176 NOTES AND MEMORANDA. 



name in the larva. From the ectoderm of the larva is formed the 

 external layer of the skin, which in Spongtlla consists of two distinct 

 layers, the epidermis and cutis. The larval entoderm forms the thin 

 single-layered lining of all internal cavities or canals (the body-cavity 

 excepted), as well as a covering to the mesodermal septfe, trabeculje, 

 &c. The mesoderm of Spongilla may be regarded as a simj^le form of 

 connective tissue in which the cell element prevails, and the structure- 

 less gelatinous matrix is very slightly developed. A syncytium, in 

 Haeckel's sense, does not exist in the Spongilla. Fusion of Spongillce 

 of dififerent forms and sizes never gives rise to the formation of the 

 so called pseudo-oral orifices, pseudo-enteric cavities, communicating 

 canals, and other cavities coated with ectoderm. 



We can distinguish in sponges two different modes of development. 

 One group of sponges show in their developmental history a well- 

 pronounced blastula stage, i. e. a hollow single-layered sac with a largo 

 segmentation cavity in the interior. Some of those sponges, as 

 Halisarca lob^daris, Dujardinti, Ascetta primordialis, A. claihrus, have 

 an archiblastula stage ; in others of the same group (Sycandra raphanus, 

 compressa, the calcareous sponges of Barrois), the modified amphihlas- 

 tula form obtained. In this form of the generation cycle the two 

 primitive embryonic membranes originate by the cells of the posterior 

 half of the sac undergoing differentiation into the primitive entoderm, 

 whilst the cells of the other half of the blastula give rise to the ecto- 

 derm of the larva. The formation of the gastric cavity in this case 

 depends, in all probability (as F. E. Schultze and Barrois have 

 already noticed), upon the invagination of the posteriorly-situated 

 entoderm, in the interior of the segmentation cavity of the embryo. 

 But whether the aperture of invagination of this provisional archi- 

 gastrula passes directly into the actual oral orifice of the sponge, 

 remains to be proved. The sponges of the second group begin their 

 ontogeny with the morula stage (all siliceous sponges — Spongilla, 

 Esperia, Beniera, Amorphian, Desmocidon, Isodictia, Raspailia ; also the 

 calcareous sjionges of Haeckel). The formation of all three embryonal 

 membranes depends in this case on the delamination process. The 

 stomach cavity of the larva is formed by the separation and dissolution 

 of some cells of the interior of the entoderm mass. In place of the 

 gastrula, the plano-gastrula makes its appearance. The generation 

 cycle with the blastula stage is to be regarded as the most simple. To 

 this corresponds also the much simpler organization of the sponges 

 of this group : Halisarca, for example, has no skeleton, is everywhere 

 covered with the ectoderm of the larva, &c. The sponges which in their 

 developmental history pass through the morula stage, are also more 

 complicated in morphological and histological respects. The place 

 of the sponges as a particular class of the Coelenterata is entirely 

 natural, on the ground of all the facts hitherto known of comparative 

 anatomy and embryology.* 



Mr. F. M. Balfour refers shortly to the above in the article next 

 mentioned, and considers that M. Ganin's account of the development 

 of Spongilla is not reconcilable with that of Sycandra, as described 

 * ' Zoologischcr Anzcigcr,' vol. i. (1878) p. 195. 



