138 
Psyche 
[December 
vertently let slip into print a manuscript notation which was 
to have been subsequently altered. It is easier to be tolerant 
of this mistake than it is to agree with Emery’s treatment 
of the taxonomic status of missouriensis. I can see no reason 
why he should have assigned it to victima instead of to 
Mayr’s rrdnutissima. In 1870 Mayr had published a key to 
the New World species of Crematogaster in which he clearly 
distinguished between the characteristics of Smith’s victima 
and his own minutissima. Emery must surely have employed 
Mayr’s key and just as surely he should have been aware 
that missouriensis is more closely related to minutissima 
than to victima. The distinct areas of cephalic punctures 
which are present in victima are absent in the other two 
forms. In recent years the recognition of a number of addi- 
tional subspecies has considerably expanded the specific 
limits of victima. Even so the above contention can be de- 
fended. Still more peculiar is Emery’s disregard for zoogeo- 
graphical considerations. When he assigned missouriensis 
to victima the latter species was known only from Brazil. 
It would certainly have seemed more logical to consider the 
possibility of relationship with a species which had been 
found in the Gulf Coast region. As far as I can determine no 
one has ever questioned Emery’s judgment in the matter. 
Despite this I believe that missouriensis should be regarded 
as a northern race of minutissima. In addition there is a 
western race which is described below. 
Crematogaster (Orthocrema) minutissima thoracica 
subsp. nov. 
The subspecies thoracica differs from the typical form and 
the subspecies missouriensis in its distinctly more shining 
thoracic dorsum. In both the other two forms the dorsum 
of the promesonotum bears, in addition to longitudinal 
rugse, a number of fine and fairly close-set punctures. These 
punctures, while not dense enough to produce an opaque 
appearance, dull the surface to a considerable extent. They 
are not present in the subspecies thoracica. In addition 
thoracica usually lacks longitudinal rugse on the pronotum 
and, when they are present, they appear to form a wavy 
border at the extreme edge of the pronotum. The longi- 
