158 Dr. E. Korschelt on the 
might in its case expect to find free-roving larve (in fresh 
water). Chiefly on this account I endeavoured during the 
summer of last year to elucidate the hitherto unknown repro- 
duction and development of this mollusk. At that time my 
investigations were undertaken too late; but with the com- 
mencement of spring of the present year I recommenced my 
observations and persevered with them until I finally suc- 
ceeded in deciding this question. I would remark at the 
outset that in the following pages I have no intention of 
giving a detailed account of the development of Dredssena. 
My observations were, as I have already hinted, directed 
chiefly towards the one point of the presence or absence of 
free-roving larv. Moreover they were really intended to 
partake of an orientating character, in order to determine the 
time and manner of the reproduction of Dreissena, and thereby 
to facilitate a subsequent minute investigation of the deve- 
lopment, since this was not possible for me this year on 
account of other tasks which could not be postponed. Mean- 
while it is already possible to determine the leading features 
of the development of Dred¢ssena, so that a communication on 
the subject will not be unwelcome. 
As a general rule development by means of free-roving 
larve provided with a velum is characteristic of the Lamelli- 
branchs. Nevertheless an exception is formed by the fresh- 
water mussels. As is frequently the case in freshwater forms, 
e.g. in the Annelids, Turbellarians, &c., these creatures adopt 
a direct development with the freshwater existence, and 
abandon the original method by means of free-roving larve. 
Thus we see that the freshwater mussels have already 
assumed the adult form on leaving the mother (Cyclas, Pisi- 
dium) or else pursue a course of development in which larve 
indeed appear, but which is at once recognizable as having 
undergone secondary modification and considerable adaptation 
to the present mode of life of the mussels (Unionide). 
Cyclas and Pisidium exercise a brooding-process, since they 
form a kind of brood-pouches within the gills, in which the 
eggs develop until the formation of the pertect animal is 
almost complete. Nevertheless these mussels pass through 
a stage in which the embryo is to be considered equivalent 
to the free-roving larva, which has been compared to the 
Trochophora of the Annelids. The velum of the larve of 
marine mussels is indicated by an area of cilia in the embryo 
of Cyclas ; in this case therefore we may speak of a rudimen- 
tary velum*. Moreover it possesses a primitive kidney, the 
* F. Ziegler, “ Die Entwicklung von Cyclas cornea, Lam.,” Zeitschr. f, 
wiss. Zool. 41 Bd., 1885. 
