﻿144 



•ENDEAVOUR" SCIENTIFIC RESULTS. 



strongylote or tylote extremities. One is inclined to think, 

 therefore — in regard to those genera included in the Mycalin^e 

 which afford reason for believing that they are degraded Myx- 

 illina^ — that the megascleres of typical species of Desmacidon 

 and Homoeodictya are homologous with principal spicules 

 whilst in some species at least which have been assigned to 

 the same genera and to Amphilectus and Batzella they are 

 homologous with auxiliary megascleres. i The presence or 

 absence of spination on the principal spicules has been found 

 lacking in generic value, and everything points to the fact that 

 the spined condition is the more primitive ; it is perhaps not a 

 rare occurrence in species in which they are smooth when fully 

 mature, that they are spined in their early developmental stages 

 (as, for example, in My.xiila diversiancorata^), or during the 

 larval period of life of the sponge (as in MyxiUa pedunculata^). 

 It is a peculiar circumstance that the principal megascleres are 

 almost invariably curved ; exceptions to the rule are provided 

 by Raspailia and its allies, but it is significant that these are 

 also aberrant in other respects. In Echinodictyuiu and li'i- 

 kentrion amongst the "Ectyoninae," and in DendoriceUa and 

 some species of other genera {e-g-, lotrocliota) amongst the 

 "Dendoricinje," they are diactinal (as in Desmacidon and 

 Hotna'odictya), and take the form of oxea or strongyla ; vet 

 they still exhibit the curvature and fusiformity which, in 

 general, are characteristic of principal megascleres. 



Certain Myxillinae are possessed of more than three 

 kinds of megascleres owing to the division of one or 

 more of the primary groups into two kinds ; and this 

 division usually appears to be correlated with, and to have 

 been the outcome of a performance by the spicules concerned 

 of two different functions. Instances of such, in which the 

 principal megascleres have undergone division, are provided by 

 Echinodictyum clalhratum^ and species of Raspailia {e.g., R. 

 raniosa, Mont., and R. tenuis, R. and D.); and in which the 

 "accessories" have undergone division by the species of PIo- 

 camia. That the "dumbbell" spicules of the last-named genus 

 are derivatives of the accessory megascleres, the indication 

 afforded bv P. plcna^ leaves scarcely any room to doubt ; and 

 this species also, by reason of its possession of stunted ab- 

 normal forms of the principal megascleres, renders it probable 



1 Examples of such are provided by Hoinceodictya doidyi (Whit.) ; Bes- 



macicton vUcatum (= D. Mellidermn, Carter); J), psammodefi and 

 Batzella ino'qu'dis. (Hentschel, Fauna Sudwest-Australiens, Bd. iii., 

 1911); and. by Anivhilectus (rnitoshft, R. and D. 



2 Lundbeck— Porifera Danish Tngolf-Expedition, Pt. 2. 1905, p. 150. 



3 Lundbeck-ior. cit., p. 149. 



4 Dendv— Rept. Pearl Oyster Fisheries, Gulf of Manaar, Pt. 3, 1905, 



p. 176. 



5 Sollas— Ann. Mag. Nat. Hist. (4). v., p. 44. 



I 



