GROWTH AND MORTALITY OF PACIFIC RAZOR CLAM 
553 
Karls Bar). The isotherms follow in general the sweep of the gulf but the beds in 
Shelikof Straits lie north of the mean annual isotherm of 40° F. which passes approxi- 
mately through those near Cordova. For this reason we have used the position along 
the coast measured in miles rather than the latitude as indicating the geographical 
position. 
Controller Bay, Alaska ( lat . 60°; 2,040 miles jrom Pismo). — This bay is a broad 
shallow body of water, with much of the bottom exposed at low tide. The upper 
part of the bay is covered with fine glacial mud, and the outer bars are of sand with 
some glacial mud. During the summer the Bering River empties a large amount of 
glacial drainage into the bay, making the water cold and filled with muddy silt. We 
consider Controller Bay a very unfavorable habitat. The beds will not support 
commerical operations. 
The clams furnishing the Karls Bar growth curve were taken from a small por- 
tion of the beds in Orca Inlet near Cordova (lat. 60° 27' N.; 2,105 miles from Pismo). 
The soil is fine sand with some glacial silt. The bed is not exposed to the surf; and, 
although it differs greatly from the southern beds, it is typical of the Alaskan clam- 
producing areas. The beds in this vicinity have been dug for 12 years, but the area 
from which these specimens were taken had not been previously exploited. 
Swickshak, Alaska (lat. 58° 5' N.; 2,405 miles from Pismo). — This bed is on the 
north of Shelikof Straits opposite Kodiak Island. The beach is of fine sand and vol- 
canic ash and is more exposed to surf than most of the northern beds. This bed was 
first dug in 1923 and our material was taken during that year and in 1924. 
Clams are taken in Hallo Bay (lat. 58° 5' N.; 2,425 miles from Pismo) on the 
northern shore where it is protected by the island in the bay. The beach is of sand, 
volcanic ash, and glacial mud. The shells were taken from among the first clams dug 
in that place. 
QUANTITATIVE TREATMENT OF DATA 
We may define growth as increase in size and take, in the present case, length as 
a measure of size. Quantitative data on the length of the clam will be considered. 
This may be presented as absolute growth; that is, average total size at each age or 
the gross annual increments. Or we may show the relative growth rate; that is, the 
proportional or percental gains at each age. The graphic representation of growth 
may be cast, therefore, in two contrasting pictures, both of which contribute to an 
understanding of growth. 
Absolute growth. — -We shall first consider absolute growth. In Figures 4, 5, 
and 6 are presented norms of growth for 10 localities; the corresponding medians are 
given in Table 8. These are plotted in the usual fashion; that is, they are the regres- 
sions of length on age, or average length for each age-group. The converse regres- 
sion, age on length, is closely similar in early life, but differs significantly in the 
later years (Weymouth, McMillin, and Rich, 1925). 
The length of the first ring varies roughly with the latitude of the beds from 
which the clams are taken, ranging from one-third of a centimeter in the north to 
2 centimeters in the south. These differences appear to result from the higher tem- 
perature in the south which favors a more rapid growth and, permitting an earlier 
spawning, furnishes a longer growing season. 
Following the formation of the first ring, growth is rapid. In the southern 
beds, over two-thirds of the length is reached during the first growing season. In 
the north growth is slower and a comparable increase requires over four years. In 
the northern forms it will be noted that the curve of absolute growth rises slowly at 
