as far as I know, 36 species of Spirastrclla ; about 8 species, placed into other genera 

 ought to be included ; hence we have 44 species. Some of these are quite insufficiently 

 described to allow us to form an opinion ; others cannot remain under Spirastrclla because 

 of their spiculation. Of the remaining 34 there are two, S. coronaria and ó\ spinispirulifer \ 

 about which it is open to discussion whether they are to be considered as species of our genus 

 or as belonging to a separate genus. The rest, according to me, represent specimens of one 

 species: S. purpurea. 



Provisionally we may give the following diagnosis of the genus: 



Sponge encrusting; or cushio n-s h a p e d, w i t h o r without tuberculiform 

 to digitiform o r conical processes; or massive, irregularly spherical. 

 Canal system eurypylous. Skeleton mainly co m pose d of bundies of styles 

 or (and) tylostyles. Typically there is a der mal crust of spinispirae, but 

 in some cases these spie u les are exceedingly scarce or even wanting 

 altogether. 



It seems to me beyond doubt that the nearest ally of Spirastrclla is the well-known 

 genus Cliona. Whereas the latter begins its post-larval life by boring into calcareous matter, 

 Spirastrclla never does so. Both genera agree in their typical spiculation : tylostyles (resp. 

 styles) and spinispirae ; they agree also in the fact that the latter spicules may become scarce 

 and finally disappear. This is very disappointing for dogmatical taxonomists; but it is a fact 

 with which I am acquainted since long and to which I believe Topsent first drew attention in 

 1888. If I state, that I learned this truth more than twenty-five years ago, when in Naples I 

 studied the development of a certain Desmacidonid sponge, it is by no means to make a 

 desperate trial for priority, but to show how the phenomenon is not a speciality of Cliona and 

 Spirastrclla, and how Topsent and myself independently arrived at the same result. Ridlev 

 suggested (1884 p. 468) whether Cliona and Spirastrclla justifïed a generic separation. I think 

 the answer must be affirmative. In spite of Topsent (1909 p. lxx) I likewise retain Poter ion 

 as a genus. All three are closely allied, but they are generically different. I am quite satisfied 

 that Topsent admits my short statement, that Poterion begins its life as a "boring sponge". 

 I may add now, that in the course of my investigations I discovered that at least once it is 

 found in its boring state by other authors, viz. by Ridlev & Dendv and described by them 

 (1887 p. 224 — 229) under the name Cliona dissimilis. Topsknt seems to me, however, "plus 

 royaliste que Ie Roy" if he calls the Neptune's cup " Cliona patera \ Am I wrong if I suppose 

 that Topsent never examined well preserved alcohol-specimens of Poterion ? I cannot discurs 

 the matter fully now, as long as I have not yet published my observations after good sections 

 of Poterion, and I can only say that there are sufficiënt anatomical grounds to keep Poterion 

 and Cliona apart. I beg my French colleague to postpone his judgement, hoping for a little 

 more clemency than in the case of the identity of Cliona eclata and viridis, the arguments 

 wherefore he frankly states to wait for "sans la moindre confiance". 



Cliona, Spirastrclla and Poterion all have in common that they are highly polymorph 

 according to circumstances. In their early youth the first and the third live in holes made by 

 themselves in calcareous matter; the second covers as a thin coat foreign objects, as it seems 



