182 Journal New York Entomological Society. [Voi. xx. 



and a partial loss of hair. The shape of the body is also deeper and 

 narrower than in the Dytiscidas, a form better fitted for movement 

 through the water, but lacking the means of rapid propulsion. 



The swimming efficiency is very feeble as compared with the more 

 specialized Dytiscidse. Instead of a soldering fast of the posterior 

 coxae and the formation of a solid joint, as in the Dytiscidje, there is a 

 remarkable plate-like development of the coxae. By this means the 

 hind legs are moved in one plane and their efficiency as swimming 

 organs increased though their horizontal range of movement is, if 

 anything, lessened. There is also a developmenut of long fringes of 

 hairs on the tarsi of all the legs and the tibia of the first and second 

 pairs. These fringes are in some measure supported by short, stiff 

 setae (Fig. 14). There is no shortening of the pr^^ximal joints of the 

 legs, nor the lengthening or flattening of the tarsi as in the Dytiscidae. 



The modifications for the securing and storage of an air supply 

 have already been fully discussed. 



Feeding Habits. — Little is known regarding the feeding habits of 

 many of the adult aquatic Coleoptera. In most cases it is taken for 

 granted that they are carnivorous. Unfortunately I have not much 

 information regarding the adult Haliplidae. In my aquaria H. rufi- 

 collis, H. conncxus, and H. cribrarius were observed feeding greedily 

 on the contents of Nitclla, the softer portions of Cliara and other fila- 

 mentous algae. They crush open the Nitclla stems with their 

 mandibles and devour the contents, leaving the cell wall. H. 

 crihrarins was especially greedy and a dozen or so would destroy con- 

 siderable quantities of Nitclla in a very short time. Whether they 

 feed on animal matter or not I am not prepared to say, though I have 

 observed them destroying dead specimens of their own kin in my 

 aquaria. This may have been due to the lack of sufficient vegetable 

 food. For in the case of H. cribrarius and H. conncxus it was found 

 that when freshly placed in aquaria with Nitclla, they greedily at- 

 tacked the young and tender tips. H. ruficollis also fed readily on the 

 smaller filamentous algae and Nitclla. 



Egg-laying Habits. — Nothing has been written regarding the eggs 

 or egg-lying habits of any species belonging to this family. During 

 the past two years I reared two species of Pcltodytcs and one of 

 Haliplus. In the genus Pcltodytcs the eggs are fastened to the strands 

 of filamentous algae, particularly Nitclla and Cliara, while in Haliplus 



