No.1.] ZHE DEVELOPMENT OF BALANOGLOSSUS. a) 
almost everywhere the inner surface of the ectoderm. A cross 
section of the larva cuts the body cavity along its length, and 
shows its close connection with the ciliated band. Such a sec- 
tion is figured in Fig. 13, and incidentally shows the shape of 
the large cells of the circular band. It is by no means an easy 
matter to determine with accuracy the origin of an embryonic 
organ, and in the present case, owing to the comparatively 
small number of embryos at the critical period, I have had no 
little difficulty in getting positive evidence on the point. Yet 
I feel assured that in the present case no uncertain answer is 
given to the problem. Longitudinal sections, cutting earlier 
embryos from right to left, are the only ones of value. From 
such a series are taken the figures shown in Figs. 14 and 15. 
The first series (Fig. 14, a, 4, c, d, e, f) are from the left side, 
and the second (Fig. 15, a, 0, ¢, d, e, f, g, 2) is from the right 
side. The following table gives the number of the section from 
which each figure is taken : 
Fig. 14 (left). @. 13th Fig. 15 (right). a. 18th 
Gs 3h Ge 25 
GAR C. .20 
ad. 46 ad. 46 
e. 69 2. (60 
J 82 J 63 
& 76 
h. 77 
All the sections in which traces of this pair (third) of body- 
cavities was found, are drawn in the two series. Asa rule the 
small collections of cells found in the larva are present in but 
a single section. Occasionally, as in Fig. 14, c and d@ (44th and 
46th sections), and in Fig. 15, g and # (76th and 77th sections), 
the same mass is cut twice. As seen in the figure, these cells 
do not differ in any respect from the mesenchyme cells of the 
biastocoel, and when but a single cell is found in the suspected 
position, it cannot be affirmed that it is to become, later, part 
of the series ; but in most cases these cells have divided, so 
that an accumulation of from two, three, four or five nuclei are 
present, and there can then be no question of the interpreta- 
tion. 
