46 T runsactions. 



Two fairly well-defined spawning-marks are shown in Plate II, fig. 2, 

 and the space between the outer spawning-mark and the edge of the 

 scale represents the growth of the fish between June and October. This 

 growth (1| in.) certainly appears large as against 2^ in. for the whole of the 

 previous year. It seems, however, to be a pretty general rule that the most 

 vigorous growth takes place in the spring, and very little after midsummer, 

 except perhaps in quite young fish. It should also be noted that the 

 posterior end of this scale is well developed. This is usually a characteristic 

 of vigorous growth, and this portion of the scale is usually the first to be 

 eroded when deterioration sets in. 



It has been objected that scales are not permanent, but are shed and 

 replaced by new scales. There may have been something in this objection 

 until Dahl pointed out that scales with the so-called " expanded centre 

 of growth " were in reality " replacement scales,'" and supplied the con- 

 necting-link in a drawing of a scale which had been displaced in its socket 

 but not actually lost. Plate I, fig. 2, shows a particularly fine example 

 of such a " displacement scale," and is of itself almost convincing proof that 

 normally scales are retained throughout a trout's life, and grow with the fish 

 by additions round the outer edge. 



The Matekial. 



The material examined consists of three samples comprising respec- 

 tively 33, 140, and 65 fish taken from the Selwyn River on the occasion 

 of the annual stripping by the Acclimatization Society in 1915, 1917, and 

 1918 respectively, and smaller samples from several other rivers and lakes. 

 I shall deal with each separately. 



Selwyn River. 



Table I (A) gives the complete figures for thirty-three fish, all males, 

 stripped in the Selwyn in June, 1915. The scales were collected by 

 Dr. C. Morton Anderson, who kindly handed them over to me. It is 

 interesting to note that these scales had been simply folded up in paper 

 for nearly two years when I received them, and had not deteriorated 

 during that time. 



The second parcel of scales was taken by myself on the 17th June, 1917, 

 and consists of scales from 140 fish, all being females except one, a 

 particularly large male weighing 10 lb. The full figures are given in 

 Table I (B). 



The average growth-curves are shown in fig. 1. The curve for the 

 1915 fish is a broken line, and that for the 1917 fish a continuous line. On 

 the same diagram are also shown the curves for 173 fish from Lake Mjosen, 

 in Norway, plotted from figures given in Dahl's book. The broken line 

 again is the curve for males, the continuous line for females. In each case 

 the males continue vigorous growth for a longer period than the females, 

 and eventually outstrip them. As the males were from fish taken in 1915 

 and the females from fish taken in 1917, I thought it desirable to test this 

 apparent difference between the sexes further, and with that object col- 

 lected scales from twenty-nine males and thirty-six females at the annual 

 stripping this winter (1918). The full figures are given in Tables I (C) 

 and I (l5). The average growth-curves are shown in fig. 2. Again the 

 raales continue vigorous growth longer and attain a greater size than the 

 females. 



