1974 ] 
Peck — Catopocerus 
395 
using bait. The salamander which obtained the beetle as food un- 
doubtedly did so from a small and/or inaccessible population in or 
near the cave. Catopocerus are not commonly eaten by terrestrial 
cave salamanders. I have examined the digestive tract contents of 
433 salamanders from caves within the range of Catopocerus in the 
southeastern U. S. (Peck, 1974: Richardson and Peck, in MS) and 
no other Catopocerus have been found. 
Evolution and Distribution 
We may assume that the genus had a past transcontinental range. 
The present division of the generic range into the ecologically iso- 
lated eastern and western forested regions may date from the Mio- 
cene when drying conditions caused a retreat of the Neotropical 
flora from western North America. This range division was probably 
widened with the later Pliocene arid trend associated with the north- 
ward invasion of the Madro-Tertiary geoflora which gave rise to 
the recent xerophytic vegetation (Axelrod, 1958). 
Since this division, there is little that can be said of speciation and 
distributional events for the eastern species. A more rich history is 
evident in the western species, but will be discussed in a later paper. 
The three eastern species with limited distributions ( C . hamiltoni, 
C. alabamae, and C. jonesi ) probably experienced a contraction to 
their present ranges, from a wider ancient range, in connection with 
Pleistocene climatic events. None of these species is closely related 
to the other. 
C. applachianus and C. ulkei, however are similar enough to have 
been derived from a single common ancestral species which may have 
inhabited the Appalachians from at least Pennsylvania to North 
Carolina. A division of the range into two main units during an 
early interglacial may have allowed C. ulkei to diverge in the north 
from Virginia perhaps to Pennsylvania, and allowed C. appalachianus 
to diverge in the south in North Carolina. C. ulkei may have then 
dispersed southward during the Illinoisan glaciation to southwestern 
North Carolina. These populations may have become isolated and 
somewhat divergent during the Sangamon interglacial. Other overlap 
of the ranges of the two species may have occurred by dispersal in the 
Illinoisan, as well as the Wisconsinan. However, judging from the 
lack of other geographic variation, most of the present distributions 
were probably gained in the Wisconsinan. This was the time when 
the larger and perhaps more vagile C. appalachianus moved north 
into Virginia and West Virginia. The populations of this species in 
