52k ae On Aquatic Carnivorous Coleoptera or Dytiscide. 
Cybistrini, where not only is the dilatation of the three joints in question on the 
front feet enormous, but it is accompanied by a shortening and coadaptation which 
render them in fact a single organ or saucer. Neither do the males show an 
incrassation of the basal joints of the tarsi such as exists in Agabini and Laccophilini, 
where an enlarged sole is gained by an increase in the whole circumference of the 
joints, the increase being in fact as great in the vertical as in the horizontal direction 
of each incrassate joint. 
Putting aside these sexual differences in the tarsi, other interesting variations 
in the basal joints, especially in the outer of them—the third joint—may be noted ; 
this is almost subquadrate or truncate-cordate in certain forms (Hyphydrus, Macro- 
vatellus, e.g.) with the upper surface bearing a short groove at the extremity for 
the insertion of the basal joint; while in others (many North American species of 
Hydroporus of groups 1 and 2), it is deeply grooved, or indeed cleft nearly to the 
base, so as to exhibit elongate lobes, the terminal joint being inserted near to its 
base. In this case the structure is essentially that of the tetramerous Coleoptera, 
as represented by the Curculionidee, Phytophaga, Longicornes, &c. : and there is every — 
reason to believe that this modification of structure is essentially similar in its 
functional value in the two cases. The Coleoptera which walk or run on the 
surface of the earth have slender cylindric tarsi, destitute of clothing beneath or 
bearing only a few rather rigid setee, while those groups which live much on foliage 
have the three basal joints of the tarsi dilated to form a sole, which is clothed 
beneath with a peculiar pubescence, while the terminal joint is not used in walking 
but is inserted so that the claws terminating it are held up from the surface ; so m 
the Hydropori mentioned we have good grounds for supposing that the species are 
much in the habit of frequenting and walking on aquatic plants, while in the 
other tribes of Dytiscidee the front and middle tarsi (when not modified for sexual 
purposes) have essentially the structure of the Carabide. As another striking 
instance of the relation between the form of the tarsi and these habits of the species, 
I may be allowed to mention the genus Stenus of the Staphylinidee, some species 
of which run swiftly on earth and mud and have filiform elongate feet, while others 
which live much on plants have the tarsi broad with a lobed penultimate joint and 
the surface clothed beneath with fine pubescence. In these Steni the tarsi however 
remain five-jointed and have not as in these specialized Hydropori acquired the 
fully characteristic structure of the tetramerous Coleoptera. In the fact therefore 
that the structure of the tarsi of some Dytiscide differs fundamentally but little, 
scarcely atall, from that of some Curculionidee, we have clearly a clue to deciphering 
certain of the environments under which they have been developed. 
It is worthy of note that where the true fourth joint of the tarsus is conspicuous, 
as in Sternopriscus and Necterosoma, it is accompanied by an elongation of the 
fifth joint, and this elongation is different sexually, beg always somewhat, often 
greatly, more developed in the males. A striking example of this may be pointed 
