51 
1897-98.] Migratory Movements of Salmonidce. 
Isolating together the state of water and the totals as to 
numbers of fish, we find that — 
During 40 normal days, there were 178 fish. 
8 J-flood days, 
5J 
72 
9 !-fiood days, 
5) 
69 
1 full-flood day, there 
was 1 
58 
320 
This result shows the greatest ascent of fish while the river is 
in quarter-flood, there being then 9 fish a day, as compared with 
7f fish a day during half-flood, and scarcely 4| fish a day during 
normal flow. The result as to full flood is necessarily of small 
value, since, during a flood of full force, observations are impos- 
sible. Reference to the table will show, however, that the two 
highest totals, viz., 26 fish and 25 fish (9th and 10th Dec.) occur 
during normal flow ; but this normal flow succeeds three days of 
quarter-flood, during which we may presume the fish were working 
up the main river. No marked rise in the number of fish recorded 
is noticeable during any prolonged period of normal flow, yet it is 
evident, by the figures which occur between two rises of the water, 
that with only the averageflow (1 2,000,000 gals, in twenty-four hours) 
both salmon and sea trout had no difficulty in entering the stream. 
Only one other point may be noticed before leaving this subject. 
A period of normal flow existed for thirty-two consecutive days from 
9th Jan. to 9th Feb. During the nine first days fish occurred, but 
a gradual diminution in number is noticeable. From 20th Jan. to 
10th Feb. no fish are recorded, then immediately, with a rise to 
half-flood, we have the three last records of salmon in Talla. 
From a consideration of these data, therefore, it seems allowable 
to infer that the condition of water-flow preferred by fish when 
ascending from a main river to a tributary, for the purpose of 
spawning, is a moderate rise rather than a flood ; and that a 
succession of moderate rises would fulfil the conditions under 
which most fish would be enabled to reach the upper spawning 
grounds of a large river such as the Tweed. Further, it seems 
clear that fish ascend in limited numbers when the water is in its 
normal condition. 
A record of water temperatures of Tweed and of Talla was also 
