118 Mr, H. J. Carter on Fossil Sponge-spicules of 



that there are three canals in the centre diverging from a common 

 one, which proves that, from the simplest ti-ificl or ternate head 

 to the most elaborately divided one, all begin with a trifid or 

 ternate development — in short, that all are some multiple of 

 three, and therefore that in it we have a distinct trifid or ter- 

 nate system from the beginning for the grouping of the large 

 spicules with which we are now concerned, which character 

 is the most practicable, at least for our purpose. 



In this view I have been alike anticipated by Dr. Bower- 

 bank (1858) and Mr. Parfitt ij. c. 1870) independently, the 

 former of Avhom, in 1869, writing on the " siliceo-fibrous " 

 sponges, observes : — " The apices of the connecting spicules 

 are exceedingly various in their form ; but they are all modifi- 

 cations of a triradiate one, even the peltate forms " (Proc. cit. 

 p. 73). I prefer much the term "ternate" or "trifid" to 

 " triradiate," because the former apply to the branching of a 

 stem (the axial canal), and the latter to a branching or radia- 

 tion from a point, since this avoids a confusing of the tri- or 

 quadi'iradiate spicules of the CalcispongiiB in particular with 

 the trifid or temate division of those of the Coralliospongiffi and 

 Pachytragiae ; and as we are most familiar with the term " ter- 

 nate," I shall henceforth use this with its necessary prefixes — 

 a grammatical violation, it is true, but one, perhaps, which the 

 desirableness of using short instead of long cumbrous terms 

 may sanction. 



From the ternate system of the peripheral spicules let us go 

 to the silicified fibre of the interior ; and here we have all the 

 figures from 10 to 29 inclusively illustrating this structure, 

 many of them, no doubt, somewhat worn by trituration at the 

 time of their deposit, but otherwise the irregular knot-branch- 

 ing of the Dactylocalycidffi, and the more rectangular hexra- 

 diate one of the Euplectellidae, together with the canalated 

 fibre of Farrea occa (Bk. Proc. cit. 1869, pi. xxiv. fig. 1), all 

 find their representatives respectively in these figures, many 

 of which, also, are almost facsimiles of Schmidt's figures of 

 Lyidium torquila^ obtained by M. de Pom'tales in 270 fathoms, 

 off the island of Cuba (Atlantisch. Spong. Fauna, p. 84). 



We now leave the Coralliospongite and go to the heads of 

 the first and second divisions of the ternate system, for which 

 group I have proposed the name of " Pachytragije " {I. c.) ; 

 and here we revert to the condition of the deep-sea sponges, 

 so far as the absence of silicified fibre is concerned ; but instead 

 of the, for the most part, soft, silky nature of their spicular 

 structure, we have the short, rigid, ternately developed spi- 

 cules of the Pachytragiae, which grow and develope themselves, 

 in many instances, on the shore-rocks, where they are ex- 



