Young salmon were observed from a barge anchored in the middle 

 of the river during a heavy migration on Jxme 24, 1946, between 8 and 

 9 p.m. At that time, fry in great nianbers could be seen drifting down- 

 stream with the current. They seemed to be evenly dispersed o\-er the 

 entire river with no tendency to travel in groups or schools, euid most 

 of them were within a few inches of the surface. Few were seen at depths 

 exceeding 18 inches, but all of them make some effort to stay head upstream 

 as they drifted with the otirront. In a 5-ininute period, 47 were counted as 

 they passed through an illuminated section of river 10 feet wide* Assum- 

 ing that the migrants were evenly distributed across the 200-foot width 

 of the river, they were moving past this point at the rate of 188 per 

 minute, or 11,280 per hour at that time. 



In the warm summer months when relatively no downstream movement 

 is evidenced by fyke-net returns, young salmon are commonly seen at 

 all points along the river in the vicinity of Lewis ton* Their presence 

 is most conspicuous in the early morning when they can be observed Jump- 

 ing from the water in quiet, shallow areas along the river^s edge. 

 Throughout the day they apparently seek seclusion in the rooks on the 

 river bottom. Several hundred young salmon congregated at a small sub- 

 surface spring near Lewis ton on days when the water temperature rose above 

 75°F« Any tendency toward downstream movement in their activities is so 

 slight that it escapes detection. The seaward migration is resumed 

 immediately following the first fall rains when water temperatures approxi- 

 mate those recorded in April, May, and June* This migration of yoting 

 entenng their second year of life usually starts in October or November 

 and continues until March and April of the following year, as shown in 

 Figure 6. 



The rate of growth of seaward migrant salmon is difficult to 

 ascertain from samples taken at a fixed location. It would be possible 

 only if certain population segments could be followed downstream and 

 sampled periodically. Sampling at a fixed location actually provides a 

 single set of data on a different population group each day. It is 

 interesting to note, however, that eis the season of downstream migra- 

 tion progresses, the average length of migrants taken increases uniformly 

 and may be indicative of the rate of growth (Table 11). Although samples 

 taken in February are too small for reliable statistical treataaent, salmon 

 of the year captured during that month in three of the four seasons were 

 larger than migrants caught in March of the same years. The average 

 length of February migrants in 1944 and 1945 was greater than similar 

 average lengths obtained in both March and April. This tendency is pre- 

 sumably indicative of the difference in time of emergence between progeny 

 of spring or summer-run adults and progeny of fall-run parent salmon. 



The difference in average lengths of seaward migrants was i 15 mn. 

 (Goo9 inches) between March and May of 1943; 15 mm. (0.69 Inches) between 

 March and June of 1944; 15 mm. (0.59 inches) between March and June of 

 1945; and 16 mm. (0.63 inches) between March and July of 1946. Salmon 



33 



