OF THE EEL AND LAMPREY. 219 



close together as they can well be placed ; and they may be seen ex- 

 ternally through the membrane composing the flake. When these fishes 

 have spawned the flakes become flaccid, but still the nidi may be seen 

 in little opaque spots. The mode of spawning I shall describe in the 

 common eel. 



In the common eel, and also in the conger, the roe is somewhat 

 similar to the above, although not exactly. Each roe is composed of a 

 membrane attached by one edge to the back of the fish, almost through 

 the whole length of the abdomen, and continued into the tail some way 

 beyond the anus : the other edge is unattached, and is longer than the 

 attached one, so that it hangs like a ruffle 1 . On the sides of this mem- 

 brane are a number of folds, similar to the inside of roes in common ; it 

 is similar to half of a common roe slit up through its whole length, 

 having the smooth membrane on one side and the flakes or folds on the 

 other. 



These roes in the lampern, lamprey, conger and common eel, have no 

 duct or outlet directly belonging to them ; therefore the operation of 

 spawning is uncommon, and probably peculiar to this order of fish. The 

 passage out appears to be by two openings, directly from the cavity of 

 the belly just behind the rectum, which unites into one, and opens into 

 the rectum on the further side of that gut just at the verge of the anus*. 

 From this formation of parts, the question is, how do they spawn ? In 

 the common fishes the parts themselves explain this operation, and in 

 the present we must have recourse to the same method. 



In the common fish we must suppose that the ova fall off and get 

 loose into the cavity of the roe or ovarium, and then are protruded out 

 of that cavity through the duct, by the action probably of both the roe 

 and of the abdominal muscles, which forces them externally. 



In the eel, &c, we must suppose them [the ova, to be] forced out at 

 the small opening above mentioned by the same kind of action. 



From the structure of the parts, this method of accounting for the 

 operation of spawning appears to be the only possible one ; and 

 although it may be difficult to conceive how the spawn, when loose in 



* All of the ray-kind have two openings from the belly, one on each side, by the 

 fin at the anus 2 . 



1 [This mode of attachment of the ova to fringes, as compared with the compact 

 masses of the ovaria in other fishes, is admirably adapted to the vermicular winding 

 motions of the eel.] 



2 [In the Hunterian preparation of the salmon, No. 2662, the peritoneal apertures 

 are shown, within the verge of the vent, between the cloacal apertures of the bladder 

 and rectum; as described in my Physiological Catalogue, 4to. vol. iv. p. 130. The 

 proper oviducts are wanting in the salmon-tribe, and the ova are excluded by the 

 peritoneal outlets, as in the eel-tribe.] 



» 



