104 



Appendices to Twenty-second Annual Report 



" shedding milt and fertilising ova," and from our experience of winter 

 netting there seems some evidence for a belief that in October, in the 

 Tay, a run of male fish may ascend from the sea with some annual 

 regularity. Yet many males may be caught in November which 

 cannot be used for the purpose of fertilising ova, or from which milt 

 cannot be pressed, and which may be kept a considerable time without 

 showing marked signs of being ripe. It may be that were such fish paired 

 naturally and swimming freely in the river they would be more rapidly 

 or more responsively fertile. Be this as it may, it is clear at anyrate 

 that every female must have the attendance of one male, if full 

 advantage is to be obtained from the breeding stock of any river. 



While carrying on the investigation into the migrations and growth 

 of the salmon by means of marking, a very considerable amount of data 

 has been collected, which bears upon this question of the proportion of 

 the sexes. Yet care has to be exercised in using this material, or in 

 drawing deductions from it, because indications are not wanting that 

 the habits of males and females may differ sufficiently to give a rather 

 erroneous result ; runs of fish may be noted which contain an extra- 

 ordinary proportion of either one sex or the other ; or males may 

 incline to ascend tributaries at a time when females heavy with spawn 

 remain in the main river. Data distinctly limited in extent must be 

 discarded, and on this account I omit from this paper mention of 

 particulars from the Rivers Ugie, Ythan, Ewe, Beauly, Conon, Alness, 

 Ness, Doon, Snizort, Torridon, and Balgay. 



The material I select is of two kinds : — (1) That concerning breeding 

 fish marked in the Annual Close Time just before or actually during 

 spawning ; (2) That with reference to spent fish or kelts. 



I regard the netting during the close time as the most reliable guide 

 to the proportion of the sexes at spawning time, and yet we are in no 

 way certain that through possible differences in habit we have actually 

 netted a representative selection of the fishes present in the river. 

 During the close seasons 1901, 1902, and 1903 I have personally 

 conducted netting experiments in the River Tay, chiefly in the lower 

 reaches from the Linn of Campsie down to Almond Mouth about two 

 miles north of Perth, but also to some extent in the tidal waters 

 between Perth and Newburgh, and in the upper reaches from Logierait 

 to Dalguise. Even in the stretch of river between the Linn and Almond 

 Mouth (about 6 miles) there are naturally many wide, shallow, and 

 rapid reaches where a net cannot conveniently be drawn, and where 

 breeding fish were constantly to be seen leaping and splashing on the 

 surface, as their habit is at this season. But netting was certainly 

 conducted as completely and as skilfully as possible by the experienced 

 fishers of the Tay Fisheries Company, to which Company I am greatly 

 indebted for much generous assistance, and at each visit to this section 

 of the river, netting was carried on for three or four consecutive days. 

 It may be that fish out on the spawning fords were more or less 

 accurately paired, and therefore would have shown a more evenly 

 balanced result as to sex than did the fish in the pools which were 

 netted. But, where possible, as for instance at Almond Mouth Pool — 

 the most productive netting station we found — the shallow ford above 

 the pool was reached with the net and the fish drawn down into the 

 pool below. Here in one day we took 123 fish, 38 of which were males 

 and 82 females, all unspawned, besides 3 female kelts, showing the 

 striking absence of males so frequently noticed. 



I subjoin the data, selected from the register of particulars of marked 

 fish, in the form of a Table, separating unspawned fish from kelts. 



