THE STRUCTURE AND AFFINITIES OF HERDMANIA CLAVIFORMIS. 241 



the stolon, into which it merges imperceptibly. The diameter at the base is about 

 2 millimetres. The younger zooids frequently appear to arise from the lower portion 

 of the body of the larger ones, rather than from the stolons (PI. XVIII, Fig. 1). 

 Since, however, the place of origin appears to be wholly indeterminate excepting in 

 that it is confined to the region a little below the heart and genitalia, any portion 

 that produces buds must be regarded as stolon even though superficially it appears 

 to be a part of the body of the zooid. Sometimes the individuals are quite straight 

 and erect, but more frequently they bend over at various angles and in various 

 directions, and the bodies are not infrequently considerably curved. 



Normally the animals are translucent; but this condition is frequently disguised 

 by foreign matter adhering to the surface. A slight greenish shade is usually recog- 

 nizable, and not infrequently a very decided dull, dirty green film covers the surface 

 of the test, particularly of the middle and posterior portions of the body. 



The organs of the body are disposed in three regions, namely, thorax, abdomen, 

 and postabdomen. These regions are not, however, obvious on surface inspection, 

 and it is only with careful study that the postabdomen, in the strict sense in which 

 the term is applied in the Polyclinidse, is recognized. In full-grown specimens that 

 are not contracted the abdomen is about five times the length of the thorax, and 

 the postabdomen is about equal to, or somewhat shorter than, the latter division, 

 though its proportional length is variable (Fig. 2). 



3. The Test. — This rarely if ever reaches the thickness of 1 millimetre. It is 

 rather soft and pliable. Its surface is smooth excepting where it has become eroded 

 through age and the action of external agents, as diatoms, etc. It is quite clear and 

 transparent normally, that is, when not covered by foreign matter or eroded. 



The cells are generally few; in fact I know of no other ascidian in which they are 

 so scattered as they are here. The cells are more than usually uniform in size and 

 shape. There are no vessels in the test, and the matrix is very uniform in structure. 



4. Musculature of the Mantle. — This is moderately well developed, especially as 

 regards the longitudinal fibres. These extend the entire length of the animal. In the 

 thorax they have a tendency to be arranged in bands; but in other portions of the 

 body they constitute a uniform layer. Circular fibres are confined to the thoracic 

 region and are most numerous at the anterior end and in the siphons. They are dis- 

 posed in bundles of about six, eight or ten fibres each. These are numerous, and 

 anastomose with one another at short but irregular intervals. 



5. Branchial Apparatus. — The general form of the siphons as they appear in a 

 specimen dissected from the test is shown in Figures 2 and 3 (PI. XVIII) . They are not 

 prominent before removal from the test, but afterwards they are decidedly so. The 



