410 Miscellaneous. 
copiously ramified, and bearing branches on the right and the left 
of the trunk, although less frequently on the latter side than in the case 
of those described by Weismann. The ccenosare not only at the 
base of the colony, but also close to the summit, is formed, or rather 
issues, from two tubes described by Weismann, the outer one being 
smooth and straight, while the inner one has undulating lines. The 
zooids are club-shaped, with about twelve scattered tentacula, of which 
some were rather longer than the rest. 
The medusoid buds proceed from a distinct pedicle, attached to 
the stalk of the supporting hydrosoma. From their structure, as 
Weismann observes, it is obvious that they become liberated and 
able to float away. I kept my colonies for two or three days, living in 
vessels in which the sea-water was constantly changed and kept in 
motion, and I was thus able to obtain a certain number of medusa- 
like zooids, which swam freely about like the medusve of a Bougain- 
villia or a Podocoryne, and on comparing them with the medusoid 
buds of Dendroclava it was evident that they were the ultimate 
evolutionary forms of the latter. 
The detached meduse of Dendroclava Dohrnii are somewhat 
larger than those of Podovoryne carnea. They have a well-developed 
umbrella with prolongations: the manubrium, or pedicle, is on the 
other hand small, not reaching farther than the middle of the 
umbrella, and is somewhat bottle-shaped with small ramifications at 
its mouth. About halfway down appear four yellow-green radiated 
spots, “ the gonophores,” which have a smooth surface. There are 
four radiated canals, which are flattened asin thefamily Tiaride. At 
the base of these there is a delicate longitudinal fibrillation. Of 
the eight marginal tentacles four are radial in continuation with the 
radial canals and fourareintermediate, the base of all being excessively 
dense, while the lower side, as well as the distal extremity, bears 
an orange-coloured spot (ocello). 
Weismann considers that this medusa should be placed in the 
family Tiaride (Heckel), and in the subfamily of the Pandeida, 
while it may possibly be referred to the genus Pandca (Lesson) or 
to Conis (Brandt). A close examination of the adult zooids shows 
that they have the characters of the Tiaride * and of the subfamily 
Pandaide, but cannot be referred to the genus Conis, as they have 
no double crown of tentacles bearing ocelli on the shorter and 
upper of those bodies. This form may possibly not admit of being 
included in the genus Pandewa; at any rate | am unable to detect 
those urticating threads (Nesselstreifen) of the umbrella which 
distinguish the latter genus. 
The polypoid form has been referred by Weismann to the family 
Clavide. If, however, we follow Allman f the fact of its having 
gonophore medusoids would lead us rather to place it in the family 
* KE. Heckel, ‘Das System der Medusen.’ Jena, 1879, 1 Theil, 1 
Halft, p. 40. 
+ A. F. Allman, ‘A Monograph of Gymnoblastic, or Tubularian 
Hydroids,’ 1872, London. 
