SALMON— EARLY AND LATE RIVERS. 63 



tlie Derwent, 35| miles long, and 268 square miles of catcTiment basin; the Bhen, 

 14| miles long, with. 59 square miles of catchment basin ; the Irt, 16 j miles long, 

 with 48 square miles of catchment basin ; and the Leven, 6| miles long, and with 

 123 square miles of catchment basin. The two first, with the longest course, are 

 early rivers ; but not so the remainder, which are rather late than otherwise. 

 Irrespective of which some of the earliest and some of the latest rivers are among 

 such as have no lakes in their course, and are of very different lengths. 



Buckland held that large estuaries at the entrance of salmon rivers must have 

 the same influence as lakes near their sources, or in their course ; but detailed 

 investigations hardly bear out this theory, while it has been remarked that in a 

 single estuary, as of the Severn, three rivers with different degrees of earliness 

 enter. It has also been suggested that the conditions which occasion the early or 

 late ascent of fish into fresh water may have some connection with the circum- 

 stances of the marine life of the salmon, the food which it is able to obtain, or the 

 enemies from which it has to escape. 



What the temperature or other conditions of the water in the river, or the 

 food that is present therein at various seasons of the year, have to do with the 

 period at which salmon ascend into it from the sea, are worthy of study.* 

 Jardine, Yarrell, and others have remarked that the temperature of the river- 

 water exercises some influence in the time of the ascent of these fish, they 

 usually first selecting those that are warmest (of coarse in moderation), while in 

 most streams the eggs would be more rapidly hatched ; thus, other things being 

 equal, the yoang fish would be hatched earlier ia the warmer than in the colder 

 streams. The Salmon Oommissioners of the Tweed remarked in 1867 that it was 

 stated by several witnesses that salmon do not enter the river freely when fuU of 



tells me he has constantly done so " (p. 20). " This year (1885) I have heard of fish spawning as 

 late as the end of March ; and one year I heard of fish, full of spawn, being caught in April. On 

 the Usk, on the 18th of May, 1861, a fish was taken full of spawn" (p. 59). Mr. John Anderson 

 informs me that in the Teith, above CaUender, he has seen plenty of salmon spawning during the 

 last week in October. Praser (The Salmon, 1833, page 7) that in the upper rivers that feed 

 Lochness they spawn from the middle of October to the middle of November. 



The Usk Board of Conservators reported in 1885, " that the tJsk has undoubtedly become an 

 earlier river than formerly. Twenty years ago but few salmon spawned in November ; the first 

 great run was at Christmas, and the second in January; considerable numbers spawned in 

 February and a few in March. Now occasional fish spawn in October, large numbers in November 

 and December, comparatively few in January, and only occasional fish in February. In conse- 

 quence the additional water-bailiffs are put on earlier and discontinued earlier. The change has 

 been gradual." There has been no change in the condition of the water. Mr. Armistead (Land and 

 Water, Nov. 10th, 1885), remarked on the foregoing: " Possibly the fact of their having more to eat 

 and less to do may have something to do with this ; but I believe that temperature has more to do 

 with the time of their spawning, and with apparent deviations from rule than we are at present 

 inchned to suppose. In the case of our river here, the Nith, sahnon, after a run of a few mUes, 

 arrive at a fork called the Cluden. They spawn earlier in the Cluden than in the Nith, and I had 

 thermometers carefully placed in both streams and examined daily at the same hour all through 

 the spawning season, and the Cluden water proved to be rather warmer than the Nith. It would 

 be interesting to know, if from any change, say in the drainage of the land, warm springs from 

 mines, or any other cause, the Usk water has in any degree risen in temperature the last few years. 

 I have known cases in which the cutting down of wood has had considerable influence on streams. 

 Those who have had much experience with domesticated trout know well enough what an 

 influence the weather has upon them, a warm or cold rain making a great difference, for that day, 

 at least, in the yield of eggs. It is now known, for instance, that a trout or salmon when ripe, 

 and even partially spawned, can still hold its ova for a month if desired in certain cases ; indeed, 

 I have known them to do it longer. On the other hand, pisciculturists know how to make fi^ 

 spawn earlier than they would do if left to nature." 



* Temperature has often a direct connection with the food which may be present, while 

 because that of a river exercises a manifest influence as to the when and where, the eggs of the 

 SalmonidsB are deposited, it does not follow that temperature is the sole cause which induces 

 salmon to enter certain rivers. 



Livingstone- Stone in the Bulletin of the United States Fish Commission, 1882, vol. ii, 

 page 12, observed of Salmo quinnat that when raised in the United States it did not spawn at 

 the Oalifornian time but at the Sahno solar time, November or nearly so. When investigating 

 the ascents of oarps in Asia into hill rivers for breeding purposes, I observed that it was an almost 

 invariable rule that they turned aside from the main snow-fed streams into the side affluents in 

 order to deposit their eggs, this being probably done in order that the young should be hatched 

 in water not replenished from melting snows because such would be deficient in food. 



