BULLETIN OF THE UNITED STATES FISH COMMISSION. 165 



May, but it often happens that the best hauls are made after the 15th 

 of August. 



It is impossible to construct a salmon-way over the Kolugil Falls. 

 A little below the falls a salmon-trap has been placed in one of the 

 branches of the river, but no fish have been caught in it. The salmon 

 follow the main stream, but according to the law no trap can be placed 

 in this. There are, however, only two persons who own these fisheries 

 up to the falls ; and as they agree perfectly, this limitation of the law 

 seems very unnecessary. 



At Melstadir, which I reached on August 24, and where I met a 

 number of people from the neighborhood, everybody seemed convinced 

 that the seals are the principal obstacle in the way of the salmon fish- 

 eries. 



In the river Vestra the salmon can go up as high as Ejukandi; in 

 the Xui)sa River they can go only as far as Efriuupr, and in the river 

 Austra they keep below the Kamb Falls. In 1S84 hardly any salmon 

 were caught in this river. On the whole, the natural conditions were 

 very much the same as in the Yididalsa. 



The Hrutafjardara three years ago yielded three hundred salmon a 

 year. In 1884 not a single salmon was caught, and in the yearprevious 

 only two. This river is not well adapted to the salmon, as they can as- 

 cend it only about 1 mile from the fiord, when the Ejettar Falls stop their 

 progress. The mouth of the river becomes more and more obstructed 

 by sand from year to year. There are probably not many good spawn- 

 ing places, and the river is easily exhausted. 



The ^ordra, which flows into the Hvita, has no salmon till a short 

 distance below Hvammer, and even hereonlj- a few are caught. It is said 

 that it is difficult for the salmon to pass the falls. I think, however, 

 that the salmon might find very good spawning i^laces in the greater 

 portion of the Xordra, at least up to the nearest falls, below Veidiltekia, 

 where at times the salmon fisheries in the deep places are very pro- 

 ductive. 



In the Hvita the salmon go onlj' as far as the Kldffos Falls. A large 

 number go into the Tvera and its tributary, the Kjara. The Tvera is a 

 much larger river than one is led to expect by looking at it on the map, 

 Salmon also go into all the tributaries of the Hvita and into the Bor- 

 garfiord. Among these tributaries containing salmon we may mention 

 the EeikiadalStT, Flokadalsa, Grimsa, Andakilsa, Glufia, Gnfa, and 

 Lauga; but the salmon in the last-mentioned stream are small. In the 

 Alfta, north of the Borgarfioid, there are very few salmon. Gudmundur 

 Palsson informed me that the Hita is very rich in salmon ; and that a 

 good many are also caught in the Kalda, Hafljardara, and Laxa. Some 

 of these streams are very small, and have hardly any water ; but it is 

 probable that they would form an excellent field for observations of the 

 salmon and the salmon fisheries. The salmon do not ascend the Hvit^ 



