72 



Appendices to Twentij-first Annual Report 



support of this view, and I am of opinion that fish which are already in 

 the fresh water are induced to ascend further for the same reasons, but 

 are checked and frequently made to descend into tidal water again by the 

 advent of floods of markedly cold water caused by snow or melting ice. 

 The recaptures of marked clean fish during the past winter have strongly 

 supported this view. In this respect the early-running fish of our large 

 rivers seem to differ from the salmon of Norway and elsewhere. 



When we extend our range of observation to rivers some distance from 

 each other it is at once possible to see that, although rivers, say the 

 Solway, are warmer than rivers farther north, this has little or nothing to 

 do with the respective times at which fish run. A glance at the accom- 

 panying charts, showing weekly means for the Nith at Dumfries and the 

 Tay at or near Perth, is sufficient to show the late river Nith warmer 

 than the early Tay during the first five months of 1902. If we com- 

 pared the Tay with the Severn or the Thames, or still more, perhaps^ with 

 the Tamar or Fowey in Cornwall, I have no doubt we would find the 

 difference much more marked. The temperature of a river seems 

 to follow close on the heels, so to speak, of the air temperature of 

 the district, and therefore we have only to pick out rivefs of a cold and of 

 a warm or comparatively warm district to obtain the conditions which 

 have been from time to time held up as those which cause rivers to be 

 respectively late or early. Yet the Thurso in the extreme north is 

 remarkably early, while the Nith in Dumfriesshire is late. And if 

 further evidence for this negative result were wanted it may be found by 

 comparing the curves here given of the Ness and Tay. This theory may, 

 I think, therefore, be entirely discarded. 



Now let us glance at the question of the relative temperature of a river 

 and of the sea into which it flows. I pointed out last year that such a 

 river as the Tay is, during the winter months, colder than the North Sea, 

 and yet that I caught clean fish ascending in November, December, and 

 January. 



I desire now to turn to the conditions found in the case of the late 

 river Nith. The records given in the Nineteenth Report were incom- 

 plete, and the position chosen for the testing of the estuary temperature 

 was considered unsatisfactory, being scarcely reached by the full influence 

 of neap tides. It being desirable, therefore, to have mure complete 

 readings from this interesting river, I asked the Rev. Wm. Andson, who 

 acts as observer, at Dumfries, for the Meteorological Society, if he would 

 be kind enough to take water temperatures. This he consented to do, 

 and a pair of Philipp's thermometers (maximum and minimum) were 

 placed in a metal box below the landing stage for pleasure boats which 

 exists above Dumfries Bridge. A water gauge is situated here, and this 

 was also read. Mr. Andson was also good enough to arrange that Rev. 

 Mr. Malcolm would take readings of the temperature of the estuary at 

 Glencaple. This was done daily with a dipping thermometer at or about 

 time of high water. Errors of instrumentation have been ascertained for 

 the three thermometers used. To these gentlemen I am greatly indebted 

 for the most careful series of readings which I am now able to present. 

 I have calculated the weekly means and plotted them on the accom- 

 panying chart, together with the readings of water level. 



It will be noticed that through January, February, and March the red 

 and the black line are remarkably close to one another, that in April the 

 estuary becomes rather warmer than the river, and in ]\Iay rather colder. 

 I have no doubt that the great expanses of sand which occur at the mouth 

 of the Nith, and which are laid bare by every receding tide, have a certain 

 controlling influence upon the temperature of the estuary; but, be this as 

 it may, we have here the exact conditions which, according to certain 



