AGE AND GROWTH OF THE CISCO 
307 
lations of the present investigation. The consistent results obtained appear to justify 
the assumption. 
3. The calculation of growth from scale measurements was based on the assump- 
tion that body length and scale diameter show a constant ratio at allUengths beyond 
that at which the first annulus is laid down. In the Silver Lake cisco alone were there 
any discrepancies between growth as calculated from fish of different age and as de- 
termined empirically that might throw doubt on the validity of the assumption. 
4. It was demonstrated that Lee’s phenomenon in the Silver Lake cisco does 
not depend on changing body-scale relationships with increase in length and age, 
but rather that the growth discrepancies are reflected in the actual scale measurements 
upon which the calculated growths were based. 
5. Among the suggestions for the explanation of Lee’s phenomenon in the Silver 
Lake cisco were: (1) Selection by gear, (2) selection due to dissimilar distribution 
within the lake of the various elements of the population, and (3) selection due to 
differential mortality correlated with individual growth rate (greater mortality 
among individuals with more rapid growth). 
6. The four populations (Trout Lake, Muskellunge Lake, Silver Lake, and Clear 
Lake ciscoes) show wide differences in the amount and rate of growth both in length 
and weight. Their order with respect to growth rate in length, from minimum to 
maximum, is: Trout Lake, Silver Lake, Muskellunge Lake, Clear Lake; the order 
with respect to growth in weight is: Trout Lake, Muskellunge Lake, Silver Lake, 
Clear Lake. 
7. On the basis of data on the growth of the cisco in this and other publications 
a “cisco-type” of growth was described. 
8. On the whole the high degree of overlap of the length distributions of consecu- 
tive age groups of the cisco makes length alone a poor index of age. 
9. Because of individual variation in growth rate the largest fish within a popu- 
lation is frequently not the oldest. 
10. Condition was described in terms of the quantity K in the equation, 
IT=.£lX10 -5 L 3 {W— weight in grams; Z=lengtli in millimeters). It was pointed 
out that coefficients of condition calculated from the cube relationship describe 
relative heaviness independently of the general length-weight relationship, and are 
more satisfactory measures of condition than the quantity C in the equation W—CL n 
where the value of n is determined empirically. 
1 1 . The four cisco populations show wide differences with respect both to aver- 
age condition and the manner of change of condition with change in length. The 
Muskellunge Lake cisco and the Trout Lake cisco show a loss of condition with increase 
in length (the rate of loss of condition is the more rapid in the former population). 
The Silver Lake cisco and the Clear Lake cisco, on the other hand, show improvement 
of condition with increase in length (the rate of improvement in condition is the 
greater in the Clear Lake cisco). The order of the four lakes with respect to the 
average condition of the fish in the samples, from minimum to maximum, is: Muskel- 
lunge Lake, Trout Lake, Silver Lake, Clear Lake. 
12. In all four populations the relationship between length and weight may be 
described over certain length intervals by equations of the type, W=CL n , where 
IT=weight, Z=length, and C=a constant. In some of the populations, at least, 
these equations cannot be extended to lengths outside the range of the empirical 
data. 
