2 Spong. 
SPONQOZOA. 
Evolution, Anatomy, &c. 
Metschnikofp (8) has convinced himself that a true ectoderm (dif- 
ferent from the skeletogenous layer or “ syncytium ”) exists in different 
Sponges (lieniera, Halisarca) ; the Sponges therefore are not bifoliate but 
trifoliate creatures, and the extracellular substance of the mesoderm is 
probably homologous with the gelatinous substance of the Medusce. The 
two layers observed in the larvae of different Sponges are the ectoderm 
and the mesoderm; the endoderm is a later production ; the spicules are 
developed in the hitherto so-called endoderm of the larva, which is in 
fact the mesoderm. These views, which in several important particulars 
rectify the results previously attained by the same author [Zool. Rec. xi. 
p. 532], agree well with the results obtained by Barrois (1) ; but, though 
this author cordially acknowledges the brilliant light ” thrown upon 
the Calcispongia by Hackel’s monograph, ];iis results are much at 
variance with the ideas concerning the evolution of these organisms 
promulgated in that work [Zool. Rec. ix. p. 474], only partially based 
on observation, and for the greater portion on combination and imagi- 
nation. The egg of Sycandra, &c., takes its origin in the skeletogenous 
layer (the mesoderm) below the endoderm; the segmentation is total 
and regular ; the blastodermal cells are differentiated in such a manner, 
that those of the larger anterior portion become elongate and prismatic, 
the posterior roundish and dark. These large posterior cells invaginate 
themselves in the hemispherical cap formed by the prismatic anterior 
cells ; but this invagination is only transitory and devoid of morphological 
importance, like the obliterated central cavity of segmentation. At this 
period, the flagella are developed from the prismatic (ectodermal) cells 
and the embryo liberates itself by pushing its way through the endoderm 
of the parent sponge. Perhaps the uppermost ring of the largo posterior 
cells constitute the starting-point for the evolution of the future meso- 
derm [?]. The larva fixes itself by its posterior extremity, and is rapidly 
transformed into a minute flattened crust-like sponge. It is now formed 
of two layers, an inferior of dark protoplasm, and a superior thin trans- 
parent layer, showing amoeboid movements and pushing out pseudopodial 
expansions for attachment ; vacuoles and irregular pores are also formed 
in this superficial layer ; there is no interior cavity, the canals are new 
formations, hollowed out in the mesoderm, from which the endoderm is 
not yet distinctly separated ; the rod-like spiculae are formed before the 
radiate spiculse ; the irregular shape of the young sponge is transformed, 
during its growth, into a more cylindrical one ; the “ osculum ” is 
formed, &c. Sometimes several larvae will attach themselves so close to 
each other, that they coalesce superficially and provisionally, or com- 
pletely and definitely with their pseudopodia. In Halisarca the seg- 
mentation of the egg is more irregular ; there is, at first, no difference 
between the anterior and the posterior cells of the embryo ; there is no 
invagination, and only a single layer of cells. When the larva is set free 
and covered with cilia, the posterior cells are distinguished by their 
shorter cilia ; they afterwards increase in size, while the other blasto- 
