ANATOMY AND PHYSIOLOGY. 
475 
the whole of the nutritive (digestive, respiratory, and secretory) functions; 
and there can be little doubt that both eggs and spermatozoa are modified 
endodermal cells. A cell draws in its collar ” and “ flagellum,” increases 
in size &c., a nuclcolinus ” makes its appearance ; when free the egg as- 
sumes the shape of an Amoaha and moves about, sometimes penetrating into 
the exodorm, or emigrates, in the oviparous species, from the stomach ” to 
be fecundated abroad. In a few instances the egg is invested by a thin cal- 
careous shell. The spermatozoa are apparently developed through repeated 
divisions of modified endodermal cells ; the “ head ” is formed by the 
“nucleus,” the “tail” by the protoplasm of the minute sperm-cells. The 
author has observed the act of fecundation, which is apparently a penetration 
into the egg of the spermatozoa, and their subsequent fusion with it. He 
discovered spermatozoa in species of all families, but only in few species 
altogether ; in these male specimens eggs also were developed. “ Gemmuloi ” 
never occur in Calcispongits. The transfusion of the sea-water is effected 
in the Ascones (the simplest type) through unstable microscopical “der- 
mal pores,” which open and close at leisure : though their place is, to a cer- 
tain degree, dotorrained by the meshes of the skeleton, now pores will form 
independently of their predecessors; under certain circumstances (when 
over-saturated with food during low tide, or_ when the water is bad) they 
will all close up ; under others (e. g. when put in clear water) they will open 
again to the utmost : general contractions or dilatations of the whole “ syn- 
cytium ” depend on similar circumstances. The “ mouth,” whether naked 
or produced into a “ trunk,” formed of exodermal tissue alone, may shut up 
through temporary fusion of the “ syncytium ” (when crowned with a circle 
of long “ spicula,” through the medium of a collar-membrane) and reappear 
again. The normal course of the fluid is inwards, through the pores into 
the stomach, by the impulse of its flagellate colls, and out of the mouth ; 
but the reverse may take place ; and when the mouth is wanting; either 
through innate “ lipostomy ” or through secondary coalescence, a portion 
of the pores must act as inhaling, another as exhaling orifices. In the 
Leiicones the gastric wall (the syncytium) is very thick, but perforated by 
ramified channels, opening into the stomach with larger “ gastral ostia,” on 
the dermal surface with delicate unstable pores, or with larger dermal 
“ostia; ” they can also open or shut, but always in the same definite place. 
Tills channel-system may be modified after different types — the dendroid, 
reticular, vesicular, or glandular; in the last two the flagellate cells and 
their functions are confined to the larger cavities of the system (the “ ampul- 
laceous sacs ” in siliceous sponges) ; in the larger tubes they are deficient, 
as they are also in the “ stomach ” of all Leucones and Sycones. In cases of 
“lipostomy” also the stomach is sometimes obliterated; in some rare cases 
it is divided into compartments by perforated exodermal tissue. In the 
Syconcs the gastric wall has originally the same structure as in the Ascoyiciff 
but crops out (through a sort of “ strobiloid gemmation,” according to the 
author) into a multitude of short radiating cones or tubes, which in some 
species remain separate, but in others coalesce more or less completely with 
their borders or surfaces, forming together a system of prismatic or cylindric 
“ marginal cavities,” and commonly of triangular, quadrangular, &c. inter- 
stitial tubes of astonishing regularity ; each of these excrescences has the 
chief structure of an’ Olynthus. The typically simple Asconcs may also pre- 
