HV E. P. HALLMANN". 



775 



brush of spicules, the outer extremities of which impinge upon and 

 sometimes slightly project beyond the dermal membrane (PL 

 xxxvi., fig. 1). The brushes occur so closely situated that their 

 spicules intercross with one another, and the latter are so num- 

 erous that collectively they produce an appearance as of a con- 

 tiLUOus, somewhat halichondroid layer of spicules underlying 

 and supporting the dermal skeleton. The intervening zone be- 

 tween this subdermal layer and the central skeleton, traversed by 

 the radial fibres, constitutes the second or intermediate region of 

 the skeleton above referred to; and the subdermal layer itself, 

 the third. It is these two regions of the main skeleton which, 

 together with the dermal layer, are removed by the action of 

 caustic potash. 



The main longitudinal fibres of the skeleton range in stout- 

 ness from about 140 to '260^, the connecting fibres between them 

 from about 45 to 18Qjx. Their constituent spicules, as a rule, 

 are not so aggregated as to form a compact axial core sur- 

 rounded by a zone or sheath of spongin free from spicules, 

 but tend to be dispersed more or less completely throughout the 

 entire spongin substance. The radial fibres (which, as already 

 stated, are composed solely of spicules) are scarcely less in 

 stoutness than the main fibres. The connecting fibres, when 

 shortest, are of a single spicule's length, and contain sometimes 

 only three or four spicules; but more frequently they are 

 longer, and occasionally the number of their contained spicules 

 is such that they approximate in appearance to the main fibres. 

 The irregular pattern of the skeleton-reticulation is due partly 

 to the fact that the main fibres run rather crookedly (and, 

 hence, at varying distances apart), and partly to the fact that 

 the direction of the longer connecting fibres between them is 

 usually more or less oblique. Between the fibres very numer- 

 ous loose megascleres are scattered, the great majority of which 

 are similar or nearly similar in kind to those composing the sub- 

 dermal brushes; but amongst them there are also very many 

 which are identical in shape and size with the dermal spicules, 

 and also a moderate number similar in every respect to those of 

 the fibres. The microscleres, both toxa and chelae, occur scat- 

 tered everywhere in moderate abundance, except in the dermal 

 layer. The chelae are of two kinds, but the smaller are very 

 difficult to perceive, in ordinary sections, owing to their minute 

 size. 



