256 THE STRUCTURE AND AFFINITIES OF HERDMANIA CLAVIFORMIS. 



Table I. 



COMPARISON OP HERDMANIA AND ITS ALLIES. 



From this table it is seen that Herdmania has rather more in common with the 

 Polyclinidse than with any other group, and I have little doubt that its closest of 

 kin is here in spite of its strong superficial resemblance to Clavelina. The similarity 

 of its intestinal tract, particularly in the length of the oesophagus and the dorsal 

 position of the ascending limb of the intestine, to that of some species of Distoma, 

 is an instructive example of parallel evolution, but cannot in itself be regarded as 

 evidence of close relationship. 



In attempting to determine the degree of consanguinity of Herdmania with 

 Clavelina we are confronted with the old question of the classificatory importance of 

 the presence or absence of a common investing testicular mass. The extension of our 

 knowledge is certainly diminishing this importance more and more. Witness, for 

 example, the unsatisfactory efforts to arrange in genera on this basis a considerable 

 series of species of the Clavelinidse, Stereoclavella Herdman ('91), Pycnoclavella 

 Garstang ('91), Synclavella Caullery (:00), and probably others. But the most 

 striking instance of the ready passage from the free-zooid state to the completely 

 invested state afforded by the "social ascidians" is probably that of Perophora 

 annectens Patter ('94). Neither is the opposite tendency wanting, namely, for the 

 zooids in genera and families of strictly compound ascidians to become freed from 

 the common investing test mass; for example, Distoma pulchra Bitter ('01). 



In Herdmania the freedom of the zooids is so complete and constant that it 

 may be regarded as a good generic character, but not more, I should say; so that it 

 cannot be appraised at a greater value than, for instance, the yolkless condition of 

 the ova, in estimating the true relationship of Herdmania and Clavelina. 



The most fundamental difference, according to our present knowledge, between 



