138 Affinity of Mollusca and MoUuscoida. By W. K. Brooks. 
ledge of adult structure, and its weakness is shown only when it is 
tested by embryology. The clearest and most forcible statement 
of this view is that given by Allman in his ' Fresh-water Polyzoa.' 
According to Allman, the Tunicata are intermediate between the 
Polyzoa below and the Lamellibranchs above. The branchial sac 
of a Tunicate represents the permanently retracted tentacular crown 
of a hippocrepian Polyzoon ; the tentacles form the horizontal bars 
of the sac, and uniting to each other at intervals enclose the 
branchial slits. Although Allman's figures are necessarily dia- 
grams, no organ is exaggerated or suppressed for the purpose of 
making the likeness more forcible ; they are very accurate and 
faithful representations of the animals, and show the closest simi- 
larity between these two forms ; the position, structure, and con- 
nections of almost ever}^ organ of the one being duplicated in the 
other. An almost equally perfect comparison may be made between 
a Tunicate and a Lamellibranch, but the recent great additions to 
our knowledge of the embryology of the Tunicate seem to show, 
with absolute conclusiveness, that we here have nothing but a very 
perfect and striking adult resemblance, reached in each of the groups 
in a different way, and therefore without homological signification. 
Whatever view of the Vertebrate affinity of the Tunicate we may 
incline to, we must recognize the fact that the branchial sac is 
morphologically part of the digestive tract, and in no sense what- 
ever a lophophore or a tentacular gill. Moreover, we should expect, 
according to all analogy, to find the affinity to other groups most 
clearly shown in the low or embryonic forms, but Appendicularia 
presents none of the peculiarities upon which the comparison is 
based. As Kay Lankester has lately referred to Allman's homology 
in a way which seems to imply that he still accepts it, I will repeat 
more briefly my reasons for rejecting it. These are, first, that the 
development of the Tunicate shows that the resemblance is not due 
to community of origin, but is reached in different ways; and 
secondly, that the adult Lamellibranchs are a specialization of the 
embryonic type, and therefore cannot lie in the direct line connect- 
ing the MoUuscoida with the Mollusca. Allman himself seems to 
have seen the force of the first objection, for in a much later paper 
(1869) he advances the view that the Polyzoa are connected, 
through Ehabdopleura, with the Lamellibranchs. His studies of 
this genus were made upon alcoholic specimens; and Sars, who 
enjoyed the superior advantages afforded by an abundance of living 
specimens, has shown that Allman was mistaken in regard to almost 
every one of the points upon which he attempted to establish the 
supposed relationship. 
These are only a few of the arrangements of the Mollusca which 
have been proposed, and the fact that, of the three selected, two are 
by Allman must not be regarded as the result of a wish to un- 
