66 



B. Arrangement of the spicules. 



It may be expected a priori that, where the elements of the skeleton are so variable, 

 the latter as a whole, i. e. the arrangement of the spicules will likewise show variation. Generally 

 we find in more or less massive specimens a diffuse skeleton, with no distinct spiculo-fibre, as 

 e. g. Ridley & Dendy described for Spirastrella massa, solida, papillosa etc. Indeed this seems 

 to be the case ; but if we compare it with more regular forms, we learn that in fact always 

 some main bundies of tylostyli can be distinguished, with so-called secundary fibres at about 

 right angles to them. In the meshes of massive irregularly shaped specimens we always find 

 spicules in all directions and this gives to the whole the appearance of a more or less dense 

 reticulation. At the periphery the tylostyles are usually smaller and project a little, more or 

 less regularly, forming as a rule distinct brushes or tufts. In specimens which form crusts, there 

 are distinct bundies of spicules starting from the substratum at about right angles and expanding 

 brush-like at the periphery — in longitudinal section fan-like. We see this arrangement distinctly 

 e. g. in 1945 ; we find there at the same time that the bundies may branch dichotomously 

 (PI. VI, fig. i). In conical or digitate processes we find likewise longitudinal main bundies, 

 branching and anastomosing. They give off small branches towards the periphery, where they 

 terminate in brushes; the main longitudinal bundies likewise terminate in a brush (PI. VI, fig. 5 

 and PI. VII). 



In specimens where spinispirae occur these spicules form a peripheral crust. If large and 

 minute spinispirae are present, the crust is formed by three or four layers of robust spinispirae, 

 mostly situated parallell to the surface, with some few spicules at right angles to it ; such layers 

 are then distally covered by a layer of minute spinispirae. This arrangement is undoubted in 

 1455 a _c and 1945. The main mass of spinispirae always occurs at the sponge surface; but in 

 the whole of the parenchyma they may occur without any definite order. 



The more the spinispirae become scanty or disappear, the more distinct are tylostyles, 

 placed perpendicularly to the surface and finally wholly replace the former. In many cases 

 they may become so densely packed that they form a pseudo-cortex, which is clearly visible 

 to the naked eye, often very thick and somewhat glossy in appearance (PI. V, figs. 1, 3, 

 15 and 16). 



The question whether in Spirastrella purpurea we have to speak about a true cortex 

 or not I cannot answer for the moment. We had better leave the question open. So much 

 is certain that we may distinguish an ectosome from a choanosome. In this ectosome we 

 often find one or more layers of spinispirae ; in addition to these spicules we find tylostyles, 

 perpendicular to the surface and generally smaller than those of the main skeleton. If spinispirae 

 are absent, the tylostyles may entirely so to say fill the ectosome. The tissue of the ectosome 

 is distinguished from that of the ectosome, because the cells are situated farther from each 

 other and consequently it is less stained. Moreover we often find fibres and long fusiform 

 cells. It is, however impossible to draw a sharp line of demarcation between ectosome and 

 choanosome. 



