92 
GEORGE LEFEVRE AND WINTERTON C. CURTIS 
of the egg, the conglutinates are very thin and as a rule are com- 
posed of but a single laj^er of eggs, while in others, as in the species 
of Lanipsilis, the plates are much thicker, especially above, and 
in horizontal section show several layers. In the genus Quadrula 
the thickness is intermediate. 
The amount of the matrix in the conglutinates also varies con- 
siderably in different species. In Unio complanatus and Obliqua- 
ria reflexa, the egg membranes, which seem to be of an adhesive 
nature, are closely pressed together in the mass on all sides, as 
shown in fig. 28, which is a detail drawn from one of the conglu- 
tinates of fig. 26. The glochidia are seen with the valves open 
but still contained within the membranes, which are closely adher- 
ing. In cases like this, it is difficult to determine whether there is 
any cement substances at all between the embryos, which may pos- 
sibly be held together solely by means of the adhesive surfaces of 
the membranes. In many other cases, however, the matrix is 
quite evident, and the embryos, which are not so closely appressed, 
are simply embedded more or less loosely in a glutinous binding 
substance. Such is the condition in Lampsilis and is illustrated in 
fig. 24, which is a portion of one of the conglutinates of fig. 27 
seen under higher magnification. 
In still other species, the eggs cannot be said to form congluti- 
nates at all, as they are merely suspended in a slimy mucus which 
never hardens sufficiently to enable the mass to maintain a defi- 
nite form when removed from the gill. This condition is most 
noticeable in Alasmidonta truncata. 
When the glochidia are fully formed, the cement, which through- 
out the embryonic development has held the embryos together, 
dissolves away, and the larvse are discharged at the time of spawn- 
ing in slimy strings of mucus which is secreted by the lining epithe- 
lium of the marsupium. 
HISTOLOGICAL STRUCTURE OF THE MARSUPIUM 
Figs. 30 and 31 are horizontal sections of water-tubes of the mar- 
supium of Quadrula ebena, and Lampsilis (Proptera) alatus respec- 
tively, and, as they are drawii under the same magnification, a 
comparison of the size of the tubes in the two cases may be readily 
