10 REPORT OF COMMISSIONER OF FISH AND FISHERIES. 



before they ascend the Yukon, where the running season is very short 

 and may not exceed a month or six weeks. The early runs in the 

 Cohimbia River are usually from one to three weeks passing from the 

 mouth of the river to Clifton, about 20 miles. They first appear at 

 The Dalles, 200 miles u]) the river, in the middle of Ai^ril, and are found 

 in great quantities at this point about the middle of June, two months 

 after they appeared in large numbers at the bar. This would indicate 

 that they j)roceed up the Columbia at the rate of 100 miles a month. 

 In the later runs they probably travel faster. 



The spawning season of the (luinnat salmon varies in different rivers, 

 and, considering the entire coast, extends over a i^eriod of fully six 

 months. In July the summer run is spawning at the headwaters of 

 the McCloud and Sacramento; in August and September, farther down 

 these rivers. In October the fall run has began at the McCloud 

 and below, and this run continues spawning through November into 

 December. In the Columbia the spawning season begins at the head- 

 waters in June; at Clackamas, 125 miles from the mouth of the river, 

 it begins about the middle of September and continues until November. 



A few days before they are ready to spawn the salmon hollow out 

 elongated cavities with their heads and tails in the gravel beds of the 

 river where there is some current, and here in due time the eggs and 

 milt are deposited. The eggs drift into the crevices in the pile of 

 stones thrown up below the hollow, sink to the bottom, and remain in 

 that i)rotected position during incubation ; here, also, the young remain 

 until the umbilical sac is absorbed. The eggs and young are liable to 

 destruction by freshets, but are comparatively safe from other injurious 

 influences. 



The quiunat salmon is not so prolitic as the Atlantic salmon, 300 or 

 400 eggs to each pound weight of the parent fish being about a fair 

 average. 



In view of the enormous annual destruction of this salmon for com- 

 mercial purposes the necessity for its propagation became manifest at 

 an early period in the history of the Pacific fisheries. Fortunately the 

 species is readily susceptible of artificial propagation on a large scale. 

 In 1873 the application of fish-cultural methods to this species began 

 on the McCloud River, California. The propagation work, since grown 

 to large proportions, now engages the attention of all the coast States 

 as well as the general government, and in 1896 was more extensive 

 than ever before. Whenever large numbers of the fry, artificially 

 hatched, have been placed in the waters of the Pacific coast, whether 

 in the tributaries of the Sacramento or the Columbia, an increase in the 

 run of full-grown fish has been observed after the time required for a 

 salmon to mature (about four years) has elapsed. 



As the salmon ascend the rivers they are caught by gill nets, pounds, 

 weirs, fyke nets, seines, wheels, and other devices, but the great bulk 

 of those taken in the Columbia and Sacramento are caught with gill 

 nets drifting with the current or tide as they head up stream. In the 



