80 
EIGHTH REPORT. 
RAINFALL IN THE GREAT LAKE COUNTRY FOR THE TWENTY-FIVE YEARS, 1880 TO 1904, 
BY FIVE YEAR MEANS. UNITY ON EACH LINE IS THE MEAN OF THE 
TWENTY-FIVE YEARS FOR THAT PLACE. 
1880-4. 
1885-9. 
1890-4. 
1895-9. 
1900-4. 
1.12 
0.90 
1.01 
0.95 
1.03 
Alpena. 
1.25 
1.06 
0.97 
0.90 
0.81 
1.18 
0.88 
1.06 
0.95 
0.93 
Lansing.. • • 
Detroit. 
1.20 
0.89 
0.97 
0.94 
1.03 
1.16 
0.84 
1.01 
0.96 
1.04 
1.03 
0.86 
1.06 
1.07 
1.01 
1.16 
0.95 
0.93 
0.91 
1.00 
1.21 
0.99 
0.96 
0.92 
0.90 
1.18 
1.04 
0.91 
0.94 
0.93 
Keokuk. 
1.09 
0.95 
0.88 
1.12 
0.97 
1.18 
0.87 
0.97 
0.89 
1.09 
1.12 
0.86 
0.90 
1.12 
0.97 
1.26 
0.95 
1.01 
0.92 
0.88 
1.17 
0.97 
0.95 
0.95 
0.98 
Peoria. ; . 
1.13 
0.91 
0.88 
1.03 
1.06 
1.07 
1.00 
0.89 
1.04 
0.98 
1.06 
1.01 
1.05 
0.94 
0.95 
Chicago.. 
1.21 
1.02 
0.92 
0.92 
0.89 
a.23) 
0.99 
0.96 
0.81 
0.99 
1.25 
1.22 
(0.96) 
1.13 
0.94 
Saugeen. 
1.00 
0.98 
1.04 
0.96 
0.98 
Mean. 
1.15 
0.96 
0.97 
0.97 
0.97 
Had we no record for Marquette for the years 1880-1884, it is evident 
that since those years were unusually wet ones, the average for the 
years since 1884 would he too small, 97 per cent in fact of the average 
for the whole period. This will be set right by multiplying the average 
'for the twenty years by 100/97ths. There is enough consistence in the 
figures of the table to show the need and value of this reduction to a 
uniform period, enough of irregularity to show that after reduction we 
have not the real value of the precipitation for the twenty-five years, 
but a value that in most cases is nearer to it than the uncorrected 
mean is. 
To satisfy any desire to know what the result would have been had 
the straight means of the observations been used, I have placed them 
-on a blank and reproduce the resulting isohyetals in Fig. 2. It serves 
at any rate to show that the changes introduced by the system of cor¬ 
rections, have been very moderate in amount, and at the same time dis¬ 
tinctly make for consistency and clearness. The method of reduction is 
shown by the following details for Grantsburg, Wis., the first place on 
the list. There are twelve years of measurements at Grantsburg, with 
an average rainfall of 32.77 inches for those years. What we need is 
ihe average fall for the twenty-five years, 1880 to 1904. At La Crosse, 
which was selected as station for reference, the average for the same 
twelve years, column headed n years, is 29.54, for the whole twenty-five 
years 29.86. At La Crosse, therefore, the whole period average was 
it It °f that of the twelve years. We assume the same to be true at 
Grantsburg and multiply the value 32.77 by that fraction, giving' 
ff X 32.77 — 33.1, the reduced value. Latitude and longitude is 
added to facilitate identification of stations on the map. The eleva- 
