G04 REPORT— 1901. 



hypotheses — that of Gegenhaur and that of Balfour, Thacher, and Mivart. Atten- 

 tion was drawn to the complete absence of intermediate stages between gill septum 

 and limb, and also to the a priori improhability of a gill septum such as we know 

 in the lower fishes, firmly fixed and ilush with the surface, developing into a 

 motor organ. It was pointed out, however, that the numerous advocates of the 

 Gegenhaur view had managed to accumulate a large mass of evidence bearing 

 upon one particular phase of the question, and which consisted of facts pointing 

 to an extensive backward migration of the paired limbs having taken place from 

 somewhere in the neighbourhood of the branchial region. 



The lateral fold view had at first the advantage of resting upon a more 

 certain foundation of anatomical fact — upon the fact discovered by Balfour that 

 in the young torpedo the two limbs are for a time connected by a continuous 

 ridge of epiblast — that in this form the paired limbs develop in precisely the way 

 in which the theorv supposes them to have developed during phylogeny. Modern 

 research has, however, shown that this longitudinal ridge of epiblast does not 

 appear at all in the less specialised Selachians ; even in Torpedo the ridge 

 appears secondarily, and its appearance at all is probably a quite secondary 

 phenomenon associated with the secondary extension of the paired fins along the 

 sides of the body in the adult. Embryology as it is known to-day does not 

 furnish the same foundation ibr such a theory of limb formation as it appeared to 

 do at an earlier period. 



The anatomical resemblances between paired and impaired fins were touched 

 upon, and it was suggested that such resemblances are probably due to homoplasj-. 

 Attention was now drawn to the fact that in the relations to one another of 

 muscles, skeleton, and viscera in the lower vertebrates there was expressed an 

 admirable mechanical arrangement ibr lateral flexure of the body. Properly 

 co-ordinated lateral flexures provided a powerful means of locomotion through 

 fluid, a method used by all the lower vertebrates. It was difiicult to believe that 

 either a gill septum or a lateral fold could aid to any appreciable extent this 

 primitive method of swimming ; the probability was that in its incipient stages a 

 limb derived in such a way must act rather as a hindrance. 



The author was of the view that the paired limbs were not at first swimming 

 organs at all, but that they were developed in correlation with movement about 

 a solid stratum. AVitli a solid point (lappid even a very small movable projection 

 would be of use in propelling the creature forward. The question was, Did such 

 projecrions of the body wall exist in th(! lower vertebrates which might have by 

 evolution become developed into paired limbs H He considered that the most 

 primitive groups of Gnathostomata were the Selachians, the Crossopterygians, the 

 Dipnoans, and the Urodele amphibians. In three out of the four groups there 

 occurred during development true external gills, projections of mesohlast covered 

 with epiblast sticking out from the visceral arches (Mandibular — 'Balancer' 

 of Ui'odeles ; Hyoid — Crossopterygians ; Branchial Arches I.-III. — Urodeles, 

 Lepidosiren, Protopterus ; Branchial Arch IV. — Lepiclosiren, Frotopterus). In 

 the Selachians their absence was correlated with the presence of the enormous 

 highly vascular yolk-sac, which made the persistence of any other dermal 

 respiratory organ of early life quite unnecessary. The true external gills were 

 supposed by some to be larval organs independently developed, but further 

 knowleage of their identical relations and development made it impossible to 

 accept any other view than that they were truly homologous structures inherited 

 from a remote ancestor. 



The structures in question are provided with elaborate muscTdar arrange- 

 ments; in a live Dipnoan or Urodelan larva they are seen to he every now and 

 then sharply flicked back ; they are, in fact, though mainly respiratory, potentially 

 motor in function. In Urodeles the corresponding structure on the mandibular 

 arch has lost its respiratory and taken on a purely supporting function. 



The author concluded that in these serially aiTanged potentially motor organs 

 of the lower vertebrates were to be recognised organs homodynamous with the 

 structures which had given rise to the paired limbs ; the limb-girdles he followed 

 Gegenhaur in regarding as modified visceral arches. The earliest stage of the 



