36 MIGRATION OF ADULT SOCKEYE SALMON. 
It seems not unlikely that the difference in rates of travel shown in 
the different quartiles may be explained as indicating racial differences 
existing in the fish composing different parts of the run. Gilbert @ 
has demonstrated that the Fraser River run is composed of a number 
of distinguishable races, each bound to a different spawning region. 
The assumption that these races may have different rates of migration 
while passing through the waters of the Sound does not seem to be a 
difficult one to accept. . 
In the case of the halves, the results from all three stations are in 
complete agreement, indicating that the rate during the second half 
is distinctly more rapid than during the first half. 
In connection with the above determination of the rate of progress 
through Puget Sound, it must be borne in mind that the rates are 
undoubtedly lower than they should be. ‘This necessarily follows 
when it is considered that seldom Jess than one day, and frequently 
three or four days, elapses between the times when the traps are 
lifted and the fish reported. Again, the fish may spend some time 
in the hearts and pot of the trap before entering the spiller, from 
which they may be taken at the next lift. A fish entering the trap 
immediately after it has been lifted would not be reported ‘before the 
next day at the earliest, thus adding at least one day to the actual 
time required to make the journey from the marking station. If 
traps were uniformly lifted at 24-hour intervals, 12 hours could with 
safety be subtracted from the average number of days en route and 
the rate calculated accordingly. The authors’ knowledge on this 
point, however, is not sufficient to warrant such treatment. It 
would undoubtedly be a step in the right direction, but would tend 
to give an opinion that the rates have been fully corrected, which 
would be quite unwarranted by the facts. It is possible that the 
ae involved is within the limits of error dependent upon other 
actors. 
RETURNS FROM TRIBUTARY STREAMS AND THE FRASER RIVER ABOVE 
MISSION BRIDGE. 
Returns were obtained from various sections of the Fraser River 
watershed which have not been included in the regions indicated on 
the map. ‘These sections may be indicated as follows: 
: Specimens, 
Main river, 3 to 27 miles above Mission............---.0--------+ 2-0-2222 2- ses 7 
Main river, near Hope, Yale, Hell’s Gate, and Spuzzum.........-------------- 16 
Main river at Lytton and Lillooet........0522.. 2000-2 20-2---- 0-22 e ese eee ee eee 2 
Main fiver at.Soda Creela.:.3¢ 5. teen toes en ee sored > ee ee eee 2 
Pate Biverr 2. -c.occcs ds oc bat acs agiReReRROeate Bee eeetes a been eee ee eee 6 
Harrison Lake system, between Harrison and Lillooet Lakes. ...-..-.---------- 5 
Harrison Lake system, Birkenhead River.........-.----------------+-------+-- 9 
Seton Lake system, Portage Creek...-.......-0--s222-- ese eee cece eee eee eee 2 
Chilcotin, Biwer .c¢ wend dos econ see ER RO ern as Bee ee Se 1 
The complete data concerning these returns are given in Table 29. 
The mean dates on which the fish captured in each of the above dis- 
tricts were marked have been calculated and appear in the summary. 
This was done in order to see, if possible, when the fish bound to the 
different tributaries were to be found in Puget Sound. The possi- 
bility of a segregation in time of the races demonstrated by Dr. Gil- 
bert has been suggested above. 

a Gilbert, Charles H.: Contributions to the life history of the sockeye salmon, Report, commissioner 
of fisheries for the Province of British Columbia for 1917, paper No. 4, pp. Q33-80. Victoria, 1918, 
