479 



runs forward along the ventral edge of the related cleft and approaches 

 but does not jom the next anterior artery. No connection of any kind 

 could be found between the dorsal ends of any of the afferent arteries, 

 but each of the seven arteries turns backward, slightly, around the 

 dorsal end of the related cleft. Heptanchus, accordingly, presents a 

 condition in which the dorsal and ventral afferent loops of Chlamydo- 

 selachus are probably aborting, for it seems improbable that they 

 are here found in process of development. 



Both anterior and posterior efferent arteries are found in each 

 of the six branchial arches, while in the hyoidean arch there is but 

 a single efferent artery, and it has the position, in its arch, of a pos- 

 terior artery. The two arteries in each of the branchial arches are 

 connected by several intermediate cross-commissural vessels, these 

 commissures all lying internal to the afferent artery of the related arch, 

 and not external to that artery as shown inBouLENGER's(1904) figure 

 of Mustelus antarcticus. 



The posterior efferent artery in each arch is connected by dorsal 

 commissure with the anterior efferent artery in the next posterior 

 arch, but it is not so connected with the anterior efferent artery of 

 its own arch, Heptanchus, in this, resembling the adult Mustelus 

 (Parker, 1886) and differing from the adult Chlamydoselachus(ALLis, 

 1911). The term commissure, here employed, is the one used by 

 DOHRN in his discussion of these vessels in selachian embryos, but 

 DoHRN says (1890, p. 399) that in these embryos the commissure that 

 connects the posterior efferent hyoidean and the anterior efferent 

 glossopharyngeus (first branchial) arteries is not a structure strictly 

 comparable to the dorsal commissures that primarily connect the 

 anterior and posterior efferent arteries in each arch; the former 

 commissure arising much later than the latter ones, and apparently 

 being simply the result of a gradual approach and final contact and 

 fusion of the two vessels concerned around the dorsal end of the 

 intervening cleft. The development of similar commissures around 

 the dorsal ends of the more posterior gill clefts Dohrn does not de- 

 scribe, but, as they are there found in the adult, it is evident that 

 Dohrn's description apphes to all the clefts and that the two sets 

 of commissures thus developed give rise, in principle, to the condition 

 found in the adult Chlamydoselachus, where there is a continuous 

 dorsal longitudinal commissure. But, in the adult of selachians other 

 than Chlamydoselachus, so far as yet described, this continuous dorsal 



