62 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



100 



no 120 130 140 



Attained length at tagging in millinneters 



50 



160 



170 



Figure 6. — Carapace length increment for one molt of 310 tagged crabs. The line represents the linear regression of 

 growth increment on size as determined by the method of least squares;. 



sizes is essentially constant and for tlie purpose of 

 this discussion we regard the growth increment per 

 molt as being 16 mm. for all male crabs 110 mm. 

 in carapace length and larger. Extrapolation of 

 the regression line beyond 170 mm. may introduce 

 error, but the i-esults are not appreciably affected 

 as only a small proportion of the ci'abs of these 

 larger sizes molt. 



AVERAGE ANNUAL GROWTH INCRE- 

 MENT OF THE POPULATION 



If all adult male crabs molted once annually, 

 their growth would be described as an accumula- 

 tion at the rate of 16 mm. per year. However, the 

 small adults molt annually, but as they increase 

 in size, molting occurs less frequently. Since we 

 do not yet know the molting frequency of indi- 

 vidual crabs, we cannot describe their growth rate. 

 We can, however, detennine the average annual 

 growth of the population by adjusting the growth 

 increase determined from tagged individuals by 

 the proportions of molting crabs observed. 



The nunibers of non-molters (old-shell crabs) 

 and molters (new-shell crabs) by size, observed in 

 samples for the years 1956 through 1959, are 

 shown in figure 7. Shell condition was not re- 

 corded in 1955. 



Since all sizes of adult male ci-abs greater than 

 110 mm. in carapace lengtli were shown to in- 

 crease by approximately 16 mm. per molt, the 

 new-shell distribution for each year was shifted 

 16 mm. to the left. This has the effect of return- 

 ing the new-shells to their size prior to molting. 

 We then smoothed both distributions by a moving 

 average of 7 mm. and calculated the proportion of 

 new-shell to old-shell crabs for each millimeter 

 size class. The result of the transformation, using 

 the 1958 data as an example, is shown in figure 8. 

 By multiplying the proportions molting by 16 

 mm., the average annual growth increment of 

 crabs greater than 110 mm. was calculated for 

 each year's data and shown in figure 9. 



AVERAGE GROWTH RATES 



In any growth study it is highly desirable to 

 define growth in terms such as the growth of in- 

 dividuals or of an age class. Until permanent 

 records of growth are found in crabs, or tagged 

 individuals are returned after prolonged periods 

 of freedom, it is unlikely that the growth rate of 

 individuals can be described. It appears possible, 

 however, to estimate the average growth rate of 

 a year class. 



