D.—ZOOLOGY 71 
Someone is certain to challenge my statement that these species- 
characters are, for the most part, non-vital but, as a fact, comparatively 
few of the characters upon which we rely for classifying the water-beetles 
can be recognised as vital for the survival of their possessors. In the 
general characters there is a uniformity everywhere among water-beetles 
because the obvious features are, first those common to all Coleoptera, 
secondly those which belong to their land-relations, and thirdly those 
which place the species in one or other of these two groups, Hydradephaga 
or Hydrophilidez. ‘These latter characters are such that we recognise 
them as eminently suited to the habits of their owners and we call them 
‘adaptive.’ The form of the adult is more or less streamlined, and in 
almost all the Hydradephaga the body is smooth and free from projecting 
hairs. The Hydrophilidz are also smooth except for a felting of hairs 
on parts of the underside. The hind legs of both groups are adapted 
for swimming and, in both groups, an air-reservoir exists which enables 
the beetles to remain under water for long periods. The larve also show 
certain adaptive characters. 
If we now analyse these characters we find that, although in some 
cases it appears that there have been special modifications of structure 
to enable the insect to live in the water, in other cases a change of function 
explains the adaptation. The modification of the hind legs into oars may 
be regarded as a modification of structure, since it is almost inconceivable 
that such legs could be of much use out of water. Moreover, the Hydra- 
dephaga include four families which, according to current opinion, have 
not had a common origin and yet all four have developed swimming legs 
in varying degrees, three of them on much the same lines. The enlarge- 
ment of the basal segment or coxa of the hind leg is to be found in various 
land insects, but the special form of the postcoxe and their fusion to the 
body has developed independently in three of these families. In one, 
the Pelobiidz, which Sharp regarded as having only recently taken to 
the water, the changes in this structure have not advanced far, but, in 
the other two, the coxe are greatly enlarged at the expense of the ventral 
body plate, the metasternum, and in almost all the Dytiscide the two 
parts form a perfectly smooth surface and present the minimum resistance 
in the water. 
The Hydrophilidz possess clubbed antenne which are of use in 
connection with respiration. These beetles carry their reserve of air not 
only under the wing-cases, like the Hydradephaga, but also on the felting 
of hairs on the under side of the body and, when renewal of air is necessary, 
the beetle comes to the surface and breaks the surface film by pushing 
up one of the antenne. This brings the ventral air-film into communica- 
tion with the atmospheric air and a renewal of the air-supply takes place, 
not only in the two communicating reservoirs but also in the main air 
tubes or trachea which open into these reservoirs. Now the clubbed 
antenna is a structure which occurs in a number of families of beetles, at 
one time all regarded as having had a common origin and classed together 
as the ‘ Clavicornia,’ but now recognised as belonging to different groups, 
indicating that the clubbed antenna has appeared more than once in the 
