376 Relationship of the Sponges to the Choanoflagellata. 



Sponges quite independently of that of the Choanoflagellata 

 seems to me to be a priori by no means excluded. 



The circumstance that in quite distinct groups of Protozoa 

 we meet with delicate membranous elevations of the plasma- 

 body, which, although not the same as, are yet similar to, 

 the collar, may perhaps indicate that the faculty of forming 

 such processes may be inherent in protoplasm in general, and 

 therefore that such processes might be produced indepen- 

 dently of each other even in different divisions of animals and 

 at different times. I have myself observed * in Placopus 

 ruber, a freshwater Rhizopod, pseudopodium-like processes, 

 which originate on the free upper surface of the animal, and 

 by the fusion of their contiguous lateral mai*gins may unite 

 to form delicate membranous funnels. The so-called undu- 

 lating membranes of many ciliated Infusoria also resemble 

 the collar in many respects ; but one does not on this account 

 assume a close relationship between those Infusoria and the 

 Choanoflagellata. 



In estimating the relationship of the Sponges to the Cnidaria 

 the consideration of the larvas will also be of great conse- 

 quence, and certainly not less important than the comparison 

 of the fully-developed animals, which has hitherto been prin- 

 cipally employed. Notwithstanding the small extent of our 

 knowledge of the two kinds of larva? and the mode of their 

 metamorphosis, we can even now assert that the difference 

 between the free-swimming ciliated larvge of the Sponges on 

 the one hand, and of the Cnidaria on the other, is on the 

 whole not more considerable than between the different 

 sponge-larvee themselves. No one can say, with regard to 

 any ciliated larva met with accidentally in sea-water and not 

 already known to him, whether it is a Sponge- or a Cnidarian 

 larva. It is only after metamorphosis that those primary 

 differences of organization by which we can easily and sharply 

 separate the two groups from each other make their appear- 

 ance. 



Thus, in my judgment, we are justified in the belief that 

 the divergence of the two lines did not commence before that 

 phylogenetic developmental stage which corresponds to the 

 ciliated larva ready for metamorphosis. But what degree of 

 organization was attained before the separation actually took 

 place it will be more difficult to decide. 



I can find in the developmental history no satisfactory 

 ground for Marshall's above-mentioned hypothesis, that the 

 common ancestors of the Sponges and Cnidaria possessed 



* Arch, fur mikr. Anat. Bd. xi. p. i<48. 



