February 18, 1898.] 



SGIENGE. 



247 



France, his attention having been called to the 

 species by extensive damage in the Vendee 

 during 1894. He shows that, although previous 

 writers in many countries have given three as 

 the maximum number of annual generations, 

 in France there may be developed, under the 

 most favorable circumstances, six such annual 

 generations. The most part of these, however, 

 are partial, and the most incomplete are the 

 third, fourth and sixth. He has shown that 

 there is a considerable retardation of develop- 

 ment where humidity is lacking, dryness hav- 

 ing been responsible for a retardation of two 

 months. An interesting section on natural 

 selection concludes with the statement that, 

 far from being adapted to climatic conditions 

 by natural selection, the species is perpetuated 

 in spite of obstacles placed in its way by ex- 

 terior conditions, and that it overcomes these 

 obstacles only by its fecundity and by the great 

 variability of its biologic cycle. Careful studies 

 are given of other species of Cecidomyiidse 

 affecting grasses and grains in Europe and else- 

 where, and especially of Cecidomyia avense, a 

 closely allied form which the author has differ- 

 entiated from the Hessian fly. 



The observations made on the biology of par- 

 asitic insects are nothing less than remarkable. 

 The larval development of internal feeding in- 

 sect parasites is, of course, very diflBcult to 

 observe. Few observations of value are re- 

 corded. The well known studies of Ganin on 

 certain Platygasters were the earliest. The 

 studies by Bugnion on the structure and life 

 history of Encyrtus fuscicollis, an internal para- 

 site of the European Hyponomeuta cognatella, 

 are the only ones of importance which have 

 appeared since Ganin. All of the species of 

 the genus Trichacis are parasitic within the 

 bodies of Cecidomyiid larvae, and the genus is 

 closely related to the form studied by Ganin. 

 According to Marchal the first larval form of 

 T. remulus corresponds to the type of the 

 curious cyclops-like larvse studied by Ganin, 

 and which certain authors regard as an adap- 

 tive form, while others see in it an ancestral 

 form. The post- embryonic development, ac- 

 cording to Marchal, is as follows : 



When they are young and motionless, and 

 have not issued from the cysts which contain 



them, these larvse are always lodged in the inte- 

 rior of the nervous system of the host larva, and 

 there they bring about alterations and prolifera- 

 tions of a very curious character. The most 

 frequent is at the posterior extremity of the 

 nerve chain, where the cyst of the parasite is 

 formed. This extremity spreads out into an 

 enormous bouquet of club-shaped giant cells, 

 which alone fills the larger part of the body 

 cavity of the host. The larva of the parasite 

 is lodged in a cyst filled with liquid, the cellu- 

 lar structure of which, with broad, polygonal 

 contour, seems to indicate an amniotic envelope 

 in a condition of retrogression. All around this 

 membrane the giant cells are grouped. These 

 exist not only in the immediate neighborhood 

 of the cyst, but all the surrounding region of 

 the nerve chain seems to have undergone the 

 same degeneration and growth of giant cells. 

 The youngest cells are hyaline and present a 

 fibrinous, longitudinal structure. The oldest 

 cells are filled with fatty globules and become 

 entirely opaque. The giant cells increase and 

 isolate vesicles, which separate and fall into 

 the body cavity in the form of protoplasmic 

 spherules, which are absolutely characteristic. 

 When one dissects a Cecidomyiid larva under 

 the microscope he can be sure, if he sees these 

 spherules floating in the liquid, that there are 

 in the preparation one or more larvae of this 

 parasite. The localization of the larvse of the 

 Trichacis in the nerve chain or in the nerves 

 of the larva presupposes that the parasite 

 pierces the egg or the young larva upon the 

 median ventral line at the time when the 

 nervous system has not begun to branch and 

 is concentrated in a single ventral baud. 

 The mass of giant cells accumulate in them- 

 selves evidently the nutritive material neces- 

 sary to the parasite. They are a kind of in- 

 ternal animal gall, developed by the presence 

 of the parasite. The Trichacis, in the condi- 

 tion of the cyclops-larva, waits in its cyst until 

 the tissues which surround it have submitted 

 to the transformations by which it profits later 

 for its food; then, when the host larva, ex- 

 hausted by its presence, is transformed into a 

 sort of a sac filled with giant cells, it issues 

 from its cyst to devour the accumulated ma- 

 terial, which, probably, has nutritive qualities 



