150 
THE YOYAGE OF H.M.S. CHALLENGER. 
reproductive organs and the heart (see fig. 3 on p. 16), The length of the post-abdomen 
is very variable (compare PL XX YTTT. fig. 2 with PI, XXX fig, 13), and as pieces may 
become detached from its end in the form of buds, it is obvious that the size may vary 
not only in different Ascidiozooids of the same colony, but even in the same Ascidiozooid 
at difierent times. The post-abdomen is composed of an outer layer of ectoderm covering 
a narrow sac-like prolongation of the connective tissue and muscle bands of the mantle. 
This sac contains a mass of mesoderm cells, and is divided into two parts by a double 
median partition or septum (PI. XXIX. fig. 11), which, according to Kowalevsky,^ is a 
prolongation of the posterior end of the branchial sac. 
The test varies greatly in its texture. It may be soft and gelatinous as in Polyclinum 
inolle, firm and cartilaginous as in Atopogaster gigantea, or tough and leathery as in 
Amaroucium albidum. In some cases, when it would otherwise be soft, its appearance 
and character are totally changed by the numerous sand-grains and other foreign bodies 
attached to its surface and even imbedded in its interior. A number of the new forms 
discovered during the Challenger Expedition have the test in this curious condition (see 
PI. XXXI. fig. 9), They form the genus Psammaplidium. 
The mantle is well developed, and the musculature is usually of moderate strength. 
There is great variability in the arrangement of the muscle bands. In some cases they 
are all longitudinal in direction, while in others the chief bands run circularly around the 
thorax. In the post-abdomen they are always longitudinal. The branchial aperture is 
usually very distinctly six-1 obed (PI. X XX I. fig. 3), rarely eight-lobed {Morchellioides 
affinis, PI. XXIV. fig. 19) or nearly circular {Aplidium fuscum, PL XXVIII. fig. 10). 
The atrial aperture may be six-lobed, but is more usually circular, or provided with a 
single large lobe placed on its anterior edge and known as the atrial languet (PI. XXIX. 
fig. 10, at.l.). This languet may be placed considerably in front of the atrial aperture 
(as in von Drasche’s Polyclinoides diaphanuin, see also PL XXXI. fig. 3), and in that 
case the margin of the aperture is circular and unlobed. There is seldom any well- 
marked atrial siphon. 
The division of the Polyclinidse into genera is an exceedingly difficult matter. 
Previous to the publication of von Drasche’s short note on Polyclinoides diaphanum, in 
1883, the family seemed to fall naturally into two groups which have been recognised 
by most authors as the genera Aplidium and Polyclinum, but the subgenus Polyclinoides, 
proposed by von Drasche, unites the characters of Aplidium and Polyclinum in a way 
which renders it practically impossible to say to which group it really belongs. Von 
Drasche assigns it to Aplidium, but he might, I think, with equal propriety have placed 
it under the genus Polyclinum. Some of the new species described in the following pages 
also exhibit the same intermediate characters between the two groups, and render their 
recognition in the wider sense in which they are employed by Giard and von Drasche no 
^ Ueber die Knospung der Ascidien, Archiv. f. mikrosk. Anat., Bd. x. p. 441, 1874. 
