AXD PHYSIOLOGY OF THE TUNICi^TA. 333 



coils. The chief purpose of these radiating vessels is apparently to 

 prevent engorgement of the coil, and to aid the reflux of the stream 

 by conveying the blood in the most direct manner to and from 

 the centre of the coil. When the spiral arrangement is more im- 

 perfectly developed than usual, the radiating vessels are very irre- 

 gularly dispersed ; but even in such cases there can be little doubt 

 that their function is as above stated. The longitudinal bars have 

 the ribbon-like character of those in Stycla, and are principally 

 confined to the folds. 



Ascidia parallelogramma has also the secondary vessels spirally 

 arranged, as originally pointed out by Mr. Alder* ; and here the 

 coils, which form slight conical eminences, are disposed in regular 

 transverse series between the transverse channels. The coiled 

 vessels do not so frequently intercommunicate as they do in Mol- 

 gula ; consequently the stigmata are much longer, being less in- 

 terrupted. The radiating vessels are not numerous, and they 

 pass from the centre of the coil to the transverse channels. The 

 longitudinal bars are more rigid and cord-like than usual ; they 

 project considerably from the surface of the gill ; and the papillae 

 which are attached to them are not elevated, but inclining back- 

 wards are united throughout their length, and give support to 

 wide membranes that extend from the surface of the transverse 

 vessels. In this species the oral lamina is replaced by a series of 

 well-developed filamentous processes. 



The most perfect, however, of the spiral type of gill is found in 

 Molgula arenosa. In this interesting species the interior of the 

 branchial sac is furnished on each side with six or seven wide, 

 longitudinal, ribbon-like bands, which are attached by one edge 

 to the transverse vessels at the points where they cross them. 

 These bands, how like soever in general appearance, ai^e not to be 

 confounded with the branchial folds in Stycla and Molgula ; they 

 are the homologues of the longitudinal bars so constantly present, 

 and with the transverse vessels give to the surface a coarse reti- 

 culation, the square meshes of which are each occupied by a conical 

 eminence. On account of the prominence of the longitudinal 

 bauds these eminences, which look like so many miniature bee- 

 hives, seem to be arranged in six or seven longitudinal series. 

 Each cone is formed of a double spiral coil of secondary vessels 

 united at the apex ; the coils are perfect, and the stigmata, which 

 are coextensive with them, appear to be scarcely, if at all, inter- 



* " Obserrations on the British Tunicata," Ann. & Mag. Nat. Hist. S. 3. Vol. li . 

 p. 158. 



