returns to its equilibrium shows the effect of the pressure upon it. The maximum rise in the water will occur when a strong easterly wind is blowing at the time of a low barometer, and the minimum when a westerly wind and a high barometer are simultaneous. These facts show that the atmospheric pressure on the water of the lakes is the same, though in a less degree, as that often seen on the coast of South America during the great fall of the barometer before a hurricane, and also why the fishermen along the coast can often predict the approach of storms by the rise or fall of the water. In order not to confuse the diagrams the temperature curves were omitted.

LAND AND LAKE BREEZE.

To ascertain whether we have upon the lakes a land and lake breeze similar to the land and sea breezes on the ocean, I have compiled from the hourly anemograph records at Milwaukee and Cleveland and from the half-hourly observations at Thunder Bay island tables B B. In these tables the number of days are given in which the wind blew from the different points of the compass during the several hours in the months of June, July and August, all stormy and cloudy days being first thrown out.

A summation of these tables for each place is given in tables C C, and the latter are graphically represented in the diagrams 22 and 25, according to the principle of indexed planes used by M. Koeuptz in his meteorology, in which the curves represent the outlines of planes passing through every fifth number in the tables, the larger number being taken as bills and the lesser as valleys.

Diagram 22, for Milwaukee for four years, shows a decided breeze off the lake from the east-southeast and southeast from 10 or 11 o'clock a. m. till 6 or 7 o'clock p. m.; from noon till 5 p. m. on nearly one-half of the days taken, the wind blew from these two points. This would be from the broadest part of the lake.

The land breeze is not as well marked, its effect being probably diminished by the hills back of Milwaukee.

In diagrams 23, for Cleveland, on the other hand, we have but little indication of a lake breeze, but have a land breeze from 9 o'clock in the evening till 9 o'clock in the morning.

The Cleveland anemograph being on a high tower upon the bluff overlooking the lake, the lake breeze rarely rises high enough to effect it, while it feels the full influence of all winds off land.

There is, however, but one year's observations here, as the anemograph broke down and has never been replaced.

Diagram 24, being very curiously banded, I reduced the number of points used to eight and replotted it in diagram 25.

This banding shows the difference between a careful observer and a machine, the observer having only recorded the wind as blowing from the eight principal points.

The latter diagram shows a breeze from the south and southeast, in which direction the main body of the lake lies, and from the north from 10 a. m. to 10 p. m., and from the north and northwest from midnight to 8 a. m.

The island lying a distance from land, and having a large body of water around it from north to south by east, the lake breeze assumes a rotary character opposed to the apparent motion of the sun, and the land breeze much weakened in its effect. This was also shown in diagram 1 to be true for all the seasons as well as for the summer months we are now investigating. The island is not large enough to have a land breeze of its own, and only received a part of that from the mainland.

Whether this lake breeze blows from the centre of the lake towards each shore, is a question we have not the means at present of investigating, but hope to have at some future time.

The last lines in the tables give the number of days taken in each month and season and the number of hours of calm.

In tables C C it will be seen that the calmest hours were from 10 o'clock at night to 4 o'clock in the morning at Milwaukee, and at Thunder Bay island from 8 o'clock in the evening till midnight, while at Cleveland the most windy hours were from 10 at night to 3 in the morning, which is a confirmation of the supposition that the day breeze does not reach the elevation of the observatory.

The records of the levels of the water at the several stations have not beeu reduced since 1863, nor have I been able to give enough attention to them to report upon them this year.

I hope another year to be able to bring them up.
Very respectfully, your obedient servant,

General W. F. RAYNOLDS,

D F. HENRY, Assistant Lake Survey.

Corps of Engineers, Sup't Lake Survey, Detroit.

Reductions of the monthly meteorological register of the survey of

the northern and northwestern lakes at Superior City, Wisconsin.

NE. NE
NW NW
NW.. NW
Calm
Calm
NW.

63

.12

Reductions of the monthly meteorological register of the survey of

Dry bulb, or tem