396 A HISTORY OF EECENT CRUSTACEA 



more than one host. This, if accepted, makes the distinc- 

 tion of species particularly easy. It is, however, a rather 

 wide generalisation. It appears to imply that the larval 

 forms always settle on the same species as that occupied 

 by their parents, or perish unless they do. Yet, if the 

 larvae of a single brood were dispersed upon hosts of nearly 

 related species, one might expect that to those placed in 

 slightly novel circumstances, some difference of habit would 

 result rather than destruction. Still the important point 

 would remain that in each species of host the parasite is 

 distinct in character, and in favour of their view that it 

 is also distinct in species, Giard and Bonnier urge that 

 ' often among closely related Epicarids there are consider- 

 able physiological differences, sometimes even morpholo- 

 gical differences relating solely to the male or the embryo, 

 differences too important to be attributed simply to the 

 difference of host.' 



In the relations between the Epicaridea and their hosts 

 a very singular circumstance has recently been brought to 

 light. Rathke and other observers had commented on the 

 unexplained peculiarity that the infested prawns and 

 crabs all appeared to be females, and moreover sterile 

 females. De Haan, as heretofore remarked, not unfre- 

 quently records sterile females of a different form from the 

 fertile females of the same crab. Giard and Bonnier have 

 shown that an infested female is not always absolutely 

 sterile ; but a few years earlier Professor Giard made the 

 extremely interesting discovery that the parasites attack 

 males just as freely as the other sex. Only, under the 

 influence of the invader the distinguishing characters of 

 the male are hindered from development, and a sort of 

 intermediate appearance is permanently retained. 



Family 1 . Microniscidce. 



This includes the least degraded forms, corresponding 

 in general aspect to the second larval form of the other 

 families. They are parasitic on Copepocla. 



Hicroniscus, Fritz Miiller, 1870, is the only genus. 



