156 GEOL. AND NAT. HIST. SURVEY OF MINNESOTA. 
shell walls themselves, and the floor is formed by a membrane sup- 
ported on, and investing in part, the strong muscles which connect 
the abdomen with the upper anterior part of the dorsal plite. Thus 
a space is left between the pericardium and the intestine which is oc- 
cupied by a special blood sinus leading toward the posterior and 
lower part of theabdomen. The posterior wall of the space is formed 
by a chitinous partition which bounds the brood space, or its homo- 
logue, and is connected by chitinous processes (stutzbalken) with the 
outer skeleton. The anterior, on the other hand, is only bounded by 
the supporting ligaments of the abdomen above described and mem- 
branous partitions. As usually described, the heart lies suspended in 
the cavity thus defined by slender muscular threads, more or less like 
those of the heart of Corethra larve and the like; and such seems to be 
the ease at first, but a more careful study shows that this is far from 
correct. On the contrary, the chief supports of the heart are mem- 
branes which, seen in cross-section with the attached grains or blood 
globules, assume the appearance of exceedingly slender structureless 
threads. The action of re-agents indicates that these supposed 
threads are not muscles, but composed of connective tissue; while by 
changing the focus the sharpness of the line is frequently not altered, 
but its relative position is changed,—a simple test which often serves to 
dispel an illusion of this sort. That there are some threads of the 
character above mentioned is not to be doubted, as in connection 
with the valves of the heart; but the proper support of the heart is 
found in the membranes which invest it in part, and are reflected 
upon the walls of the shell and, anteriorly, of the intestine. It is not 
yet possible to fully describe the insertion of these tissues, as there is 
so large a number, especially about the anterior opening, where they 
lie in all directions and at all angles, and are so transparent that only 
their vertical sections appear as dark lines. Thus the same membrane 
appears and disappears, only to re-appear in a different position, 
where it might be readily taken for a distinct membrane. In general, 
however, I hope to make no serious error in the following summary. 
Before going into detail, however, it will be necessary to consider the 
intimate structure of the heart, as well as its general shape and posi- 
tion. (See Plate LL.) 
The general shape is that of an irregular oval with the greatest 
convexity posterior (Daphnia, etc.), or it may be strongly bifid and 
thus somewhat Y-shaped (Hurycereus, etce.). It is held in position in 
the pericardial cavity by the membranes above alluded to, to which 
it is attached at definite points, the principal of which are two slight 
enlargements on the lower posterior portion, which are in part op- 
posed to each other and also to a superior posterior point of insertion. 
