272 J. T. CUNNINGHAM. 



of Elasmobranchs in the position of the original vessel of the 

 arch. This vessel in the latter forms lies near the outer bor- 

 der of the arch ; in Petromyzon it lies as near as possible to 

 the inner surface. The arch elongates and becomes flattened 

 antero-posteriorly ; the muscle-tube undergoes a correspond- 

 ing compression. The cartilaginous arch arises anterior to the 

 muscle-tube, but soon divides this tube in the middle of the 

 arch completely, separating an adductor on the inner side from 

 a constrictor on the outer, as in Selachians, The cells of the 

 anterior wall of the muscle-tube have a remarkable peculiarity. 

 They persist, in embryonic form, as long tubes, which run the 

 whole length of the arch, and show a transverse striation only 

 on the exterior. All the muscles run the whole length of the 

 arch and unite, dorsally as well as ventrally, with those of the 

 other side ; the important point about this is that if the carti- 

 laginous rods were to disappear the condition would be the 

 same as that which actually exists in Myxine. The chief dif- 

 ference between the gill laminse of the adult Petromyzon and 

 those of Selachians is that the former are directed towards the 

 exterior, the latter towards the interior, and this difference 

 appears at their first origin in the embryo. It is probable 

 that the adductors serve as inspiratory muscles by lifting up 

 the ventral side of the branchial region, and so expanding the 

 branchial cavities, while the constrictors are expiratory, their 

 contraction driving the water out. 



Thus it is shown that the branchial skeleton of Petromyzon 

 is composed of true cartilaginous branchial arches. It is true 

 that these arches in the Cyclostomata are not segmented, nor 

 are they in the Teleostean ; and this shows that Petromyzon is 

 derived from a form more premature than the Selachian, in 

 which the segmentation had not yet occurred. The same truth 

 is indicated by the homology of the hypophysis with the nasal 

 duct, an homology which, as Dohrn frankly acknowledges, was 

 first asserted by Goette in his 'Entwicklung der Unke.' 

 Petromyzon must have branched ofl" from a condition in which 

 the hypophysis was still an independent prseoral pair of gill- 

 clefts. Tliat the gills of Petromyzon are homologous with 



