As we have said above, the greater part of these breeders ascend in August and September, and the throng of the spawning process takes place in November and December.  The earlier spawning begins in September with only a few pairs, generally grilse; and from that period the numbers increase till the first week of December, when the operation has attained its height.  It then gradually decreases until February, when perhaps only a few pairs are seen at work.  Mr Young informs us that sea-trout are seen spawning a week earlier than grilse, and grilse a week earlier than salmon.  He does not mean that all grilse spawn before salmon begin, but that they are observed working a week before the latter have commenced.

Mr Shaw informs us, (in his last letter,) that it is an exceedingly rare occurrence to find an unspawned fish in the rivers of Dumfriesshire in the month of March.  On one occasion, however, about twenty years ago, he observed a female salmon spawning in the Nith about the 10th or 12th of March, but unaccompanied by any male.  He can also call to mind a pair of salmon having been observed spawning in the Ettrick so late as Selkirk March fair, which is held during the first week of April.  This, however, we believe to be a very rare occurrence, notwithstanding Mr Scrope’s statement, that he has in the Tweed “caught full roaners as late as May.”  These seem to be anomalous or accidental instances, and we are not aware that any evidence has been brought forward to prove that they still seek the spawning beds in pairs at that period, or produce what may be called autumnal fry.

The usual spawning period in the south-west of Scotland extends from about the middle of November till the middle of February; but the busiest months of that period are December and January, when the salmon spawn in great numbers in the Nith, about Drumlanrig.  From the circumstances of the largest salmon visiting the rivers at that season, Mr Shaw is induced to think that they are likewise the oldest; and that, as they increase in years, they desire to remain the longer in the sea, visiting the fresh waters only during the breeding season.  The spawning period of sea-trout, he informs us, is from about the middle of October until the middle of December, the principal period being the whole of November, when the various streams and tributaries are taken possession of both by sea-trout and herling, spawning in deep or shallow water, according to their individual size.

But in reference to the point in question, that cold accelerates the spawning process, let us take for a moment the general basin of the Oykel waters into view.  We know that for several seasons back, the earliest spawning in that quarter has occurred in the Carron, in September.  Now, it is certain, that during that month the Carron waters are warmer than those of the Shin.  So also the Oykel (properly so called) is itself two degrees warmer in October than the Shin, and yet the latter is the later of