1902 - 3 .] Mr Wm. Murray on Salmon in American Rivers. 475 
average yield in 1886 amounting to 21,700 lbs., but the increase is 
94,000 lbs. A comparison of the other averages of the table upon 
the same or varied assumptions will show similar results. 
Saguenay . — In the Saguenay in 1886 the yield had fallen very 
low indeed. The output of fry was, however, roughly speaking, 
doubled between the years 1882 and 1894, and if this were the 
only influence at work, then the yield should have increased in a 
somewhat similar degree, but on the contrary it has risen in 1898 
to six times its volume in 1882. 
Miramichi . — In the Miramichi sixteen years of salmon culture 
seem in 1890 to have had no effect in improving the yield ; the out- 
put of fry was a little increased in the period 1886-1890, and there 
is a large and disproportionate increase in the yield in 1894, which is 
followed in the next period by a fall, notwithstanding the planta- 
tion of more fry. If it be allowed that these tables warrant the con- 
clusion that there have been at least two causes contributing to the 
increases in the yields of the rivers considered, namely, salmon 
culture and the Canadian system of administration, the question 
arises, which of these two has had the greater effect? assuming 
that both of them have had some influence. In order to help to 
decide this point the following table is given. 
Grand River and Gaspe (Table, p. 47 6). — The Gaspe and Grand 
Divisions are two adjoining districts in Canada, in one of which 
artificial culture has been resorted to, whilst in the other it 
has not. In both the Canadian Government have abolished 
fishing stations where possible, and in the Grand River Division 
anglers, by purchasing netting rights in the estuaries, have assisted 
the efforts of that authority. The result shows that whereas the 
yield of the Gaspe Division, aided by artificial culture, has in- 
creased 173 per cent., that of the Grand Division, where very few 
fry have been planted, has risen as high as 364 per cent. It is un- 
fortunate that tables of this kind cannot be multiplied, but unless 
the whole circumstances of the rivers are known and can be dis- 
counted, as it is hoped has been done in the case given, comparisons 
may mislead. The writer has printed elsewhere a table of the Grand 
Cascapedia river, adjoining the Ristigouche District, in which there 
has been little or no salmon culture. The Cascapedia shows a de- 
cline of f per cent., yet the table should reach back ten years 
