LIFE HJSTOliY OF THE COD 109 



SESSIONAL PAPER No. 38a 



in the summer of their fourth year, i.e., three-year class, being determined from 

 the other curves to bo mentioned hiter. All the fish except those in the last 

 humps (66 to 71) may be safely taken as 3-year-olds. Another curve was plotted for 

 all the fish caught between June 11 and 24. This showed oiriy a slight indication 

 of division into year-classes. Most of the remaining fish (those caught in July and 

 August) were plotted in another graph. The interesting things about this 

 curve were the way the little fish (which were afterwards found to be less than one 

 year old) fell into two groups representing two different catches taken about five 

 weeks apart. The averages of these two humps indicate a growth of 3-6 centimetres 

 in that time. The one year class is represented by only one fish and the two and three 

 year classes can easily be distinguished. 



Since the value of the next part of this paper depends so much upon the age 

 determination of the fish, it will be necessary to explain how the scale is an indicator 

 of the age of the fish. The youngest scale ever observed by me consisted of a single 

 central plate, quite homogeneous, with a single ring of smaller plates around the 

 margin. It was taken from a fish which measured 3-00 cm. The next smallest is 

 from a fish 4-10 cm., and it clearly shows the central plate with three 

 rings of smaller plates around it. From this it seems reasonable to suppose 

 that when the fish starts out in life its scales consist of single plates, and as it grows 

 it adds rings of smaller plates around the central nucleus of each scale. Since tlie 

 number of scales on the fish does not generally increase throughout life, the linear 

 growth of the scales may be expected to be proportional to the linear growth of the 

 fish. It is found that when the rate of growth of the fish is greater, i.e., in tlie 

 summer, the plates laid down are slightly larger than those laid down when the rate 

 of growth is less. A glance at any old scale reveals a more or less regular alternation 

 in the open and close bands, in the first three years at least, signifying a regular 

 periodicity in the growth of the fish. It has been demonstrated beyond doubt by 

 other investigators that this periodic retardation and acceleration in the growth of 

 the fish is brought about by the alternation of winter and summer, the close band 

 representing a winter's growth and the open a summer's. Another factor which retards 

 the growth of the fish and consequently leaves a mark on the scale is the spawning 

 period. In some kinds of fish, according to other investigators, the -spawning rings 

 can be clearly distinguished on the scales, but in the cod my experience has been 

 that they only lead to confusion between winter and summer rings, making it almost 

 impossible to tell with any degree of certainty the age of the older fish that have 

 spawned many times. Doubtless there are other things which affect the growth of a 

 fish besides the seasons and spawning periods and consequently the markings on the 

 scales; for example, scarcity of food, or temporary incapacity of the fish to obtain 

 food. Indeed I have seen some cod scales in which it was practically impossible to 

 notice any distinction between summer and winter rings, and others in which there 

 appeared to be more winter rings than would be expected from the size of the fish. 



In spite of this drawback in the method, the ages of nearly all the fish from which 

 scales were taken were determined and are appended in the tables above. In order 

 to appreciate the significance of the ago in relation to the length, the fish were plotted 

 again in their different year classes. This showed that the majority of fish caught 

 for commercial purposes are between two and four years old, and that the greater 

 number of them fall into the three-year class. 



In taking measurements for cod very few opportunities are offered for determiti- 

 ing the sex. Those that were taken are shown in the tables and use made of them in 

 constructing the rate of growth curves. 



To further test the assumption that the rate of growth of the scale was proportional 

 to the rate of growth of the fish, the following construction was made: A scale was 

 drawn with the camera lucida. From the centre of the scale. A, a line was drawn 

 to the periphery, B, usually in the direction of the long axis of the scale. Another 



