258 
THE VOYAGE OF H.M.S. CHALLENGEE. 
branchial sac, and a well-developed atrial siphon is present ; and Tetradidemnum, in 
which there are four rows of stigmata, and the atrial aperture is provided with a languet. 
This classification appears, however, to refer only to the species which Della Valle was 
investigating, and which seem from his descriptions and figures to be all referable to the 
old genus Didemnum. Consequently his Trididemnum and Tetradidemnum may be 
regarded as subdivisions of Giard’s Didemnum. 
Von Drasche,/ in his recent scheme of the classification of the Synascidise, does not 
•recognise Eucoelium. His family Didemnidse contains two genera only, which are 
named Didemnum, Giard, and Leptoclinum, Milne-Edwards, and he gives as a new 
distinguishing feature that the former possesses four rows of stigmata while the latter has 
only three rows.^ In his large work on the Synascidise of the Gulf of Eovigno (1883), 
von Drasche divides the family into Didemnum and Leptoclinum, and then subdivides the 
latter genus into Leptoclinum and Didemnoides; he rejects the genus Euccelium on the 
ground that the characters ascribed to it by Savigny and Giard are not sufiicient to 
distinguish it from Didemnum and Leptoclinum. 
If the various classifications referred to above are combined as far as possible, 
they will form the following scheme, which contains all the subdivisions of the family 
which have been proposed : — 
Family. 
Didemnid^, 
Genus. 
Didemnum. 
■ Eucoelium. 
Leptoclinum. 
Subgenus. 
Trididemnum.. 
Tetradidemnum. 
Leptoclinum. 
Didemnoides. 
It is clear, however, that Della Valle’s Tetradidemnum, with its four rows of 
stigmata and its atrial languet, is closely aUied to von Drasche’s Didemnoides, and cannot 
be retained as a subdivision of Didemnum if that genus is used in von Drasche’s sense. 
If the number of rows of stigmata is to be regarded as the most important distin- 
guishing feature in the family, then three sections must be recognised, viz., (l) with 
three rows of stigmata, Trididemnum (or Didemnum in the limited sense) ; (2) with 
four rows of stigmata, Tetradidemnum, Leptoclinum, and Didemnoides; (3) with six 
rows of stigmata, Eucoelium (see Savigny’s figures). Leptoclinum and Didemnoides may 
then be separated by the thickness of the colony, leaving Tetradidemnum (in regard 
to which we have not yet sufficient information) with three possibilities before it, viz., 
(l) it may possibly form thin in crusting colonies, and in that case it should be included 
in Leptoclinum, (2) it may form thick masses and would then be identical with 
’ Zool. Anzeiger, 1882, p. 695. 
2 I am inclined to think that he has accidentally misplaced the figures and means the reverse, viz., Didemnum, 
3, and Leptoclinum, 4. See also, Die Synascidien der Bucht von Rovigno, p. 9. 
