32 Transactions. 



The arrangement is as follows : The carotid and systemic arches are 

 normal, but the pulmo-cutaneous arch bifurcates close to its origin, the 

 posterior branch of which has the usual relations dividing into the pul- 

 monary artery and the cutaneous artery. It is the anterior of the two 

 branches which is exceptional, and appears to be the persistent tliird 

 branchial arterial arch of the tadpole. This anterior branch arises from 

 the base of the pulmo-cutaneous soon after it leaves the synangium : it is 

 equal in diameter to this arch, and runs parallel with the other arches 

 nearly to the point at w^iich the last arch divides to form the pulmonarv 

 artery and the cutaneous artery. But as it approaches this cutaneous 

 artery its diameter decreases and it bends backwards towards the fourth 

 arch, to which it is joined by a very slender vessel. It then bends forward 

 again and is continued into the cutaneous artery, alongside which runs as 

 usual the petrohyoid muscle, to which it gives off twigs. The relation of this 

 third arch to the cutaneous artery would suggest that the latter is derived 

 from it, were it not for the precise account of the development of the latter 

 given by Marshall. At the first bend of the arch is an angle as if a vessel 

 or ligament passed forwards to the systemic arch, but I can find no trace 

 of this. There is no connection between this third arch and either the 

 systemic arch or the dorsal aorta. 



On the left side I find that the condition of affairs is essentially the 

 same, but the third arch is much more slender than on the right side. 

 Less injection has penetrated the vessel, which suggests that possibly some 

 resistance is exerted at the connection between it and the cutaneous artery. 

 Nevertheless, the connecting vessel is distinctly red with injection, but is 

 much narrower than its basal region. As on the right side, this third arch 

 bends backwards (more abruptly than is shown in the drawing, for it is 

 better seen when the arches are stretched apart) in order to reach the 

 cutaneous artery, which on this side is normal and of equal diameter 

 throughout its course. 



The delicate pharyngeal artery, from the systemic arch, is plainly visible 

 in the specimen below the third arch, but I have omitted it from the 

 drawing for the sake of clearness. 



According to Marshall,* during metamorphosis " the third aortic arch 

 in the third branchial arch of the tadpole atrophies altogether. In young 

 frogs of the first year it loses its connection with the aorta and then gradu- 

 ally shortens up, the distal end becoming a solid cord, and the proximal 

 or cardiac part retaining for a time its lumen. Before the end of the first 

 year this vessel has entirely disappeared." 



In the larva this third afferent arch goes, of course, to a gill, and has 

 no connection distally with the fourth arch : it is this union on the ventral 

 surface that is rather puzzling in the present case, and especially the very 

 slender union between the cutaneous artery and the parent fourth arch. 

 It raises the question whether the ontogeny of the frog is really a true 

 recapitulation of the phylogeny of the Anura, or whether the cutaneous 

 artery is originally derived from the third arch, which in the embryology of 

 those species of frog that have been studied has imdergone some modifica- 

 tion, leaving the third to become connected with the fourth arch at some 

 stage in the history. 



The cutaneous artery seems to be peculiar to the Anura, as no reference 

 is made to such an artery in any description of the anatomy of salamander 



* A. M. M.\ESHALL, Vertebrate Embryology, p. 178, 1893. 



