. Examples of the highest associations 
were: C. papuensis with C. pullus (27), C. 
fraudatrix with An. farauti (26), C. halifaxi 
with C. pullus (25), and A. vexans with An. 
punctulatus (25). 
REFERENCES 
Anonymous. 1943. Gazetteer to maps of New 
Guinea. 272 pp. U. S. Army Corps of En- 
gineers, Map Service, Washington, D. C. j 
Arnett, Ross H., Jr. 1950. Notes on the 
distribution, habits, and habitats of some 
Panama culicines (Diptera: Culicidae). 
Part IV. Ecology. N. Y. Ent. Soc. Jour. 
58(2): 99-115. 
Bates, Marston. 1949. The natural history 
of mosquitoes . xv+379 pp. Macmillan, New 
York. 
Belkin, John N., K. L. Knight, and L. E. 
Rozeboom. 1945. Anopheline mosquitoes 
of the Solomon Islands and New Hebrides. 
Jour. Parasitol. 31(4): 241-265. 
Bonne-Wepster, J. 1938. Geographic re- 
lationship of the non-anopheline mos- 
quitoes of New Guinea. Dienst der Volks- 
gezondh. Nederland. Indie Meded. 27(1, 2): 
206 - 212 . 
Bonne-Wepster, J., and S. L. Brug. 1932. 
The sub genus Stegomyia in Netherland 
Indie. Geneesk. Tijdschr. v. Nederland. Indie 
72(2): 39-119. 
1937. Nederlandsch-Indische Culi- 
cinen. Geneesk. Tijdschr. v. Nederland. Indie 
77(9, 10): 515-617. 
1939. Larven van Nederlandsch- 
Indische Culicinen. Geneesk. Tijdschr. v. 
Nederland. IndiT79(20): 1218-1279. 
Boyd, Mark F. 1930. An introduction to 
malariology. xiv+437 pp. Harvard Uni- 
versity Press, Cambridge, Mass. 
