466 PARENTAL CARE AMONG FRESH-WATER EISHES.- 



Htlle if any weiglit, Imt they move slightly ami appear to serve to steady the 

 hody. 



On one occasion an individual displayed much activity and gave 

 Mawson a good opportunity to observe its peculiar movements. 



Last night I heard a peculiar sound, and on looking around I saw one of the 

 fish travelling about the room. He had escaped from the tin which was in my 

 bedroom, had fallen from the table to the floor, and travelled along the corridor, 

 about 12 meters (about 40 feet) to the sala. I watched him travelling for 

 two hours, during which time I estimate that he moved at least 90 meters. 

 Toward the end of the two hours he seemed to flag a little, but in the earlier 

 part his method and speed were fairly seen. He seemed to start with a sudden 

 movement of the head or the barbels, then wriggled briskly for .5 or 10 

 seconds, advancing about a meter. Then he would rest for about 10 seconds, 

 and start as before. This was kept up during the whole two hours, and I 

 left him still moving. This morning, five hours later, I found him dead. While 

 he was moving I spilled some water on the floor, but he crossed it ; hence I 

 concluded that it was nmd rather than water of which he was in search. The 

 fish <ire eaten and considered good food. 



Many of the hassars bury themselves in the mud as the stream in 

 which they have lived dries up and are able to survive quite a pro- 

 tracted dry period, reviving when the rains again fill the depressions 

 with water. 



The roundheaded hassar of Demerara {Iloplostefmnm littoralc) is 

 one of the earliest fishes whose nest-building habits have been de- 

 scribed. When it was generally asserted by naturalists that '' no 

 fishes are known to take any care of their offspring," from far-off 

 Demerara John Hancock, an English surgeon resident in that colony 

 during the second quarter of the last century, in 1829 sent " Notes on 

 some species of fishes and reptiles," and among them were observa- 

 tions on fresh-water fishes which he called hassars — one kind, the 

 flathead, a Donts, and another, the roundhead, the Hoplosternuni 

 littorale. The latter was claimed to assiduously guard its eggs and 

 young. The sexes, it was asserted, cooperate and make a regular nest, 

 formed of grass, in which " they lay their eggs in a flattened cluster, 

 and cover them over most carefully." 



According to Hancock, too, '' they lay their eggs only in wet 

 weather." He had been " surprised to observe the sudden appearance 

 of numerous nests in a morning after rain " had occurred, "■ the spot 

 being indicated by a bunch of froth, Avhich appears on the surface of 

 the water, over the nest ; below this are the eggs, placed on a bunch of 

 fallen leaves or grass," which the fishes " cut and collect together. 

 By what means this is effected," he continued, " seems rather mys- 

 terious, as the species are destitute of cutting teeth. It may possibly 

 be by the use of their serrated arms, wliich form the first ray of the 

 pectoral fins." 



The nest building was later (in 1886) described by Captain Vipan, 

 who observed the same species in Trinidad. He had a pair confined 



