PROPAGATION OF FRESH-WATER MUSSELS. 619 
just described is absent and the eggs are either not agglutinated at all or are 
only loosely held together in a slightly viscid substance. 
The extent to which the gills are modified as a marsupium varies greatly in 
different genera. In many, as in Anodonta, Pleurobema, Strophitus, Symphynota, 
and Unio, the entire outer gill of each side is thus utilized and when filled with 
embryos or glochidia, presents a much swollen, padlike appearance. In others, 
for example, Lampsilis, Obovaria, Plagiola, only the posterior portion, varying 
from one-third to two-thirds of the entire outer gill, is used as a receptacle for 
the eggs, a sharp demarcation existing between the marsupial region and the 
anterior respiratory part. The former is highly modified in structure and may 
be readily recognized, even when empty. In Obliquaria reflexa a varying 
number of enlarged and distended water tubes contain the dense, glutinous 
cords in which the embryos are embedded. Finally, there are at least two 
genera, namely, Quadrula and Tritigonia, in which all four gills are utilized as 
brood chambers. This condition would seem to be the most primitive one, as 
the gills show but little structural modification and when empty are indistin- 
guishable from those of the male mussel. 
In all of the species of the genus Lampsilis which we have examined the 
posterior margin of the valves in the female is less pointed than in the male, and 
by reason of this difference in the shape of the shells the sexes may be readily 
distinguished externally. In most of the other genera that have come under 
our observation, however, the shells of the two sexes are identical, and an inter- 
nal examination is necessary in order to determine the sex of an individual. 
The entire embryonic development takes place within the marsupium, and 
at the end of this period the glochidium, which has become liberated from the 
egg membrane within the gill, is fully formed. In those species which are 
known as “summer breeders,” the glochidia are discharged at once, while in 
many others, the so-called “winter breeders,’’ they are carried in the gills for 
several months before being set free in the water. 
The glochidia, as has long been known, do not complete their development 
unless they become attached to fishes, and the metamorphosis occurs while 
they are living parasitically in a cyst formed by the epidermis of the host. 
After discharge from the marsupium, the glochidia under natural conditions 
lie quiescent on the bottom, their valves gaping widely apart, and unless chance 
brings them in contact with a fish they will die. Once attached to the gills or 
external parts of a fish, however, the rest of the development, if conditions are 
favorable, is assured. 
This unique peculiarity in the life history of fresh-water mussels has been 
taken advantage of in artificial propagation, as it has been found that a single 
fish, even though small, may successfully carry to the close of metamorphosis 
several hundred glochidia. One fish may, therefore, be made to do the work 
