6 



racteristic of a respiratory raucous membrane, that a practised eye 

 would immediately discern their true office. 



In the cartilaginous fishes, the arrangement of the gills presents 

 many varieties ; those only to be described here occur in the rays 

 and lampreys ; in the former, the branchial arches are composed of 

 soft cartilage, and do not hang loose in the branchial chamber as in 

 most osseous fishes, but are attached by both extremities, so that the 

 chamber is divided into five compartments distinct from each other, 

 each having an orifice on the external surface of the body. The 

 branchial lamellae are given off from the arches in the form of folds, 

 parallel to each other and continuous with the mucous membrane ; 

 in the first four cavities there are two gills, one on the anterior and 

 the other on the posterior surface of the arch, but in the last cavity 

 there is only one gill, and that on the anterior part of the arch. 



In the lampreys there are no branchial arches, but seven branchial 

 cavities or pouches, perfectly distinct from each other, each of which 

 has two openings, one communicating with the pharynx, and the 

 other situated on the external surface of the body ; the interior of 

 each chamber is lined with the branchial mucous membrane, which 

 is plicated transversely, in the same way as it is on the laminae of the 

 osseous fishes, and each fold has its characteristic net-work of capil- 

 laries. In the river lamprey {Petromyzon fluviatilis), the capillaries 

 are quite as small, if not smaller, than in an eel of the same size, but 

 it is in P. marinus, and more especially in the rays, that the most 

 largely developed net-work is found ; in these latter, the blood cor- 

 puscles are also of larger size than in other fishes. Portions of one 

 of the laminae from P. marinus are represented as seen under a power 

 of twenty diameters, at fig. 4, and under that of fifty at fig. 4 a, and 

 similar specimens from the ray, in figs. 5 and 5 a ; these, although 

 only magnified to one -half of the extent of those of the eel, exhibit 

 the capillaries, nevertheless, much more plainly. 



Of all the fish in which I have been able to inject the capillary plexus 

 satisfactorily, the eel exhibits the smallest vessels ; in an eel about 

 twelve inches in length, the laminae were on an average not much 

 more than one-quarter of an inch long, and on each lamina I was 

 able to count upwards of eighty plicae, each having a plexus of capilla- 

 ries so minute, that a good defining power of fifty diameters was re- 

 quired to make them out, and one of a hundred diameters to show 

 them satisfactorily, the diameter of a single capillary being the 

 four-thousandth of an inch, so that it becomes a question whether 

 vessels so small as these were ever seen by the early microscopists. 



