ONE of the primary problems in fishery biology is that of distinguishing discrete populations of food fishes. This is frequently done by morphometric studies of local forms, and Ehrenbaum (1929) has pointed out that such studies have merely a theoretical but also ... practical importance in the interests of rational utilization of the stock. It has been suggested by Lea (1929) that characters might serve as accessory race characters. He and his associates have shown that the scale patterns in Norwegian herring, Clupea harengus Linnaeus, reflect differences in types, and also reveal place of origin and the stock to which an individual or a population belongs. Ford (1933) experimented with regression equations of the type y = a ?bx as a technique for the comparison of in length in different geographical areas. Other biologists, for example Biickmann (1938) also have discussed the application of this equation to phenomena. This technique is based on the correlation between calculated length of herring at the formation of a given winter ring (In) and the calculated length at the time the next ring is laid down (In. +). Thus , + 1= a bl,. The formula also indicates that follows a geometric progression with annual increments becoming smaller and smaller in constant ratio. Differences due to annual changes in conditions of are recognized, however, and the geometric law is regarded as the smoothing out of real fluctuations which are likely to recur year after year. Differences in between herrings of different geographical areas were expressed by Ford in a series of regression equations and calculated limiting values, or asymptotic limits of growth, which changed from south to north. These equations were derived from data on mean calculated lengths of herring from 11 to as old as 113 caught in the North Sea, the English Channel and off Iceland. The values for the slope of the regression lines varied from around 0.6 in Lowestoft herring to 0.74 for the Icelandic forms, with predicted ultimate size (y-intercept -complement of slope) similarly differing from 27.5 cms. in the south to 36.2 cms in the north. Walford (1946) described a convenient graphic method for plotting that part of the that lies beyond the inflection point of the ordinary sigmoid curve that is derived by plotting length against age. He transformed this portion of the usual curve into a straight line by plotting length at age n + 1 against length at age n. This was shown to apply commonly (but not always) to the of individuals of various types of organisms from clams to man, as well as to averages of groups of individuals in a population. Walford (op. cit.) derived two growth the same as the slope and the limiting value derived by Ford from the regression equations. These characteristics, descriptive of the of different organisms, he