630 REPORT— 1902. 



and to show that if we allow for adaptive change, its characters, well knowD, 

 are constant within the limits of its simpler forms. ^" 



It is now more than forty years ago that the late Laeaze-Duthiers described 

 for Dentalium a larval stage, characterised by the possession of recurrently 

 ciliated zones, which by reduction, with union and translocation forwards, give 

 rise to the trochal lobe.'-* It is now known that in the American pelecypod 

 Yoldia Umatilla a similar stage is found, in which a ' test,' of five rows of ciliated 

 cells, is present ; '-^ and of the young of Dondersia hanyulensis the like is true. 

 But whereas in the Yoldia the ciliated sac is ultimately shed, in the Myzomenian 

 the escape of the embryo is accompanied by rupture, which liberates the 

 anterior series of ciliated zones in a manner strongly suggestive of forward con- 

 centration, leaving the posterior circlet with its cilia attached.'-"^ 



This ' test ' has also been seen in two species of Nucula, and pending fuller inquiry 

 into the Myzomenian and a reinvestigation of Dentalium, I would suggest that 

 this recurrently ciliated sac is representative of a larval stage antecedent to the 

 trochophore, for which the term protrochal may suffice. This term has indeed 

 been already applied to a larva ot certain Polychseta, which might well represent 

 a modification of that for which I am arguing;'-' and quite recently it appears to 

 have been observed near Ceylon for a species of the genus Marphysa}''^ 



The discovery of this larva in Dondersia was accompanied by that of a later- 

 formed series of dorsal spicular plates, which for once and for all, in realising 

 a chitonid stage, demolish the heresy of the ' Solenogastres,' mischievous as 

 suggesting an affinity with the worms. Like that of the supposed cephalopod 

 affinities of the so-called ' Pteropods,' it must be ignored as an error of the past. 



Returning to the protrochal stage, whatever the future may reveal concerning 

 it, by bringing together the Lamellibranchiata, Scaphopoda, and Polyplacophora, 

 it associates in one natural series all the bilaterally symmetrical Mollusca 

 except the cephalopods. In doing this, it deals the death-blow to the supposed 

 Rhipidoglossau affinity of the Lamellibranchiata ; ''^ and in support of this con- 

 clusion I would point out that the recently discovered eyes of the mytilids are 

 in the position of those of the embryo Chiton,"'^ and that just as Dentalium, in 

 the formation of its mantle, passes through a lamellibranchiate stage, so are there 

 lamellibranchs in number in which a tubular investment is found.'^' 



This protrochal larva has an important part to play. It may very possibly 

 explain phenomena such as the compound nature of the trochal lobe of the 

 limpet,"- the presence of a post-oral ciliated band in the larva of the ship-worm,'^^ 

 and of a prse-anal one in that of various molluscan forms. '^* In view of it, we 

 must hesitate before we fully accept the belief in the ancestral significance of 

 the trochophore. And it is certain that an idea, at one time entertained, that the 

 Rotifer {Trochosphara) which so closely resembles it as to hear its name, is its 

 persistent representative,"'' is wrong, since this is now known to be but the female 

 of a species having a very ordinary male. 



Through the Rhipidoglossa we pass to the Gastropods, which are one and all 

 asymmetrical, for even Fissurella, Patella, and Doris, when young, develop a 

 spiral shell; while Huxley in 1877 had observed that the shell oiAjjli/sia, in its 

 asymmetry, betrays its spiral source. 



The notion, which until recently prevailed, that among these gastropods the 

 non-twisted or so-called euthyneurous condition of the visceral nerve-cords, as 

 exemplified by the Opisthobranchs, is a direct derivative of that of the Chitons 

 has been proved to be erroneous, since the nerves in Actaon and Chiliiia, like 

 those of the prosobranchs, are twisted or streptoneurous."'^ And as to the torsion 

 of the gastropod body, recent research, in which my lamented demonstrator the late 

 Mr. F. Woodward played a leading part, involving the discovery of paired reno- 

 pericardial apertures in Haliotis, Patella, and Trochus, has resulted in proof that 

 the dextral torsion which leads to the monotocardiac condition, does not affect all 

 organs lying primitively to the left of the rectum, as we have been taught. Con- 

 cerning the renal organs, it is the primitively (pre torsional) left one which remains 

 as the functional kidney, its ostium as the genital aperture. Nor is the primitively 

 right kidney necessarily lost, for while its ostium remains as the renal orifice, its 



