8 THE VOYAGE OF H.M.S. CHALLENGER. 



bomba, and some other species. Beginning with the gastric surface, we find here numerous 

 large canals, which ramify tolerably regularly, like the branches of a tree, towards the 

 dermal surface. The branches, broad at their beginning, become smaller and more 

 numerous in their course towards the dermal surface, and pass at last into the finest 

 small canals, opening by pores on the dermal surface. The canals are, either throughout 

 their whole extent or for the greater part of it, covered with flagellated endodermal cells. 

 Only the outermost ends of the finest canal-branches (near the dermal pores), and the 

 innermost ends of the largest canal-trunks (near the gastric cavity), remain free from the 

 flagellated epithelium. The anastomoses of the branches are either wanting altogether 

 or few in number." 1 The diagram illustrating this description refers to Leucandra 

 bomba (loc. cit., Bd. iii., pi. xl. fig. 9). I was not able to obtain this sponge. I hoped to 

 find it in the Godeffroy Collection in Hamburg, which was kindly sent to me by the 

 administration of the museum ; but this expectation was not realised. I found, however, 

 amongst the Challenger Calcispongias three specimens of Leucetta primigenia, of a form 

 even more instructive with respect to the question of the existence of the dendroid 

 canal system than Leucandra bomba. "Leucetta primigenia," says Hasckel, 2 "as the 

 conjectural radical form of the Leucones, is so closely allied in the properties of its 

 skeleton to the general radical form of all the Calcarea, i.e., Ascetta primordialis, that 

 it can be derived immediately from this latter. The wall of Ascetta primordialis 

 requires only to grow thicker, its variable dermal pores require only to become constant 

 canals, and to ramify in the wall, in order to realise the transformation of Ascetta 

 primordialis into Leucetta primigeniar We shall soon see that Leucones have had 

 quite a different course of development ; at any rate, with regard to the canal system 

 of Leucetta primigenia, Prof. Hseckel's statements do not correspond with the reality. 

 A close examination of the three specimens above mentioned showed that their canal 

 system possesses just the same character, and, wdth the exception of some trifling differ- 

 ences, the same peculiarities as that of Spongelia, Aplysilla, &c. Its detailed description 

 will be given later; at present we have only to notice that a dendroid modification of 

 the canal system does not exist at all. 



" The retiform (' netzformig ') modification of the canal system takes origin from the 

 dendroid in the following manner : the anastomoses of the ramifying canals grow more 

 numerous and occur not only between the finer branches, but also between the larger 

 ones. When this structure has obtained its highest development, the wall-parenchyma of 

 a Leucon seems to be pierced by a dense net of canals, like a gland rich in blood-vessels. 



This modification is not rare it is to be found, for instance, in Leucetta 



trigona, Leucaltis Crustacea, Leucandra cataphracta, and Leucandra stilifera." In these 



words Hasckel describes his "retiform" type, illustrating it by a diagram referring 



to Leucandra stilifera; but I am forced to deny its existence, as also that of the 



1 Kalkschwan me, Bd. i. p. 228. 2 Loc cit., Bd. ii. p. 122. 



