ANGLER FISHES GILL. 607 



of the laboratory of the United States Bureau of Fisheries at Beau- 

 fort had a couple which were observed by Gudger to feed " vora- 

 ciously, eating pieces of oyster, bits of shrimp, and small fishes, alive 

 or dead." Others thrived in aquaria at Woods Hole, and were 

 noticed by Hugh Smith. The several individuals 



were clumsy in their movements, and often made prodigious efforts to go short 

 distances. At tlie same time, they approached their prey stealthily, and liter- 

 ally pounced upon it. Their wide mouth enables them to capture and swallow 

 fishes that are entirely disproportionate to their own size. 



This was well illustrated b}' the habit of eating their fellows, of 

 which Smith observed several cases. 



On one occasion a specimen 6 inches long captured and swallowed intact 

 another nearly 4 inches long, and did not seem particularly incommoded 

 thereby. They persistently bit off the dermal flaps of their fellows, so that 

 after a few days nearly all of them were more or less completely stripped of 

 these appendages. This habit was exhibited when there was an abundance 

 of minnows on which to feed. Cannibalism in this species has been noted by 

 other observers. 



The spawning time of the sargasso fish may extend over a consider- 

 able period. Its newly matured eggs have been observed from mid 

 and late summer (July and August) to late in the fall (October). 

 Evidence of the maturation of the eggs becomes manifest by swelling 

 of the abdomen and, according to Gudger, sometimes " in front of the 

 arms, becoming as square as if it had been cut to shape with a knife." 

 Soon after this condition has been attained the eggs are discharged 

 in a jellylike mass, which becomes swollen on contact with the water 

 and enlarged into a narrow raft, 3 or 4 feet long, although the mother 

 fish may have been " only 3 or 3^ inches long," and " had only about 

 one-third of the volume of the eggs and jelly combined." 



The act of spawning had not been described till Dr. Hugh Smith 

 observed it at Woods Hole, There, in 1897, " several spawned in the 

 aquarium in August, but the eggs were not fertilized. The eggs were 

 buoyant and combined in long bands or strings like those of the goose 

 fish {Lophiiis) y At the Beaufort laboratory, in 1894, a sargasso 

 fish, which had been seven weeks in captivity, laid a long string of 

 eggs on July 25. In every subsequent year oviposition was repeatedly 

 observed. 



The prolonged time during which spaw^nmg may occur appears to 

 be partly due to the difference in the development and maturation of 

 the two ovaries. In the case of a female, especially observed in 1906^ 

 Doctor Smith found that there was one issue of eggs in a raftlike 

 mass on the 6th of September and a second on the 10th of October, 

 and in 1905 another female matured one raft of eggs in August and 

 later one in September. Nevertheless, occasionally, according to an 



