128 The Ottawa Naturalist. [Oct. 



specimen 8 inches long in Messier Straits, but Dr. Gilchrist's 

 specimen 2 feet long obtained in September, 1903, about 40 

 miles north-east of Cape Point, proved to be a mature female 

 specimen in which the ovaries were very advanced and crowded 

 with reddish spherical eggs, numbering probably not less than 

 30,000.* 



The eggs were formed in the hanging transverse folds of 

 the inner ovarian surface, and later they collected on the floor 

 of the chamber of the ovary. They flowed freely from the fish, 

 and Dr. Gilchrist was led to regard them, at first, as ordinary 

 demersal eggs, deposited by the fish on the bottom of the sea. 

 To his surprise he found, on closer examination, very young fish 

 hatching out within the parent. Eight small larval fish were 

 curled up among the loose ova. In the mouth of one larva he 

 found some oil-globules, and in another a mass of soft food- 

 matter, in which were oil-globules and spots of black colour. 

 The mass was carefully removed and turned out to be part of a 

 young fish which was being devoured by another baby fish, and 

 the rest of the body of the victim was found close to its devourer. 

 Alcock had already made the important announcement that in 

 Saccogaster, a deep-sea species, developing embryo fish were 

 found inside the parent and hinted that they fed on the sur- 

 rounding ova; but Dr. Gilchrist's discovery proved that some 

 embryo fish actually swallowed and fed upon other embryos of 

 the same brood, and thus lived and grew inside the ovarian 

 chamber. The larger larvae 10 mm. (f of an inch) long, lived on the 

 smaller newly-hatched young, not simply upon the surrounding 

 eggs. These'larval cannibals showed well-developed breast fins, 

 and anal and pre-anal fin-lobes, but the tail had not any caudal 

 fin-lobes. 



Most fish, of course, produce eggs or spawn, and the young 

 develop and hatch after they have been laid by the parent. 

 The formation of the young inside the deposited egg of a fish, 

 may take from 2 days to 6 or 8 months in different species, the 

 shad being an example of rapid development (a few days) , while 

 the salmon or trout take a long period of time (many months). 

 But in the parent forms of many viviparous fish the young may 

 be found not only already hatched out and lively, but may be 

 very advanced, and exhibit the almost mature form and appear- 

 ance. I have frequently examined specimens of viviparous 

 species both on the Atlantic and Pacific coasts, and can confirm 

 Dr.Gunther's description that the young, in such fish as Zoarces, 

 on the Atlantic, and Cymatogaster, on the Pacific, coasts, 



*Dr. Gilchrist had in August, 1903, secured a fine specimen 2 feet long. 



