GoDBY. — (jfroirfJi of Broini Trout in ('a/iferburi/. 43 



pass through a period of rapid growtli followed by a period of comparative 

 stagnation. This periodic growth has generally been attributed to changes 

 of temperature and corresponding changes in the abundance of food-sujiply ; 

 and in regard to many species of fish it has been demonstrated that the 

 maximum rate of growth roughly coincides with the maximum temperature 

 of the water. There is evidence, however, to show that this periodic growth 

 is well marked in the scales of some deep-sea fish, which can hardly be 

 subject to any marked seasonal changes of temperature, and in the case 

 of the squeteague {Cynoscion regalis) Taylor (6) has shown that the period 

 of stagnation roughly coincides with the spawning season in midsummer. 

 It seems probable, therefore, that the period of stagnation is determined 

 more by a voluntary fast during the spawning season than l)y any actual 

 shortage of food, and that individuals which have not arrived at sexual 

 maturity subject themselves to this annual fast, though not to the same 

 extent as the mature specimens. This voluntary-fast theory is further 

 borne out by the observations of Masterman (5), who in a most careful 

 critic|ue of the previous work on salmon points out that a certain number 

 of salmon captured at sea throughout the summer show no evidence of 

 summer feeding. He concludes that some salmon start their spawning- 

 fast many months before entering fresh water. This may cause the age 

 of salmon to be underestimated in some cases, and certainly throws grave 

 doubt on Johnston's claim that he can tell approximately the month of 

 entering fresh water. In the case of trout there is no evidence of prolonged 

 fasts, except during the spawning season, which occurs in midwinter, and 

 it is of little importance whether the cause be lack of appetite or lack oJ' 

 food. There is some evidence to show that in Canterbury the u\aximum 

 rate of growth, especially amongst the larger fish, occurs in spring rather 

 than in summer. It is probably quite safe to assume, however, that the 

 period of stagnation occurs in the winter. 



Roughly speaking, a trout-scale (Plate I, fig. 1) consists of a transparent 

 plate of more or less elliptical form, having its centre of growth approxi- 

 mately at one of the foci. Surrounding this and roughly concentric with 

 the outer edge of the scale are a number of lines or " circuli." The scale 

 grows by the addition of these circuli round the periphery, which are added 

 in greater numbers and more widely spaced during the periods of rapid 

 growth. This alternate spacing and crowding produces light and dark 

 zones, one light and one dark corresponding to a complete year's growth. 

 The dark zones are called "' annuli," or " winter bands." 



In the case of spawning fish the stagnation is more complete, and the 

 winter band is narrower and more clearly defined. In salmon (Salmo salar) 

 the act of spawning leaves a clearly defined scar or " spawning-mark " on 

 the scales, due to disintegration or reabsorption of the scale, especially along 

 the lateral edges and the outer surface containing the circuli. A true 

 spawning-mark is not very common in trout, but the character of the 

 winter bands gives a fairly reliable indication of spawning. Plate I, fig. 1, 

 shows one such winter band. 



The exact cause of the spawning-mark in salmon is still in dispute. 

 Johnston (4) attributed it to the vicissitudes of river life, whereby the fish 

 shrank in girth, and says, '" The compression of imbricated scales tends 

 to increase the amount of overlap, and from this or dermic influences we find 

 that their margins become ragged or frayed." Masterman (5) has shown 

 that this fraying or erosion in many cases starts long prior to the fish's 

 entry into fresh water, and concludes that the phenomenon is one of 



