ONTOGENY AND PHYLOGENY. 
Spong. 9 
of stones, feeding on the various kinds of organisms which affect these 
surfaces ; the flattened, cap-like form of the gastrula at one stage sug- 
gests so much (Trichoplax is mentioned in connection with this). It is 
to be borne in mind that flagella persist throughout life in connection 
both with ectoderm and endoderm in Oscarella , and that attachment of 
the gastrula only occurs at single points of the margin, so that the 
flagella of the ectoderm will bring nutrient particles to the Sponge, 
and these will then be carried into the gastral cavity through the split-like 
interspace between tho basal layer and the edgo of the mouth. The 
pores (secondary polystomy) may have been produced by diffusion 
streams at the points where the ectoderm and endoderm were in con- 
tact ; thus, at first respiratory, they would, from their advantageous 
position, come to function as mouth openings, and the original mouth 
would close up ; the formation of the oscule (a character sui generis ) 
then becomes a difficulty. 
The sponges are genuine Metazoa , and the two primary layers are 
truly homologous with those of higher types ; but they are not Ccelenterd , 
there is a diametrical opposition between the two groups, as ontogeny 
teaches, for the gastrula in one attaches itself by the oral pole, and in 
the other by the aboral. The animal part of the Sponge undergoes but 
slight differentiation as compared with that of the Ccelentera : there 
are no grasping organs (tentacles), no true muscle fibres (the contractile 
fusiform cells of the Sponges cannot be so regarded, since the contrac- 
tile substance is not yet differentiated from the rest), and no ner- 
vous system (the arguments of Lendenfeld are inconclusive) ; there 
is also an absence of nematocysts. Finally, the symmetry of the Sponges 
is lower than that of the Ccelentera. The two groups agree in the blas- 
tula and gastrula stages, but no further ; hence the Sponges are a separate 
type — the Porifera . 
Goette (10) obtains very different results from his investigation of 
Spongilla fluviatilis ; unfortunately the memoir is too long to abstract. 
The following are some of the conclusions: — (a) The ripe egg is the 
product of one or several cells, which fuse to form a new Mono- 
plastid. (6) From the egg proceeds a sterroblastula ; this gives rise 
to a sterrogastrula, consisting of a small-Celled ectoderm, enclosing a 
large-celled endoderm, which encloses an eccentric cavity (endoderm 
cavity), (o) The embryo, by ciliation of the ectoderm, becomes the larva, 
which undergoes its metamorphosis within the parent, during swarming 
or after attachment, (rf) In the metamorphosis the ectoderm is com- 
pletely lost, and the actual Sponge proceeds entirely from the endoderm. 
(e) Its peripheral layer serves for the attachment of the Sponge, and then 
produces over the whole surface the epidermis ; at its base it passes into 
the endoderm mass ; the pores and oscules are cell-gaps of this layer. 
(/) In the endoderm mass, the flagellated chambers develope indepen- 
dently of each other and of other cavities by the budding of single cells, 
the canals and subdermal cavities arise as tissue-gaps, which are lined by 
epithelium derived from wandering cells. The spicules are iutercellular 
formations. 
Au account is given of the development of the gemmules, and the 
