308 
THE VOYAGE OF H.M.S. CHALLENGEE. 
test, an upper and a lower, between which the bodies of the Ascidiozooids are suspended, 
and they emphasize this peculiarity as one distinguishing these colonies from those of 
other Compound Ascidians ; but in all the specimens of Diplosomidse which I have 
examined, the relation between the test and the Ascidiozooids of the colony is essentially 
the same as that found in all other Ascidiee Compositse. The only difference is that 
the test is usually extremely soft and gelatinous, and that it is considerably broken 
up by the presence of numerous large spaces and canals which are prolongations of the 
common cloacal cavities, and serve to place the atrial apertures of the Ascidiozooids in 
communication with the exterior. This is of course a condition which is found to a 
greater or less degree in most Compound Ascidian colonies, and although the result of 
its coDsiderable development in the Diplosomidse is doubtless, as described by former 
authors, to reduce very greatly the amount of test in the centre of the colony, and leave 
only the surface layers as continuous and conspicuous expansions, still it is no novel 
arrangement, but merely a slight modification, and the test must be regarded as a true 
investing mass surrounding the Ascidiozooids on all sides, just as it does in the other 
Ascidise Compositse. 
The test is usually exceedingly transparent, but in some species it contains conspicuous 
pigment cells. Calcareous spicules are rarely present, and are confined to the superficial 
layer of the test. Giard was of opinion that they were never present, and made the 
absence of spicules the character by means of which he distinguished the Diplosomidse 
from the Didemnidse.^ As a general rule, it is a good distinctive feature, but it will not 
hold good universally, as von Drasche has described two members of the Didemnidas, 
Didemnum inarmatum and Didemnum tortuosum, in which the test has no calcareous 
spicules, and I have added a third species, Didemnum inerme, which is in the same 
condition; while, on the other hand, von Drasche found one of the Diplosomidse, Diplo- 
soma pseudoleptoclinum, in which calcareous spicules were present in the upper layer of 
the test, and in a small clump on each side of the thorax of the Ascidiozooids, and 
in one of the species to be described below, Diplosomoides molle, calcareous spicules 
are also present. Consequently, this character breaks down as an infallible diagnostic. 
The general shape and structure of the Ascidiozooid is very much like that which 
prevails in the Didemnidse. The thorax is usually longer than the abdomen, and the 
atrial aperture is very inconspicuous. The branchial sac is of large size, and the stigmata 
are well developed. 
The stomach is ellipsoidal, and has no folds in its wall. Its long axis generally runs 
dorso-ventrally, and the intestine forms rather a wide loop. 
The testis is usually in the form of two large spermatic vesicles, opening into a vas 
deferens which is not spirally coiled. This is an important point of distinction between 
' Rtclierches sur les Synascidies, &c., p. 608. Jourdain, writing quite recently (Oompifes rendits, June 15, 1885), 
lia-s retunied to this incorrect view tliat the absence of spicules is the chief distinguishing feature of the Diplosomidas. 
