262 SCHULTZE, ON A LIVE -BOX. 
fig. 2 a longitudinal section. It is particularly advantageous 
to bevel off, beneath, that end of the cell at which the head of 
the animal is placed, whose movements are thus effectually 
prevented. The caudal extremity, in order to ensure its 
lying flat in the shallow part of the excavation, must be 
slightly twisted at the base; a proceeding, however, which 
ELiGe as 
eae Fi Gace ae 
does not appear much to incommode the animal. It will be 
as well to have several cells of the kind of different forms in 
readiness. At any rate, the deeper part of the hollow should 
be much shorter when it is intended for the Frog-tadpole 
than in the case of the Newt’s, for which latter again the 
part for the reception of the tail should be somewhat 
shallower, or entirely omitted; and, on this account, the 
hinder wall of the excavation should be made very sloping or 
somewhat rounded, so that the flat tail may rest directly on 
the even surface of the slide. In the case of young fishes the 
hollow should be still narrower and shallower, and in this 
case it is advisable to bevel off the whole of the hinder 
border. 
As the grinding out of a hollow of this kind in a thick 
piece of glass would be attended with difficulty, and conse- 
quently with considerable expense, I have always constructed 
the apparatus of three flat pieces of glass. The lower of these, 
as seen in fig. 2, is nothing more than a common slide, and 
upon this, as a basis, the two others, in which the requisite 
incisions have been made, are affixed by means of Canada 
balsam [or marine glue]. The dotted transverse line in fig. 1 
shows the point of junction of these two pieces. When used 
the hollow is filled with water, into which the animal is 
introduced, with its head beneath the anterior border, and 
the tail in the shallow depression at the other end, or on the 
surface of the glass, as the case may be, the whole being 
covered with thin glass in the usual way. 
