26 BULLETIN OF THE BUREAU OF FISHEBIES 



Cascade Creek: This was relatively more densely populated than most of the other streams. 

 The stream was explored for about three-quarters of a mile. Above the first quarter mile the 

 stream rises rapidly, flowing through a precipitously walled canyon. Spawning fish were seen as 

 far as we went, but there the bed of the stream became still steeper and it is doubtful that many 

 fish were above. It was estimated that there were 2,000 to 2,500 fish in the stream. 



Meadow Creek: This stream was explored for about 500 yards. Only about 500 live fish 

 were seen and about ten times as many dead ones. 



This completed the observations on the lake spawning grounds for 1926. It is 

 apparent from the notes that the spawning during the latter part of August was 

 remarkably light, especially in view of the fact that there was a heavy escapement 

 through the weir during the month. The following note made on August 24 bears 

 on this: 



Many salmon have been seen during the past few days, jumping and "porpoising" in the lake 

 tself — alongshore and out over the deeper water. It seems probable that these are waiting in 

 the lake to ripen before seeking the mouths of the streams. Certainly we have seen no such numbers 

 in the streams as have been counted through the weirs during the past month. 



However, during the trip down to the portage, on August 27, great numbers of 

 salmon were also seen in the river below the lake. Evidently, many of the fish 

 that entered the river during August had remained in the river. The river was 

 examined on August 10 and was found to contain only a few fish, but on the 27th 

 "we saw fish everywhere in the river. * * * In the first 2 miles below the lake 

 they were very numerous, and many of them were spawning. However, this was 

 by no means true of all the fish in the river, as we noted fish entering the lake at the 

 outlet, and there were many quite obviously passing upstream below the region 

 where the fish were spawning. Under these circumstances it was difficult to esti- 

 mate the number of fish actually spawning in the upper 2 miles of the river, but I 

 believe there were between 50,000 and 100,000. In addition, there were as many 

 more, approximately, in the river passing up, either to spawn in the upper reaches 

 of the river or to go on into the lake." 



The summer of 1926 was exceptionally dry throughout most of Alaska. This 

 was not only a matter of common observation but is shown in the records of the 

 Weather Bureau. Throughout the Karluk System the streams were very low, 

 probably lower than they had been for many years. The lake, too, was notice- 

 ably lower, the level late in August being about 1 foot below what appeared to 

 be the normal level, judging by the grass and other vegetation alongshore. Just 

 what effect this will have on the success of the spawning is problematical; it may 

 be slight or it may be great enough to offset in considerable measure the effect of 

 the fine spawning escapement. On August 9 the following notes were made: 



In Thumb River, where the spawning had been heaviest, many of the nests were exposed by 

 the lowering of the water. We dug in some of them and found mainly dead eggs, although a very 

 few live ones were found. 



Similar conditions were noted in all of the larger streams— O'Malley River, 

 Falls Creek, and Canyon Creek. All of these flow for a considerable portion of their 

 lower courses through flats, where the stream beds are relatively wide. All of the 

 other streams, except those small ones of the character of Spring Creek, drop down 

 much more abruptly from the mountains, and the beds are more closely confined. 

 These, therefore, were not as much affected by the dry season. 



