repeated process of cleavage, by which means it quickly multiplies itself, though still more rapidly by the subsequent encystment and breaking up of the monads into spores."

It is quite unnecessary to recapitulate the varied phases of the controversy, which has become voluminous, on behalf of the theory advanced by Haeckel on the one hand and Professor Clarke on the other,1 since this would scarce possess interest for the general reader; but should any one feel desirous of investigating the subject, he will find an admirable summary in the chapter, "On the Nature and Affinities of the Sponges," in Mr. Saville Kent's "Manual."

According to those who accept the affinity of the sponge animal to the flagellate infusoria, first distinctly propounded by Professor Clarke, and further developed since his death, the essential spongestructure consists of three elements, namely, the collared flagellate monads, the hyaline mucous-like stratum (cytoblastema), and the amœboid bodies, or cells, to which is added the skeleton or framework already alluded to. It is claimed, on behalf of the flagellate monads, that they should hold the foremost position in the economy of the sponge, to which the mucous stratum and amœboid bodies are subsidiary.

1 See "Nature of Sponges," by Henry Slack, in Popular Science Review, vol. xi. p. 167. 1872.

The essential feature of the flagellate monads is, that they possess a film-like collar of membrane, not unlike a little funnel, which is capable of extension or withdrawal, enclosing within it a terminal flagellum, or whip-like thread, and at the other extremity contractile vesicles (fig. 18). As to the collar, it is stated to be not a mere funnel-shaped expansion of inert sarcode, but a most active organ, having, during life and when fully extended, a continuous stream of fine granular protoplasm for ever flowing up the exterior, and down the interior, surface of the collar, identical with the cyclosis exhibited in the pseudopodia of Foraminifera. It has also been demonstrated that " this collar, with its characteristic currents, is an exquisitely contrived trap, for the arrest and capture of its customary food, which, driven by the action of the central flagellum against the outer margin of the collar, adheres to it, and passes, with the onflowing protoplasmic stream, into the animal's body." These collared monads are always found lining special cavities, excavated within the hyaline mucous-like stratum (or cytoblastema). There is, of course, in different sponges, variability in

the size and form of these cavities, and their precise location. In some they are distributed in a general manner throughout the internal canal-system; whilst in others they are confined to spheroidal chambers, excavated within the substance of the body of the sponge, but freely communicating with the inflowing and outflowing streams.

The second element is the common gelatinous base, or hyaline mucous-like stratum, in which the two other elements are imbedded. Although of the same consistence throughout, this stratum consists of two layers; that is, of an investing membrane in which no monad chambers exist, and a deeper, thicker substratum in which they are immersed.

The third element consists of the innumerable amœboid bodies, or cells, scattered more or less abundantly throughout the substance of the mucouslike stratum. These bodies have no distinct cellwall, and unless specially sought after, are scarcely to be distinguished from the stratum in which they are imbedded. They vary much in outline, each furnished with a refractive nucleus, and are best seen in the investing membrane. "Like Ameba they are constantly undergoing a change of outline, and may also be observed to shift their position from one part to another of the matrix. Oftentimes their long, slender pseudopodia, radiating towards those of their neighbours, unite together, forming a complex network, which presents a remarkable resemblance to ganglionic corpuscles. It is undoubtedly through the stimulus received and transmitted by them that the characteristic contraction and expansion of the pores, or oscula, and other portions of the sponge body, are accomplished." 1 The relations between these amæboid bodies and the flagellate monads is at present little more than conjecture; but Mr. Kent does not regard the former as independent structures, but rather as larval, or metamorphosed, phases of the collar-bearing monads, it having been clearly observed that the latter, when their course is run, lose their collar and flagellum, and become amœboid.

This endeavour to explain, as briefly as possible, the economy of what may be termed the sarcodous, or fleshy portion of sponge structure, has been necessary in order to exhibit the complexity of what at one time was considered a very simple matter, and dismissed with little more than an intimation that it was called "sarcode," and was the amorphous flesh of the sponge. Even now there is undoubtedly much more to be known, and further investigation, proceeding as rapidly as during the past few years, will elucidate that which is still dark and uncertain. If, in each individual sponge, we should come to recognise a colony of unnumbered workers, myriads of monads, living in company, toiling for the benefit of the commonwealth, and weaving a home for themselves in the great deep, enlarging, extending, and increasing the colony day by day, it will simply be a repetition, in another form, of the same story as told by the zoophytes, the sea-fans, the sea-mats, and the architects of the Coral Islands.

"Each wrought alone, yet all together wrought
Unconscious, not unworthy instruments,
By which a hand invisible was rearing
A new creation in the secret deep."

The entire sponge, in its living state, is enveloped by a sort of skin or dermal membrane, coated in the inside with sarcode, and strengthened in various ways by fibrous tissue, or spicules. This membrane has the power of opening and closing pores on any part of its surface, through which the animals breathe, or receive nutriment. Beneath these pores are large irregular cavities, which receive the water imbibed by the pores, and convey it by an inward current into the canals, which branch and ramify through all parts of the sponge, becoming smaller and smaller as they divide and recede. There is always a double series of canals, those which convey the water charged with nutriment, by an inflowing stream, to the remotest parts of the sponge, and those which collect the exhausted water, and transmit