. Coelenterate Nature of the Sponges. 91 



fully developed, have, like the Choanoflagellata, a peculiarly 

 differentiated frill (collar) , which surrounds the basal part of 

 the flagellum like a funnel, and also pulsating vacuoles in the 

 interior. According to Kent the ciliated cells of the free- 

 swimming larvee also show the same characters ; but this 

 unfortunately has been seen by no one but himself. 



Schulze, indeed, admits that it would seem hardly natural 

 to suppose that so peculiar a structure as the collar of the 

 flagellate cells had originated spontaneously twice in different 

 groups of animals ; but that it is to be regarded as inherited by 

 the more complex, and therefore probably more recently diffe- 

 rentiated form, from the simpler and therefore probably older 

 form ; but upon the whole he finds that^ even if we leave out of 

 consideration a series of differences, which certainly exist be- 

 tween the Choanoflagellata and flagellate cells of Sponges, it 

 would not be a justifiableconclusion to deduce from a similarity, 

 however close, of unicellular Protozoa with certain cells of the 

 trilamellate Sponges, that the latter pertained to the former. 

 Moreover, in reality, in the blastula of the Sponges flagellate 

 cells of this kind, furnished with collars, are always deficient, 

 although their presence there might justly be expected, if 

 they originated from the Choanoflagellata. After taking the 

 trouble to examine how Savile Kent could have come to the 

 erroneous assumption of the presence of the collar in the 

 ciliated cells of the sponge-larvEe, Schulze arrives at the con- 

 clusion that the Sponges are true Metazoa, for they have 

 sexual reproduction, and in their larvae two different cell- 

 layers, an outer and an inner one, may be clearly distinguished. 



Schulze then discusses the hypothesis of the Coelenterate 

 nature of the Sponges. To the radiate structure occurring 

 occasionally in larvse and also in adults he ascribes no great 

 importance ; the Ascones never formed radial diverticula of 

 their central cavity, and if these were produced in Sycones as 

 sacciform distensions of the body-w^all, it must be borne in mind 

 that the Sycones before they formed the radial tubes possessed 

 ontogenetically the pure type of the Ascones, so that the latter 

 must consequently be regarded as ancestors of the former. 

 Hence it seems very probable that the most ancient sponge- 

 forms possessed an OlynthusAikQ form without radial diver- 

 ticula of the central cavity j and the developmental history of 

 the Sponges, so far as we are yet acquainted with it, presents 

 no sufficient ground to justify the assumption put forward by 

 me of common ancestors of the Sponges and Cnidaria, with 

 radially arranged mesenterial pouches, tentacles with urti- 

 cating capsules and indifferent aquiferous pores. It may be 

 true that the difference between the free-swimming larvce of 



