Underground Conduit 



Figure 5. — Diagrammatic sketcli of trapping system on bypass conduit on the Yakima River at Prosser. When 

 valve (A) is open, fish pass directly to the Yakima River. To activate the trap, the valve is closed forcing the 

 water and fish upward through a hole in the top of the conduit (B). Water flows over the top of the interior 

 wall (C), through inclined plane screen (D), over a second interior wall (E), and on to the Yakima River. 

 Fish and some water continue into trap (F). 



fishing equipment in the winter and spring of 

 1958-63 but found no first-year fish in the mid- 

 river stretches. Furthermore, no first-year fish 

 migrated seaward past Roza Dam, 19 km. above 

 the confluence of the Yakima and N aches Rivers, 

 in either 1961 or 1962 (according to Donald 

 Greenland).'' Data presented here pertain only 

 to spring chinook salmon, or fish that migrate 

 seaward in theh second year. 



Length-frequency polygons of downstream-mi- 

 grating spring chinook salmon (fig. 6) show that 

 the modal length increased from 120 mm. in 

 1959 and 1960 to 125 mm. in 1961, 130 mm. in 

 1962, and 135 mm. in 1963. Present data are 

 inadequate to explain the reason for the increase 

 in modal length. Possible explanations include 

 the availability of more food per young fish and 

 differential growth and abundance of the several 

 tributary subpopulations beheved to be present. 



Analysis of the length and weight data (table 3 

 and figure 7) clearly shows that the larger fish 

 tend to migrate early in the season. The condition 

 factor tends to increase — that is, the fish become 

 hea\aer per unit length as the season progresses. 

 This change is e.xplained, at least in part, by the 

 extended sojourn of the later migrating fish in 

 water where the temperature is rising and food is 



' Donald Greenland, Fishery Biologist, Bur. Cormner. Fish., Columbia 

 Rivtr Program Office, Portland, Oreg. Personal communication. 



becoming more abundant, and where their feeding 

 becomes more intensive. 



TIMING OF SEAWARD MIGRATION 



The daily timing of the downstream migrations 

 in 1959-63 was similar for all years (fig. 8). Gen- 

 erally the catches remained relatively stable be- 

 tween periods ending at noon, 4 p.m., and 8 p.m. 

 and increased in the period ending at midnight. 

 The hourly distribution of the catch between 

 midnight and 8 a.m. is unknown, but the mmiber 

 of chinook salmon captured during the 8-hour 

 period approximated the number taken from 8 

 p.m. to midnight. Thus, migration tended to be 

 at night. 



The week or weeks of heaviest seaward migra- 

 tion of spring chinook salmon at Prosser varied in 

 1959-63, but always fell between April 14 and 

 May 19 (fig. 9). 



SURVIVAL TO SEAWARD MIGRATION 



We estimated the number of seaward migrants 

 from the formula: 



iv=i:iv<=z;ci(32.,5//,) 

 i=i 1=1 



where N is the estimated number of spring chinook 

 salmon migrating seaward, N, is the estimated 



EGG-TO-MIGRANT SURVIVAL OF SPRING CHINOOK SALMON IN YAKIMA RIVER, WASH. 



353 



