REPOET ON THE TETEA CTINELLID A. 
Cl 
will in their case be one and the same thing, but it is different in the more primitive 
groups, for here the resemblance of a group to a tree is unfortunately too apposite, since 
so many of the leaves borne by different branches are essentially similar to one another, 
i.e., similar genera exist in the lower groups such as Sponges, which have nevertheless 
a different ancestry ; thus the Monaxonids are evidently a highl}^ polyphyletic order, 
some of them having arisen from near the root of the Demospongise, and others from its 
highest branches, and still others from intermediate points. And if we attempt to 
classify them according to descent we shall have to place in different groups genera 
which are structurally similar, and this would be equivalent to the method of a chemist 
who should attempt to classify a homologous series of compounds, not according to their 
structure, but according to the methods which he had employed to produce them. It 
may be urged that the similarity alleged is apparent merely, and I will not attempt to deny 
that this is probable ; future investigations may reveal important differences at present 
not dreamt of, but till these are made clear, it would seem better to classify by the 
similarities we do perceive than by a supposed phylogeny that may be wholly illusory. 
It is possible that the concrescence of the choanocytes which occurs in all the higher 
Tetractinellids may eventually serve to distinguish them from the Monaxons proper. 
Those Monaxons which have descended from the various families of the Tetractinellida 
might then be included by some such definition as the following: — Tetractinellida, 
Demosponges characterised hy desmas, triaenes, or tetraxons, and ivhere these are absent 
by the concrescence of the choanocytes. 
But at present our knowledge of the minute characters of the Monaxons is not com- 
plete enough to enable us to judge of the value of this possibly distinctive character ; in 
such Suberites as I have examined there is no concrescence and I hold this provisionally 
as completely separating this family from the Tetractinellids and as uniting it with the 
Monaxonids ; so too there is no concrescence in the Tethyidse, which must therefore also 
be assigned to the Monaxonids ; on the other hand, in the purely Monaxonid Sponge 
Amphius, which on morphological grounds we conjecturally derive from the Stellettidse, 
concrescence has been observed, and there would thus appear to be good reason for 
assigning it to the Tetractinellida as a degenerate or simplified form.^ The characters of 
the choanocytes in Placospongia have not yet been made out, and thus although the 
presence of the sterraster in this Sponge would suggest its affinities with the Sterrastrosa, 
yet till we know whether the choanocytes are concrescent or not it may be as well to 
suspend judgment. For the present, therefore, I am content to use for the larger groups 
of Sponges names which are better defined by the contents of the group than by any 
form of words. 
^ There is no concrescence in Placina, and it is absent from species of Tetilla, bnt these are the simplest 
members of the order ; the Suberites and Tethyidse are Corticate Sponges and without concrescence, while all Corticate 
Tetractinellida present this character. 
