70 THE ZOOLOGIST. 
from dentata make still further progress elsewhere, then D. 
hagenst could not be included any morevin the generic diagnosis 
of Dinarda, since the keeled border of the elytra of the latter has 
hitherto been regarded as essential. We must even define the 
whole group of Dinardini differently, since that keeled border of 
the elytra formed its most essential character! It has been 
objected by some defenders of the “‘ constancy theory” that the 
previous observations on Dinarda offer no evidence on behalf of 
the evolution theory, since here it is only the question of evolu- 
tion ‘‘ within the species.’’ In fact, it is perhaps better to 
reckon, as 1 showed more completely in 1896,* our four two- 
coloured Dinarda forms as races, rather than as species in the 
strictest systematic sense. But, in any case, they offer races, 
which stand at different stages in their evolution; D. dentata 
and mdrkeli, as far as concerns their constancy, have arrived 
much nearer to ‘“‘true species” than D. hagensi and pygmea. 
Further, indeed, as has been shown above, the keeled border of 
the elytra in the typical D. hagensi has disappeared ; but when 
in this evolution a character can disappear, which has hitherto 
been held as essential, not only for the species but even for the 
genus in question, even for the whole group of genera, then the 
objection that it is only a case of evolution within the species will 
fall to the ground. 
Let us now consider a similar process of differentiation, such 
as is the case which has furthered, and is still furthering, the 
evolution of our two-coloured forms of Dinarda, through the 
adaptation to different species of the genus Formica, extended to 
the adaptation of allied Dinardas to ant-guests of different genera 
and subfamilies. In this case, too, we get a phylogenetic under- 
standing for the differentiation of the genera Dinarda and Chitosa. 
Dinarda nigrita, which occurs in the Mediterranean province, 
which Cassey has lately raised to the new genus Chitosa, is in 
the highest degree probably sprung through the adaptation of a 
Dinarda-like ancestor to the Myrmicide Stenamma (Aphenogaster) 
. testaceopilo-nigrita, which is the actual host of Chitosa nigrita. 
The same essential differentiation process that we find in the 
evolution of Dinarda also takes place here, only that it already 
* « Dimarda-Arten order Passen”’ (Wien. Entom. Zeitung, xv. 4 und 5, 
Heft. pp. 125-142). 
