On Aquatic Carnivorous Coleoptera or Dytiscide. 955 
ineequalis we see the other extreme in these respects, and we find a small and 
acuminate prosternal process, with very approximate coxze and exposed posterior 
coxal articulations. If we take a third form of the Hydroporides, Dytiscus 
duodecimpustulatus of the genus Deronectes for example, and continue our com- 
parison, we find in it a general shape and swimming legs more adapted for 
aquatic locomotion ; the head is completely without margin and the labrum is 
quite exposed. The middle and hind legs are less approximate than in D. ineequalis, 
the prosternal process is much broader and the posterior coxal cavities more pro- 
tected. As the result of the comparison of the three species we find then, that 
having regard to the general form and the development of the swimming legs, a 
supposed line of descent, thus—1. Hydrovatus; 2. Ceelambus; 3. Deronectes ; 
while if we look to the distance of the coxz, the form of the prosternal process, 
and the protection of the hind coxal cavities, the order would be—1. Hydrovatus ; 
2. Deronectes ; 3. Coelambus ; and if again we look to the structure of the head, we 
find that Hydrovatus is the intermediate form, and the line of descent would be 
-——1, Celambus (J). ineequalis) ; 2. Hydrovatus; 3. Deronectes. It appears, there- 
fore, that a natural classification in which all the structures are dealt with gives no 
support to the idea of a genetic relationship between these three forms; while in 
an artificial classification—a classification in which some one or two characters 
should be taken into account to the exclusion of others—the three forms would 
stand in different relationships to one another according to what point of structure 
should be selected as the basis of the classification. It does not however follow, 
that in a natural classification it would be incorrect to speak of any one of these 
three species as being higher than the others, for if we were acquainted with all 
the points of structure, and knew accurately their functional value, we would be 
able to assign to each organ a numerical value, indicative of its grade of develop- 
ment (as for instance, in the case of the swimming legs, 1 to Hydrovatus, 3 to 
Coelambus, and 5 to the Deronectes), and by adding all these together we would 
be able to say with certainty which was absolutely the higher form; but because 
we could do this it would not in the least follow that the lower form was ancestral to 
the higher. 
This is equally applicable to the relations of the anomalous Pelobius and 
Amphizoa to the other Dytiscide ; although they are less highly developed as 
Dytiscidee there does not appear any reason for supposing them to be truly 
ancestral forms to their more highly developed allies, and thus the fact that certain 
of the lower forms have numerous points in common does not indicate a genetic 
relationship between them; the similarities found are indeed indicative not of such 
relationship, but rather of similarity in the stage of development of some one or 
more of the points of structure. 
TRANS. ROY. DUB. SOC., N.S., VOL. II. 6G 
