488 
SEITARO GOTO 
belongs to the British Museum, and was found, without any label 
or indication of its locality, among the late Dr. Bowerbank's 
collections." Carter ('78) however gives the locality as Poly- 
nesia with a query, and Steinmann^® gives it as 'Svahrscheinlich 
die Philippinen." The figures both of Carter and Steinmann 
clearly show its close resemblance in the general form of the chiti- 
nous skeleton to Hydractinia sodalis, and it is not going too far 
to assume that the internal structure must also be closely similar, 
especially as Carter's description of the skeleton of Hydractinia 
echinata^® shows essentially the same arrangement of parts, as in 
Hydractinia sodalis. Here it may be mentioned that Carter's 
description of '73 is much more accurate than that of '77, as may 
be seen by comparing the two descriptions with the results ob- 
tained by Colcutt ('98) through the use of modern methods. The 
later paper of Carter ('77) appears to me to have been influenced 
too much by a desire to bring out the supposed affinity with cer- 
tain fossil forms (other than Hydractinia), which must in my 
opinion be said to be at most only remote. Renewed examination 
of the various fossil forms mentioned by Carter and Steinmann, 
especially if microscopical sections are prepared, will probably 
throw light on this question. In Carter's specimen of H. arbores- 
cens the hydroid skeleton did not entirely cover the shell which 
was that of Fusus sulcatus,^^ but this is clearly a point of sec- 
ondary importance. Again in Hydractinia levispina^^ is clearly 
seen, as pointed out by Steinmann, the tendency of the hydroid 
skeleton to grow beyond the mouth of the gastropod shell to form 
a '^shell" of its own, and the presence of the same tendency in Hy- 
dractinia echinata can be inferred from Carter's description.^^ If 
now we compare the chitinous skeleton of Hydractinia with those 
of Solanderia, Clathrozoon and Hydrodendrium, they are all found 
to be on essentially the same plan derived from the well known fun- 
damental structure of the hydroid colony. Are we to infer from this 
that they are all closely related phyletically? There can be no 
28Loc. cii.: 78, p. 299. " gteijunann : '78, p. 109. Carter: 73, p. 2. 
" I.e., according to Steinmann. Carter gives it as Phos senticosusor Fusus 
eulcatus. 
"Carter: 73, pi. 1, fig. 1. ^'Loc. cit.: 78, p. 109. 
