( 1143 ) 
If we can accept but one species of Sprrastred/a, the question 
arises whether perhaps certain groups of specimens point to a closer 
relationship to each other? Is it possible to distinguish groups according 
to the depth or to certain properties of the bottom on which they 
are found? The first question can be answered in a positive sense, 
the others not yet. Indeed I believe that about seven such groups may 
be distinguished, although by no means sharply, as they are all 
intimately connected. But if we unite a certain number of specimens 
which seem to be nearly related to each other, into groups and try 
to bring them into connection with their conditions of life, we soon 
find that hardly any conclusion can be drawn. For there are speci- 
mens from the same locality which do not belong to the same 
group; consequently we cannot speak of so-called local varieties. 
The general structure of Spirastrella purpurea is, in spite of 
great differences, still fundamentally the same for various specimens. 
I will give a sketch of one form, viz. of a specimen corresponding 
to Drnpy’s Suberites inconstans var. digitata. A longitudinal section 
through one of the long finger-shaped processes shows us a central, 
more or less cylindrical cavity, generally opening at the top of the 
process. Such wide central canals are often met with in Porifera. 
The terminal opening is usually called “osculum” and the canal is 
considered to be of an excurrent nature. Indeed it is observed 
in many living Sponges that a current of water flows out from 
the “osculum”, the water entering through numerous little apertures 
on the sponge surface. Microscopical examination of such Sponges 
has shown that the latter apertures communicate with “flagellated 
chambers”. These chambers on the other side communicate again 
with a system of cavities, which open into the central canal, mentioned 
above. Now we know that the shape of the flagellated chambers and 
the position of the choanocytes determine the direction of the water- 
current. Consequently we can conclude from the shape of the 
chamber, which of the communicating cavities is excurrent, and 
which is ineurrent. The whole canalsystem can thus be reconstructed 
by studying series of sections. This has been done for several Sponges, 
but not as yet for the majority. Since it was found in certain cases, 
that the large central canal belonged to the exeurrent system, the 
conclusion was drawn per analogiam for other cases, that the said 
canal is excurrent. 
However, in Spirastrella purpurea certain features seemed an 
obstacle to considering the central canal as a “cloaca”. True, in 
several specimens the canal opened with a comparatively wide 
mouth, but in others the aperture is rather small. Moreover the canal 
