BY E. F. KALLMANN. 419 



to the surface) aeanthostyles are abundantly developed; these 

 superficial aeanthostyles are located entirely upon the external 

 aspect of the fibres supporting- them, and are thus directed per- 

 pendicularly to the surface with tlieir apices outwards. 



In the outermost region of the main skeleton, a considerable pro- 

 portion of the short spongin-fibres, whose disposition is more or 

 less at right angles to the surface, ensheathe each the basal portion 

 of one or several of the outwardly-projecting long tylostyli to 

 which is due the already-mentioned hispidity of the surface. The 

 dermal skeleton proper consists of scattered clusters and bundles of 

 styli and oxea (auxiliary spicules), which are mostly directed more 

 or less parallel to the surface, and, contrary to what usually is the 

 case in Raspailia, are never disposed in outwardly-directed diver- 

 gent tufts situated around the points of exit of the long projecting 

 spicules. 



Spicules.— (a)The principal megascleres are styli and tylostyli 

 and intermediate forms, together with relatively very few oxea. 

 The styli and tylostyli (the latter of which are the more numerous) 

 are sharp-pointed and more or less curved spicules, typically with 

 the curvature most pronounced in, and often restricted to, their 

 basal moiety; the very slenderest are not infrequently flexuously 

 curved (flagelliform). They range from about 4'_'0 to 1580 /x in 

 length, and attain a maximum diameter of 15//. The two forms, 

 styli and tylostyli, show some degree of differentiation from each 

 other, but not sufficient to admit of their separation into two 

 groups. The styli are, in general, the shorter and relatively stouter 

 spicules (being rarely less than 11 /x in diameter), and, unlike the 

 tylostyli, are usually a trifle stouter towards the middle of their 

 length than at the base. The tylostyli, which usually have only a 

 slightly developed phyma, are very variable in stoutness (the slen- 

 derest of them being less than 3 /x in diameter), and are seldom 

 below 800 /x, and rarely, if ever, below 500 /x in length. The 

 slenderest spicules are usually not expanded at the extreme base, 

 but at some short distance above it, and then not as a rule bul- 

 bously, but elongately and somewhat irregularly; and a consider- 



