It is as distinct and bright in dried specimens (B. M. 12, M. L. B. 132, M. L. B. 22) as in such 

 which are preserved in spirit (B. M. 32, B. M. 33). As suggested by Ray Lankester it is callcd 

 spongioporphyrin by Mac Munn (1900). [Cf. infra]. 



Previous authors of course remarked a curious feature in the appearance of our sponge, 

 viz. the agglomeration of foreign, mostly calcareous objects on the surface, more especially at 

 the base. We will see that such calcareous débris and sand often occur also inside the sponge. 

 If it is true that Spirastrclla purpurea begins its post-larval life by growing over and between 

 corals and coralline algae in the way we described it above (p. 37) the fact is easily explained. 

 [Cf. infra] . 



E. Anato m y. 

 PI. IV, figs. 2, 3, S and 9. PI. V— XIV. 



It may be expected a priori that Spirastrclla purpurea in its general anatomy is as 

 variable as it is in its external appearance. It is evident that the system of canals, the arrange- 

 ment of the spicules, the concistency of the parenchyma etc. are different in different stages 

 of development; that all these systems of organs have another aspect in thin crusts than in 

 cushion-shaped forms or in massive, spherical forms or in such with longer or shorter processes. 

 In fact we will see, as we occasionally already noticed, that the skeleton is very variable, that 

 in some instances there are large canals visible, in others not. 



It is exactly because of this enormous variability that it seems advisable to describe a 

 couple of typical examples before we can give a comparative anatomy of the various modifi- 

 cations of Spirastrclla purpurea. 



1. Description of the anatomy of a typical example of tropus tegens. 



I will choose for this purpose specimen 1945. If a section at right angles to the surface 

 is made of a carefully decalcified portion, we see with the naked eye a number of holes, indicating 

 the places where calcareous matter of the substratum was before decalcification. If the calcareous 

 substratum is an uninjured, healthy coralline alga, large portions of the cellulose tissue remain 

 visible in its natural shape. But in other parts, where the plant is destroyed by boring organisms, 

 an amount of débris of all sizes remained. In sections of decalcified specimens such places will 

 look like empty holes. Bordering these holes, resp. covering the algae, corals etc. the sponge 

 substance is seen, which, to the naked eye does not show conspicuous holes or canals. Slightly 

 magnified, the latter become visible (PI. V, fig. 9). If we magnify our section a little more, it 

 is easy to recognize that these holes and canals belong to the canal system of the sponge, and 

 that this cannot be said of the larger holes, mentioned above. In fig. 1 on PI. VI I have given 

 a portion of the section of PI. V, fig. 9, now magnified about fifty times. The small apertures, 

 which are seen on the sponge surface (PI. V, fig. 22) appear to be for a great part incurrent 

 openings, stomata, which lead by short canals into wider reservoirs (PI. VI, fig. i,i). These 

 reservoirs run almost parallel to the sponge surface, often unite with other similar ones and 



