c 
THE VOYAGE OF H.M.S. CHALLENGER. 
Zittel are not the Tetractinellida of Marshall, this change in termination is both justifiable 
and convenient; Vosmaer’s Tetractina is, however, anticipated by my Choristida. 
The examination of the material brought home by the Challenger served to deepen 
my conviction of the close relationship of the Choristida and Lithistida, and since the 
researches of all other naturalists who have made a close study of these two groups have 
led them to similar views, ^ I shall retain both the name and the group first proposed by 
Marshall. 
The Tetractinellida are not by any means all pro'sdded with tetractinose or even 
trisene spicules. Trisene spicules though present in many Lithistids are absent from 
more ; and some sponges which may fairly be included in the Choristida [Placospongia 
for instance) are equally devoid of them, and also of tetractine spicules. 
The trisene spicule also is far more characteristic of the group than the tetraxon, and 
thus it would be quite as appropriate to term it the Trisenellida as the Tetractinellida. 
This, however, is not a matter of any importance, and as to the absence of trisenes and 
tetraxons, such cases are remarkably rare in the Choristida ; and though much more 
common in the Lithistida (whole families being without these spicules here), yet as the 
whole of this order is bound together by the constant presence of the desma, and since 
the ancestral forms from which all the others have descended are characterised by both 
trisenes and tetraxons, or tetraxonid desmas, the term can be appropriately used, just as 
Monaxonida may for a group which includes Sponges (Ceratosa) without Monaxonid 
spicules. 
There is one serious inconvenience attaching to the presence of Sponges in the 
Choristida which do not possess either trisene or tetraxon spicules, for these make it 
impossible to frame a short definition which shall include them along with the rest ; they 
find no acknowledgment in the definition here adopted for the Tetractinellida, and my 
apology for this is two-fold ; in the first place it is not absolutely certain, but only highly 
probable, that they are degenerate Tetractinellid forms, so that in including them in 
the group I have attached to them a query, and in the next place if they are truly 
immediately descended from the Tetractinellids, it may still, as a matter of convenience, 
be permitted to refer them to the Monaxonids, since they actually are Monaxonid, than 
to destroy the simplicity of our classification ; their phylogenetic descent may be left to 
be indicated by genealogical trees. 
These remarks open up the whole question as to the views which should guide us in 
framing a classification ; at present it appears to be generally admitted that a system 
of classification should be founded rather with reference to the blood -relationships of 
organisms than to the mere sum of their resemblances and differences in structure. 
With regard to the higher forms of life there is everything to be said in favour of this 
view, the more especially since a strictly morphological and a phylogenetic classification 
^ 0. Schmidt, Spong. Meerh. Mexico, p. 13, 1879 ; Ddderlein, Zeitschr.f. wiss. Zool., Bd. xl. p. 69, 1884. 
