Vor. II, Pr. IT] WHEELER—GALAPAGOS ISLANDS ANTS 263 
of South America, but no such difference in affinities can 
be detected among the ants, because most of the neotropical 
species to which the Galapagos forms are most closely allied, 
are very widely distributed and because our knowledge of 
the ants of Ecuador, Peru and Chili is less complete than 
that of the ants of the West Indies, Central America and 
Brazil. I suspect that a similar dearth of knowledge of the 
western South American species of other groups may ac- 
count for the high percentage of West Indian and Central 
American elements recorded by several authors as obtaining 
among the Galapagos organisms, as e. g., by Banks, who 
mentions only five of the 54 Galapagos spiders as being 
known from Western South America as compared with 14 
from Central America, Colombia and the West Indies. 
Special interest attaches to the two species of Camponotus, 
macilentus and planus, as each of them is represented by 
distinct varieties on each of several of the larger islands. 
In fact, Albemarle Island possesses two varieties of macilentus 
and Indefatigable Island two of planus. The distribution of 
the various forms is shown in the following table: 
Islands. 
macilentus, typical.... Charles....... planus, typical 
Chathamt/222. 3 var. peregrinus 
var. narboroénsis..... Narborough... var. fernandinensis 
ee GUA Raa ccRaal Nlbemiarel eaesar: var. 1sabelensis 
VA DUICEMMIIS ty t 
var. duncanensis...... Dincan ys. ess var. pingonensis 
Indefatigable. . NEM TEAC 
var. SaNTACTUZENSIS 
VAT OODEHSIS ON Wises: Flood eet. 
var. barringtonensis... Barrington.... var. fidelis 
Valk) TOCOUCHSIS 1 Vad) TJENOSTESS A aN ae 
var. bindloénsis....... Bindloeas Oa. 
NaI Me anne ein atin ta Mower senile i: 
Although a similar “harmonic” distribution has been 
observed in birds, reptiles and plants, the only group of 
invertebrates in which it has been recorded, is, to my knowl- 
edge, the Acridians. Snodgrass cites three species of 
