Though the observations are not taken at enough places, nor generally extended through a sufficient number of years to give the differences of temperature with a great deal of accuracy, yet we can see from the tables and sketches the great influence the lakes have upon the temperature in their vicinity.

The tables also show another effect, which it was almost impossible to show by the curves on so small a scale; that is, the higher temperature in winter and lower in summer of places on the lake shore compared with that of places twenty or fifty miles inland.

The general form of the curves for the seasons will be seen to be about as follows:

Winter-January, February, December.-The first three curves rise a little near Lake Superior, and then drop toward the eastern end of Lake Ontario. The next two or three rise rapidly and pass through the northern part of Lakes Michigan and Huron, and then slightly drop toward the eastern end of Lake Ontario. The next is inflected nearly the same but in less degree, and the lower nearly straight.

Spring-March, April, May.-The upper and lower curves are about the same as those of winter, but the center ones have a tendency to drop round the south end of Lake Michigan, and then rising in the lower peninsula of Michigan, drop again round Lake Erie, and then rise parallel to Lake Ontario.

Summer-June, July, August.-The first curves pass near the west end of Lake Superior, then drop round the south end of Lake Huron and rise to the eastern end of Lake Ontario. In the middle curves the deflection round the south end of Lake Michigan, the rise between Lakes Michigan and Huron, and the further deflection round Lakes Erie and Ontario, are more marked than in the spring curves. The lower curves are nearly parallel to the lines of latitude, but slightly inflected, the same as the middle ones.

Autumn-September, October, November.-The higher curves rise slightly toward the western end of Lake Superior, then drop north of Lake Huron to the eastern end of Lake Ontario. The next rise rapidly; passing through the northern part of Lake Michigan, are deflected downward to Lake Erie, and rise again toward the eastern end of Lake Ontario. The next are slightly inflected in the same manner as the last, and the lowest are nearly straight, bending a little south in the center. Thus the effect of these large bodies of water is to depress the summer and to elevate the winter curves, making the climate in their neighborhood more equable.

It will be noticed that some of the summer curves run almost all around Lakes Michigan and Erie, while the winter curves rise far up into the lower peninsula of Michigan. This shows why, as has long been known to agriculturists, wheat and corn can be as easily raised in the northern portion of this peninsula as two or three degrees of latitude to the south, in Central Illinois or Ohio.

Cleveland, which is about sixty miles south of Detroit, has nearly the same mean temperature, while Buffalo, which is about forty miles north of Detroit, has its mean temperature nearly three degrees less.

I hope at some future day to have fuller records of the temperature, and may then be able to correct some of the irregularities of the present

curves.

In Table IV will be found the continuation of record of daily evap oration for the past year, that for the winter months being computed from the temperature according to the method adopted in the last report As mentioned last year, these evaporators are usually placed on top of

the meteorological house, and are fully exposed to the sun. The temperature of the water is, therefore, greater than that of the lake surface, and the evaporation proportionately large.

From the few observations made in 1867, I estimated the evaporation from the lake surface to be sixty-four per cent. of that recorded by the evaporators. The more extended observations made this year show that that estimate was too large.

In Tables XXXIX, XL, and XLI of my report on the out-flow of the lakes for 1868 are given the simultaneous readings of an evaporator placed as usual at the stations and one in the river.

In the St. Lawrence the river evaporator was so protected that continuous readings could be taken, but at the other places the incessant passage of steamers made it very difficult to keep the evaporator from being disturbed by their waves, and on a great many days the readings were rendered worthless. But there are, I think, enough results to show that the river evaporation is about one-half that in the ordinary evaporators.

The temperature of the water in the rivers is generally higher than the lake surface, and, therefore, the evaporation should be greater, but the evaporation recorded at the river stations is for some reason greater than that recorded at the stations on the lakes, so that the excess will probably balance, and I have therefore taken fifty per cent. of the evaporation as recorded in Table V as the true evaporation.

TABLE IV.--Mean daily evaporation for each month in decimals of an inch.

SUPERIOR, WISCONSIN.

[For the months of January, February, March, April, November, and December the evaporation was computed from the mean temperature.]

Year. Jan. Feb. Mar. April. May. June. July. Aug. Sept. Oct. Nov.

Dec. Mean.

0.117 0.098 0.045 0.135 0.190 0.095 0.046

0.145

0.130

0. 159

0.135 0.083 0.049
0.135 0.134 0.052

0, 029

0.010 0.070

0.145
0.172

0.161

0. 143 0.036 0.107 0.151

0.099 0.090 0.068 0.031 0.015 0.183 0.159 0.082 0.058 0.032 0.012 0.138 0.135 0.071 0.078 0.030 0.017

0.066

0.074

0.061

0.151 0.138 0.093 0.056 0.031 0.016 0.069

ONTONAGON, MICHIGAN.

[For the months of January, February, March, April, May, November, and December the evaporation was computed from the mean temperature.]

[For the months of January, February, March, November, and December the evaporation was computed from the mean temperature.]

TABLE IV.-Mean daily evaporation, &c.-Continued.

[For the months of January, February, March, April, November, and December the evaporation was computed from the mean temperature.]

Year. Jan. Feb. Mar. April. May. June. July. Aug. Sept. Oct. Nov.

[For the months of January, February, March, and December the evaporation was computed from the mean temperature.]

[For the months of January, February, March, and December the evaporation was computed from the mean temperature.]

[For the months of January, February, March, November, and December the evaporation was com puted from the mean temperature.]

[For the months of January, February, March, and December the evaporation was computed from the mean temperature.]