202 
PACIFIC SCIENCE, Vol. XX, April 1966 
small mean differences for all three species; the 
mean differences for sockeye and pink salmon 
were approximately 0.02 and for chums 0,04 
Although the left pectoral fin counts generally 
exceeded the right, the tests for differences were 
not statistically significant for any sample ' or 
species. 
CONCLUSIONS 
The results of this study show that sockeye, 
chum, and pink salmon, which are essentially 
bilaterally symmetrical animals, exhibit some 
asymmetries with respect to all four paired 
meristic characters considered. From the speci- 
mens examined no conclusive relationship was 
shown between the occurrences of asymmetries 
and the sex or the area from which samples 
were obtained; however, these possible sources 
of variation were not rigidly tested, for reasons 
given earlier in the text. 
In taxonomic and racial studies involving 
lateral line scale counts, interchanging counts 
from left and right sides of chum salmon intro- 
duces an additional source of variation and, in a 
situation of considerable numbers of substitu- 
tions, may be a serious source of bias. In sock- 
eye salmon, substitution of one side for another 
appears feasible; however, variation in the prob- 
ability levels of individual samples suggests that, 
for more complete assurance of reliability, ex- 
amination of a larger number of sockeye salmon 
specimens would be desirable. 
Substitution of gill raker counts from sockeye 
salmon is not advisable. In chum salmon the 
observed differences in left and right side counts 
are not statistically significant; however, the 
differences observed in one high seas sample 
containing 70 specimens raise some doubt in 
situations in which loss of a character, resulting 
in reduced sample size, is a lesser problem than 
maintaining minimum variations in the data to 
be collected. Gill raker counts from opposite 
sides of pink salmon vary equally, and substi- 
tution of the opposing structure will not seri- 
ously affect the analysis. 
In the species considered, branchiostegal 
counts from left and right sides can not be 
reliably interchanged. Also, the extent of the 
asymmetry observed in this character is of a 
greater magnitude in female salmon. Further 
study of sexual differences in left and right side 
variations is required, however, to yield conclu- 
sive evidence of the existence of sexual di- 
morphism in respect to bilateral variations in 
branchiostegals of Pacific salmon. 
Pectoral fin rays from the left and right side 
structures are essentially equal in number in all 
three species. Substitution of sides in obtaining 
counts of pectoral fin rays would contribute 
little variation to mean counts of this character. 
REFERENCES 
Fukuhara, Francis M., Sueto Mural. John 
J. LaLanne, and Arporna Sribhibhadh. 
1962. Continental origin of red salmon as 
determined from morphological characters. 
Intern. North Pacific Fisheries Comm., Van- 
couver, B. C, Bull. 8:15-109. 
IIuBBS, Carl L», and Laura C Hubbs. 1944. 
Bilateral asymmetry and bilateral variations in 
fishes. Pap. Mich. Acad. ScL, Arts, and Let- 
ters 30:229-310, 1 pi. 
Jordan, David Starr, and Barton W. Ever- 
MANN. 1896. The Fishes of North and Mid- 
dle America. U. S. Nat. Mus. Bull. 47, Part 1. 
1240 pp. 
Vladykov, V. D. 1954. Taxonomic characters 
of the eastern North American chars (Sal- 
velinus and Cristovomer) . J. Fisheries Res. 
Board, Canada 11 (6) : 904-932. 
