England. The difference between the average July and the average January temperatures seems to be in Lancashire only 180, which, compared with the apparent difference of 45° in Massachusetts, is small. There is found in Lancashire a set of conditions of temperature, prevailing winds and geographical surroundings, which is hard to believe can be duplicated. The most frequent wind is that from the south-west, and this, blowing from off the Gulf Stream, is naturally full of moisture. The relative and absolute humidities run even, as compared to New England, as will be seen from the following table of temperature readings observed in Stalybridge, England, for the month of July, 1887:

Water Vapor Maximum, 7.010 grains; minimum, 4.373 grains; mean, 5.581 grains. Prevailing south-west winds.

It can very readily be seen from this table why there is such a uniformity of moisture and relative humidity. We find that in this month the wind blew from the south-west twenty-two days. In order to fully cover the point under discussion, I will present the following table, showing the number of days in one thousand that the wind blows from each of the eight points of the compass in Lancashire :

Direction,
Days,.

N., NE., E., SE., S., SW., W., NW.
82, 111, 99, 81, 111, 225, 171, 120.

I will also present the following table, showing the number of days in nine hundred and sixty-five that the wind blew from each of the eight points of the compass in New England:

N., NE., E., SE., S., SW., W., NW.
62, 60, 105, 150, 200, 225.

Direction,
Days,.

91,

70,

It will be seen from these tables that the prevailing winds of England are south-west and west, while the prevailing winds of

New England are west and north-west; coming from a nonhumidity-producing region. This certainly has a very important bearing upon the great manufacturing centres of New England; as well as the fact that our temperatures and humidities are constantly fluctuating.

HYGRODEIK READINGS.

Wishing to compare temperature readings in different mills at the same time, I have been very kindly assisted by Messrs. WALTER E. PARKER of the Pacific, E. W. THOMAS of the Tremont and Suffolk, and J. W. KENT of the Wamsutta. These readings were carefully observed at the same hour of the day, except the Wamsutta, where the time of observation was a little later. The amount of absolute humidity has been worked out from signal-service tables, so that we can see at a glance what a cubic foot of air contains. The maximum, minimum and mean of each mill has also been worked out, so that we can see the loss and gain sustained.

We find from these temperature readings that the loss and gain of water vapor in the atmosphere was as follows: Pacific, 67.76 per cent.; Wamsutta, 47.88 per cent.; Suffolk, 48.17 per cent.; Tremont, 53.43 per cent.; Hamilton, 51.62 per cent. This certainly seems to be a very large per cent., but we must take into consideration the fact that the atmospheric changes were very marked. Take, for instance, July 8 in the Hamilton Mills. At the reading 2 P.M., the thermometers indicated 9.067 grains per cubic foot of air, and at 2 P.M., July 10, 4.527 grains, or a loss of nearly 50 per cent. of moisture. We will at this time make an estimate of the loss and gain of water vapor in the Tremont Mills. The highest point reached 9.067 grains: now, if we multiply the number of grains by the cubic feet of air (which we will call 300,000), and divide by 7.000 (the number of grains in a pound), we shall obtain 388 pounds of water. If we desire to carry this into gallons, we first ascertain what a gallon of water will weigh, which is 8.322 pounds. Now, if we divide 388 by 8.322, we shall get a product of 46 gallons of water contained in the room.

Again, we find that the lowest point reached was 4.223 grains per cubic foot. Now, if we multiply 4.223 by 300,000, and divide by 7.000, we shall get a product of 180 pounds of water. This 180, divided by 8.223, gives us 21 gallons of water.

TEMPERATURE AND HUMIDITY.

There are a good many mill men who are interested in the subject of temperature and humidity, who have not made a study of it sufficient to fully understand how to work out a problem and ascertain the relative humidity and amount of water vapor in a cubic foot of air. We often hear people speak of the temperature, but seldom of the relative humidity and dew-point. The dew-point is that temperature at which the air would become saturated when cooled, if no water were added to or taken from it. One way in which the relative humidity of the air is determined is to first find the dew-point of the air, and then by a simple division obtain the relative humidity. In order to ascertain the amount of water vapor in a cubic foot, and the per cent. of relative humidity in the air, a dry and wet bulb thermometer or hygrodeik is used. One bulb is covered with a wick or muslin, which connects it with a glass or cup full of rain or distilled water; the other bulb is left dry, and free to the air. I have found in my observations that the best position to place the instrument is some six feet from the floor, in the centre of the room. In order to work out the dew-point and humidity of the air, I will reproduce two valuable tables. The first is that of weights of saturated steam, compiled by Charles F. Porter. The second is the dewpoint and relative humidity table, compiled by the Signal Service, Washington, D. C. Scientists have concluded that the weight of water which a cubic foot of atmosphere can contain, and which it does contain when it is saturated, is the weight of a cubic foot of saturated steam of the temperature of that atmosphere; so that, if we know that the air is saturated, we can immediately know the weight of water in a cubic foot, by looking into one of the tables of the weight of water in a cubic foot of such steam at a temperature of the atmosphere.