288 REVISION OF THE MONAXONID SPONGES, 1., 



tliey at all substantiate his statement that these fields " are 

 divided from each other by sharply defined incisions " unless the 

 word " incisions " is used in a quite unusual sense. The two 

 specimens are of nearly the same height, viz., 120 mm., which is 

 20 mm. less than the maximum height recorded by Lendenfeld. 



Internal structure. — A transverse section across a branch per- 

 mits three regions to be distinguished with the naked eye : (i.) a 

 pale-coloured external layer, or cortex, which, in different parts 

 of the sponge, varies in width from about 0*8 to 1*5 mm.; (ii.) a 

 deeply brownish-coloured subcortical layer, usually much wider 

 than the cortex, but in widtli rather variable: and (iii.)a broad 

 axial region or core, also brownish-coloured, distinguishable from 

 (ii.) by reason of its being traversed longitudinally by numerous 

 spicule-strands, the cut ends of which show clearly on the surface 

 of the section. In the figure of the type-specimen {PI. xv., fig. 2) 

 a branch is seen in longitudinal section, showing the relative 

 extents of these three regions. In this example, however, the 

 subcortical tissue has mainh^ disappeared owing to maceration 

 (which it undergoes more readily than do the other tissues), and, 

 as a result, a system of fibres, crossing the subcortical layer and 

 passing into the cortex, is brought into view. Owing to these 

 fibres, the cortex cannot ])e peeled off separately, but, in its 

 removal, drags with it most or all of the underlying layer. In a 

 longitudinal section of a branch, ordinarily, the subcortical region 

 is not recognisable as a layer distinct from the core-region (Len- 

 denfeld includes them both under the tenn " pulpa ' ); but the 

 demarcation between the subcortical layer and the cortex is well- 

 pronounced, owing to their difference in colour, and to the 

 presence, immediately beneath the latter, of a narrow zone of 

 lacunar and canals. Lendenfeld 's statement that these lacunae 

 (and canals) form a " nearly continuous cavity " beneath the 

 cortex, appears to be somewhat exaggerative. 



On treatment with a macerating agent, such as caustic potash, 

 the two outer layers of the sponge readily soften and come away, 

 leaving intact the stout core, the thereby exposed surface of 

 which bristles sparsely with long spicules projecting, nearly per- 

 pendicularly, 1 to 3 mm. beyond it. The core is very resistant 



