ON THE PRESENCE AND ABSENCE OF COCOONS 
AMONG ANTS, THE NEST-SPINNING HABITS OF 
THE LARVZ AND THE SIGNIFICANCE OF THE 
BLACK COCOONS AMONG CERTAIN 
AUSTRALIAN SPECIES.* 
By WrLtt1AM Morton WHEELER. 
It has long been known that in three of the five subfamilies 
of ants, the Doryline, Myrmicine and Dolichoderine, the larve 
Spin no cocoons before pupation, and therefore remain nude, 
or uncovered in this instar. There is apparently no exception 
to this rule in these three subfamilies. The cocoon is present, 
so far as known, in all members of the subfamily Ponerinae, 
with the single exception of a small African ant, Dzscothyrea 
oculata, and is lacking among the Camponotine only in a few 
whole genera and a few sporadic species of other genera. 
The pupe of the genera Prenolepis and Cicophylla are 
naked, and in some species of Formica (e. g. the circumpolar 
F. fusca L. and its vars. subsericea Say, neorufibarbis Emery, 
etc. in North America) one often finds only naked or both 
naked and cocooned pupze in the same nest. This is also 
the case, though much more rarely, in certain species of Lasius. 
A few species of the large paleotropical genus Polyrhachis also 
lack the cocoon in the pupal stage, as I shall show in the sequel. 
These facts naturally suggest the inference that the cocoon, 
inherited from the ancient wasp-like ancestors of the ants, has 
been retained in the most primitive subfamily, the Ponerinae, 
from which, by common agreement, the four other subfamilies 
are descended, and that it has been lost in three and is tending 
to disappear in the fourth of these. Certain authors, indeed, 
seem to believe that the disappearance has been comparatively 
recent in the phylogenetic history of the Formicide. In this 
sense I interpret the following remarks of Janet (1896): ‘‘We 
are no doubt witnessing in the ants the disappearance of this 
protective envelope, which is rendered unnecessary by the 
incessant care devoted to the progeny. From the point of 
view of the evolution of instinct, it is interesting to note that 
*Contributions from the Entomological Laboratory of the Bussey Institution, 
Harvard University, No. 95. 
323 
