320 NATURAL HISTORY. 



A folding of the entire wall of the Sponge follows (this is an irregular form of budding), and converts 

 the originally simple central cavity into smaller canal-like spaces, in other words, it becomes a 

 branched excurrent canal system ; the interspaces between the folds outside the Sponge wall become 

 the incurreiit canal system. 



Histology. The ectoderm appears to maintain its pavement epithelial character very constantly, 

 but sometimes its cells become flagellated, as in Halisarca and Plakina. The endoderm undergoes 

 no great variation. The mesoderm, on the contrary, differs a good deal in different Sponges; 

 in many it consists, as in Euspongia, of a clear jelly-like matrix, embedding branched granular 

 corpuscles ; in others it becomes densely charged with minute granules, maintaining throughout the 

 character it presents locally about the flagellated chambers of Euspongia, while in some it appears to 

 consist of separately-outlined granular cells. The clear granules, which fill some of the wandering 

 amcebiform cells, are, in some cases, certainly starch. The fusiform cells of the mesoderm are often 

 abundantly developed, and sometimes form a thick layer beneath the skin, having the appearance of 

 fibrous connective tissue, but where the main water-canals pass through it, this layer is modified 

 to form around each of them a distinct sphinctral muscle. 



TJie Skeleton. The character of the skeleton is wonderfully diverse, and since it is fairly constant 

 within each species it affords us the best means of classification. Some Sponges, such as Halisarca, 

 are entirely destitute of a skeleton, others (Lithistids) are possessed of one of stony hardness, which 

 no one would think of applying to skin except as a counter irritant. The skeleton may consist of 

 a network of horny fibres, the axis of which is either filled merely with soft granular matter, or in- 

 cludes also foreign bodies, often to such an extent as to convert the fibre into a veritable rope of sand ; 

 or, instead of foreign bodies, a core of proper spicules, i.e., spicules produced by the Sponge itself, may 

 be present; and the spicules may increase in number, and the horny matter diminish in quantity to 

 such an extent, that the fibre comes to consist only of spicules. The skeleton frequently consists whollv 

 of spicules, but these are far from being always arranged in a fibrous form. The spicules, which are of 

 most diverse forms, are composed of an organic basis (spiculin), densely impregnated or chemically 

 combined with a mineral salt carbonate of lime in the case of calcareous spicules, silica in that of 

 silicious spicules. This distinction in mineral composition was discovered by Robert Grant. The 

 spicule usually consists of a clear glassy wall, concentrically-layered, enclosing a soft thin axial thread. 



It will be convenient to state here that according to the character of their skeleton, the Sponges 

 may be divided into the following four orders : 



Myxospongin.* Soft Sponges, skeleton absent. 



Calcispongice. Skeleton consists of calcareous spicides, never united to form a fibre. 



Silicispongice. Skeleton characterised by silicious spicules, which may or may not be united into 

 a fibixms skeleton. 



Cerospongwi, Skeleton consists of a network of horny fibres, sometimes including foreign 

 particles, but never proper spicules. 



The simplest form of spicule is needle-shaped (acerate), pointed at both ends (Fig. 10, a). It grows 

 lengthwise from the middle along a single axis in the direction of the ends ; it is thus uni-axial but 

 bi-radiate. Supposing it to cease to increase in length at one end, it becomes an acuate spicule 

 (Fig. 10, 5), still bi-radiate, but the radii of unequal length ; if one radius does not develop at all and 

 is represented only by a globular enlargement, a pin-headed acuate results (Fig. 10, c], which is both 

 uni-axial and uni-radiate. If a third ray grows out from the side of the acerate spicule, a tri-radiate 

 but bi-axial spicule is the result (d) ; should all three rays diverge, so that no two are in the same 

 straight line, we have the tri-radiate and also tri-axial form (e) so characteristic of the Calcispongiae, 

 though by no means confined to them, since it occurs normally in many of the Silicispongiae, e.g., the 

 Plakinidse, Plectronellidre, and Sphinctrilla, and, as a variation, common, but abnormal, in a great 

 number of other instances. If a fourth branch or ray is produced from the centre, not in the same 

 straight line as any of the others, a quadri-radiate (also quadri-axial) form appears, arid this 

 characterises the sub-order Tetractinellidce, though it appears also in Sponges belonging to other groups. 

 The four rays may remain of the same value (/), and be disposed without any ascertained relation to the 

 form of the Sponge and its canal system (Pachastrellidse), or one ray may become distinguished from 



* Greek, myxa, slime. 



