310 MK. A. HAKCOCK ON THE ANATOMY 



other, and have, except where there is an abdomen developed, all 

 the viscera and the lacunary portion of the blood-system placed be- 

 tween them. On the other hand, the mantle and test in Ascidia and 

 Molgula are always free, except at the distal extremity of the respi- 

 ratory tubes, where they are united ; there is also an attachment 

 at the point where the vascular trunks enter the test. But in the 

 genus Stycla (Savigny's third tribe of Cynthia) the test is always 

 more or less firmly attached to the mantle throughout, though at 

 the respiratory orifices the adhesion is greatest. In all the 

 species, however, that have been examined, with the exception of 

 one (a small undescribed species) , these two envelopes may be sepa- 

 rated without much difficulty in specimens preserved in spirit. 

 In the exceptional case alluded to, the mantle is exceedingly deli- 

 cate ; and hence probably arises the difficulty of separating it 

 from the test. In this genus, as well as in Ascidia and Molgula, 

 blood-vessels pass from the body to the test. It is therefore 

 likely that vessels will be found ramifying in the outer tunic in 

 all the simple Ascidians. In Pelonaia the adhesion of the mantle 

 and test is not by any means so remarkable as was originally 

 supposed ; and, indeed, in this form they are as easily divided as 

 they usually are in Stycla. Also in Clavelina these two tunics 

 are slightly adherent throughout, while in Salpa they appear to 

 be as free as they are in Ascidia*. 



It should be mentioned that, in a living state, unless the 

 mantle be violently contracted, there is no actual vacant space, 

 or space filled with fluid, as has been asserted, between it and 

 the test ; even in those species which have these tunics compara- 

 tively free the two surfaces lie in close contact. When the 

 animal is dead, however, and preserved in spirit, the body 

 enclosed in the mantle does not by any means occupy the entire 

 space within the test, but lies somewhat shrivelled, and fre- 

 quentl}^ quite free, just as commonly happens with the animal of 

 the Lamellibranchs under similar circumstances. 



The chief function of the test, like that of the shell in the 

 higher moUusks, is no doubt to protect the comparatively soft 

 and delicate portions of the animal that lie within it. But it 

 will also act, by its resiliency, as a counterpoise to the muscular 

 contractility of the mantle, which lines it as it were. In those 

 species, such as Stycla tuherosa, in which the mantle and test are 

 adherent throughout, this action is readily understood ; it is not, 



* I have examined onlv one species, namely <S'. sfinos^a ; and the specimens 

 were preserved in alcohol. 



