XXVlll 
THE VOYAGE OF H.M.S. CHALLENGER. 
the primitive paragastral canal, and subsequently from the margins of the apertures of 
the derivative canals. 
Canal-System of the Siphonia type. — This remarkable type of Canal-system is met 
with in such widely differing sponges as the fossil Lithistid Siphonia,^ which is 
represented in the Challenger collection by the recent species Eusiphonia superstes ; a 
Monaxonid sponge, Petrosia [Sipho7iia) typa, Blainville,^ which from the remarkable 
resemblance of its canal-system to that of Siphonia was mistakenly identified by 
Blainville with that sponge, and in Taxoploca {Emploca) ovata ^ a fossil Hexactinellid 
sponge. In these and similar sponges the form is more or less ovate or cylindrical, and 
the main excurrent canals radiate from an axial cloaca, which opens in an apical oscule ; 
those that proceed from the base of the cloaca continue its direction downwards, those 
from the sides extend outwards and downwards in curves which become more parallel to 
the surface the nearer they approach it ; these longitudinal curved canals have all origin- 
ated at the surface of the sponge, and those now lying deep within it serve to indicate its 
lines of growth ; since they have originated along meridians of the surface of the sponge 
these canals may be termed “ meridional." They are of the same nature as those 
discussed in the last paragraph, f.e., paragastral; whether they are formed as an evagina- 
tion of a sponge-plate or simply of the endoderm, is not certain, but that they are either 
one or the other follows from the following considerations, — in the first place if the 
skeleton of a fossil Siphonia, or of a recent Neosiphonia, be examined it will be found that 
the most superficial canals are incomplete on the outer side, forming mere grooves which 
extend from the margin of the oscule over the adjacent surface, those lying a little 
deeper are converted into complete canals adjacent to the oscule, into which they open 
by completely circumscribed apertures, but come to the surface further away from it as 
grooves, and so terminate ; from these observations it follows that the meridional canals 
originate near the margin of the oscule and subsequently extend towards the antoscular 
pole. If next a spirit specimen of a Lithistid presenting this type of canal-system 
be examined, it will be found that the most superficial excurrent canals proceed from 
the oscule as subdermal cavities, bounded below by choanosome and externally by 
ectosome, and are unprovided with corresponding incurrent sinuses; it is difficult to 
explain them as foldings, and in all probability they are outgrowths of the endoderm. 
That the deeper-lying canals were originally superficial is shown by breaking up a 
specimen of a recent Neosiphonia or Petrosia, when it will be found that successive 
concentric layers can be peeled from these sponges, each layer as it is removed exposing 
a previously existing surface ; the meridional excurrent canals are exposed at the same 
time, and they are always found to lie conformably with the surface on which they appear. 
^ Sollas, Quart. Journ. Geol. Soc., vol. xxxiii. p. 805, pi. xxv., 1877. 
2 Sollas, tom. cit., p. 795. 
® Sollas, op. cit., vol. xxxix. p. 541 ; as I find that the name Emploca was preoccupied at the time I adopted it, I 
take this opportunity to substitute for it another — Taxoploca. 
