TWO SPECIES OF HYDRACTINIA 
479 
arranged in several rows and cover nearly the entire surface of the 
hypostome. Now, in some other places and colonies spiral zooids , 
of the latter form are much more numerous, and the tentacles 
are frequently longer. Again,, there are sometimes found near 
these spiral zooids true gasterozooids which are much more slender 
than their fellows and with shorter tentacles, forming in fact a 
transition to the true spiral zooids. The latter have the gastric 
cavity well developed, but in no case have I been able to find the 
mouth, although the above observations would not justif}^ one 
in entirely den3dng its presence, the more so as there are all degrees 
of transitional forms. 
The blastostyles are found in large numbers on many of m}^ 
specimens, fig. 9, which were all collected in the first part of the 
month of April, both on the general surface of the chitinous shell 
and on the spines. They appear, however, to be absent from the 
apical part of the spines. They are considerably smaller than the 
other polyps (figs. 6, 9), fusiform in shape, with a small but variable 
number of short tentacles without any definite arrangement, and 
with an inconspicuous hypostome destitute of a mouth. They are, 
as already mentioned, generally unisexual, but occasionally herma- 
phrodite ones occur. For instance, among 51 blastostyles taken 
at random from three colonies, viz., 13 from colony a (all?), 20 
from colony b (mostly 9) and 18 from colony c (all c?), there 
was only one hermaphrodite gonophore (from colony h). The 
gonophores appear to arise from a somewhat narrowly defined 
zone about midway along the length of the blastostyle (fig. 9). 
The female gonophores are spherical, the male ellipsoidal, but 
the elongation in the latter is so slight that the difference is notice- 
able only in sections passing through the long axis of the gono- 
phore. When there are many gonophores on a blastostyle they 
are arranged roughly in a circle, although some of them may be 
superposed. In the following will be given an account of the devel- 
opment of the gonophores as made out from a complete series 
of stages. My results differ from those of previous observers 
working on Hydractinia echinata. 
