54 
THE VOYAGE OF H.M.S. CHALLENGER. 
(see PI. Ill, fig. 1). The vertical position and the shape of the Ascidiozooids are like 
those in a typical Botrylloides. 
The test is of a pale buff colour in most of its extent. It is slightly lighter and more 
transparent on the margins. It attains no great thickness in any part, but forms an even 
layer over both surfaces of the incrusted shell. In vertical sections the vessels and their 
terminal knobs appear to be more abundant in the lower part of the test and between 
the Ascidiozooids than on the upper surface. This is rather a puzzling circumstance, and 
is apparently incapable of explanation according to my view^ that these vessels and 
terminal knobs have a respiratory function. Possibly their curious arrangement in this 
species is due to the fact that there is comparatively little test on the . upper surface of 
the colony, as the Ascidiozooids are numerous and closely placed, while on the surface at 
the posterior ends of the Ascidiozooids there is a thickish layer of test in which lie most of 
the vessels and their terminal knobs, and a number of buds and young Ascidiozooids, 
The terminal knobs are in some places, especially at the sides of the Ascidiozooids, 
exceedingly plentiful on the ends of the vessels, from which they bud off in all directions 
(see PI. III. fig. 2). In some cases the terminal twigs are like small bunches of grapes. 
The cells in the homogeneous test matrix are neither of large size nor very abundant. 
The turning in of the test to line the branchial siphon is very clearly seen both in a 
surface view and in vertical sections. Figure 5 on Plate III, shows the appearance 
presented in a surface view with Swift’s 1-inch objective of a specimen stained in picro- 
carmine, from the outside. The dark band (sph) is the sphincter muscle, and the test is 
seen to form fifteen or sixteen lobes, where it turns in to line the aperture. In vertical 
sections these lobes can be traced down to the anterior edge of the tentacles (PI. III. 
fig. 4). 
The mantle is very like that of Botrylloides purpureum. The branchial sac is large, 
and is elongated at right angles to the surface of the colony. Some of the transverse 
vessels are wider than others, but there is no regularity in their arrangement. A few 
muscle fibres are generally to be seen in the transverse vessels. The internal longitudinal 
bars are strong (PI. III. fig. 3, i.l.), and are slightly thickened at their points of union 
with the transverse vessels, but there are no papillae. In a fully developed sac there 
are at least twelve rows of stigmata, and there are twelve to fourteen stigmata in the 
largest rows. 
The endostyle is large and conspicuous (PI. III. fig. 4, en.). The dorsal lamina is 
very distinctly cihated along its free margin (PL III. fig. 8, d.l.). The tentacles (PL III. 
fig.s. 4 and 7, tn.) are regular in their alternating sizes and arrangement. 
The dorsal tubercle is rather larger than usual (PL III. figs. 7, 8, d.t.). Its edges 
are formed of low columnar cells with cilia directed inwards towards the centre of the 
opening. In some .specimens, as shown in the figure (PL III. fig. 7), the neural canal 
’ On a New Organ of Respiration in the Tunicata, Proc. Lit. and Phil. Soc. Liverpool, vol. xxxix. p. 39, 1884. 
