D.—ZOOLOGY ry, 
The parameres are unequal in size in the Fragmentati, but otherwise they 
are alike throughout the family. In most genera the parameres are free 
from one another except at the base, whereas in some they are united 
along one side, the one to the other. But the shape and chetotaxy of the 
parameres varies in different genera and different species, as does also 
the form and size of the zdeagus. It is most probable that the variations 
that occur would in no way affect céitus between many species in any one 
genus and, similarly, in many cases céitus between species of different 
genera would be possible, so that, although the idea underlying the 
systematic value of the armature is the mechanical isolation of the species, 
it seems as if the modifications are mostly evolutionary and of no vital 
importance. 
The function of the parameres is presumably as claspers during céitus 
and, if so, appearance suggests that there is a varying degree of efficiency ; 
but in this connection it is interesting to notice that in three genera— 
I have so far discovered no more—what appear to be sucker-hairs have 
appeared on these structures. These are of a simple form and may not 
be very efficient, but they suggest a developing line of evolution. In the 
genus I/ybius, all the seven British species possess these hairs, which are 
rather more rudimentary in one or two species—e.g., I. ater and obscurus. 
In Rhantus, out of thirteen species examined, only three possess them and 
in one, adspersus, they are so concealed among the simple hairs of the 
apical brush that it is impossible to imagine that they can function at all. 
Out of twenty-two species of Agabus examined, only one, chalconatus, 
possesses the suckers, which, in this case, form a very efficient-looking 
tuft near the apex. 
I have selected some of the main characters upon which the classification 
of the Dytiscidz is based and I have endeavoured to show that, in at 
least many of them, a progressive development can be recognised. These 
characters are not vital to their possessors, since various stages in their 
development exist side by side, so that natural selection can have had 
nothing to do with their progress. Such lines of increasing complexity 
are recognisable throughout the world of living things. In plants, for 
instance, the gradual reduction of the gametophyte and the evolution of 
the sporophyte is a progression through many families. The tendency to 
cluster flowers into heads, brought to perfection in the Composite, runs 
through a number of families, as does also the opposite tendency to isola- 
tion and specialisation of the individual flower. In animals, increase in 
size, repetition of form, development of horns and innumerable other 
tendencies, all appear to be of the same order as the progressive lines in 
the Dytiscide. 
The discoveries in connection with chromosome control of characters 
suggest that the orthogenetic tendencies may be the outcome of mutations, 
under control of mathematical laws and caused by external stimuli, 
although some authors look upon directional evolution as an inherent 
property of the organism like growth and reproduction. On the other 
hand, it seems possible that if function can cause variation in structure, 
these evolutionary lines may be responses to physiological activities, 
