580 MR. J. A. MILNE ON THE 



limits being about the middle of March and the middle or end 

 of September. 



The foregoing remarks may suffice to explain the regular 

 markings on salmon scales, and how they show the age and time 

 of migration of the salmon. )Scales, however, may show much 

 more. 



Mr. Johnston noticed that the scales of the older fish some- 

 times seem to have stopped growing, and become jagged and 

 broken at the edges, and then to have gone on growing again. 

 When this has happened the lines of the new growth do not 

 exactly follow the contours of the lines of the old growth, and 

 thus a clear mark is formed round the anterior portion of the 

 scale, often also apparent as a thickening round the posterior, 

 the unlined, portion. Further, it was noticed that when such a 

 mark occurred the number of lines between the winter bands on 

 each side of it was frequently far fi'om normal. Tlie normal 

 number differs somewhat in different fish, and also within certain 

 limits on scales from difTerent parts of the same fish ; but nevei- 

 theless it is fairly constant, and any wide departure from the 

 usual state of things required explanation. A reason for the 

 mark suggested itself to Mr. Johnston from the known fact that 

 salmon, whether or not they take food in fresh waber, do not 

 take it in sufficient quantity, after the smolt stage is passed, to 

 nourish them. Therefore neither salmon nor their scales can 

 grow in fresh water. It was further noticed that most scales 

 taken from kelts — that is to say from salmon which, having 

 spawned, have not yet returned to the sea — Avere broken and 

 torn at the edges, from these facts Mr. Johnston argued that 

 the sort of marks illustrated at and D on the accompanying 

 photographs (text-figs. 100 & 101) showed that the fish bearing 

 them on their scales had entered fresh water and stopped growing- 

 there ; that they had spawned and become shrunken after 

 spawning, so that, from mechanical reasons, the scales being im- 

 bricated, i.e. overlapping each other like tiles on a roof, had become 

 frayed at the edges ; and, lastly, that on the salmon's return to 

 the sea the new growth had started again evenly round the scale 

 and thus left the mark. Spawning operations mean a winter 

 at least spent in fresh water and, therefore, if the fish enters 

 a river early in the year there will be fewer than the normal 

 number of lines between the winter bands formed on each side 

 of the spawning mai-k. If it comes in early in the spring there 

 will be none in that year, and in the year following there will 

 be fewer than usual on account of the fish taking some time 

 after its return to the sea to make up lost condition before it 

 starts again to increase in size. Of course, the later the 

 salmon came into the river the more lines would have been 

 formed behind the mark, and the later it returned to the sea the 

 fewer would there be in front of it. 



Besides the irregular arrangement of the lines on the scales 

 when a fish has spawned, there is a thickening round what was 



