676 PROCEEDINGS OF SECTION H. 
Taste VII. 
Showing amount of summer rainfall in years of high mortality :— 
Mortality per Inches of Rain, Nov. 
100,000. to April. 
1866 102:7 6°73 | 
1877 99:3 17°52 | 
1878 111-9 15°66 
1883 90°6 14-03 
It is not easy to find, in these tables, any confirmation of the 
doctrine that the prevalence of typhoid stands related to the 
rainfall, very low mortality in different years being the accompani- 
ment, either of an exceptionally wet or an exceptionally dry 
summer. In the same way, a very high mortality has fallen in 
snmmers which also exhibited both extremes as regards rainfall. 
In so far, therefore, as the level of the subsoil water can be 
considered to vary with the amount of rainfall, it must be taken 
as shown that the amount of typhoid in Melbourne does not 
depend on the rise and fall of that level. This must be regarded 
as proved in respect of successive years. It may almost be held 
to be true even of the seasonal variations, since, as is shown by 
the last tables VI. and VII., the summer rainfall of Melbourne 
is equal to the average of the whole year, the supply by pipes 
from without being much greater, quite a large proportion of it 
soaking into the ground. 
In how far variations in the barometric pressure have any 
influence on the prevalence of typhoid remains to be shown. It 
it does exert an influence it must be by favouring or checking the 
escape of emanations from the soil ; and the effect must be direct, 
i.e., the barometric pressure, on this hypothesis, ought to be excep- 
tionally low in epidemic years. For the purpose of discovering 
in how far this is the case, I have presented in two tables, to 
correspond with VI. and VII., the barometric pressure given, 
being the average of the mean pressures for the months November 
to April. 
