THE POSITION OF SPOXGES, ETC. 437 



port Marshall's views, and thinks it probable that the oldest 

 sponges had no radial diverticula of their central cavity 

 but a simple sac-like form, like an Olynthus among the 

 Ascons (1885, p. 191).^ 



I have left to the last the peculiar view expressed by 

 von Lendenfeld, since it cannot be assigned a place in the 

 classification of the sponge theories which I have adopted. 

 In his Monog7'aph of the Horny Spono^es (1889) he com- 

 pares sponges as " Mesodermalia " with other coelenterates 

 or " Epithelaria". Facing p. 885 he gives a figure of a 

 tree, representing the pedigree of sponges and other animals. 

 From a massive stem labelled " Protista" comes off a thick 

 branch on the right, labelled at its origin " Animalia," and 

 a little higher up, but before it has divided, " Coelentera". 

 Just above where this latter name is affixed come off two 

 branches, Epithelaria and Mesodermalia. So far every- 

 thing seems quite clear, especially when we read on p. 887 

 that the sponges form "a separate phylum in the grade 

 Coelentera of the Metazoa". But on p. 886 we read that : 

 " The sponges have probably been developed independently 

 of other Metazoa from Protozoan ancestors, the nearest 

 existing relations to which are the Choanoflagellata," and 

 that " the mode of the phylogenetic development of sponges 

 was similar to that of other Metazoa ". In spite of this view, 

 however, the author "does not see that there is any justifica- 

 tion for the establishment of a special sub-kingdom for 

 the sponges. They are evidently Metazoa and no doubt 

 Coelentera in the sense given above, namely, the grade 

 Coelentera as opposed to the grade Coelomata " (p. 888). 

 This method of reconciling statements apparently so con- 

 tradictory as that on the one hand sponges are Coelenterates, 

 and on the other hand separately descended from the 

 Choanofiagellata, implies a conception of the animal king- 

 dom essentially different from that of all the other authors 



1 Chun, who discusses the question of sponge affinities from the stand- 

 point of a specialist upon Coelenterata, may be referred to as one who, 

 taking a high rank as an authority upon a group of animals other than 

 sponges, supports the theory of their coelenterate nature. See Brown's 

 Klassen imd Ordunngen d. T., Bd. ii., Abth. 2 (Ccelenterata), pp. 86-96. 



