THE apparatus invented by Sir John Herschel for studying the calorific intensity of the solar radiation on the earth's surface is generally known as an actinometer. The causes which modify the above-mentioned intensity are essentially variable, but the principal ones are the position of the earth in its orbit, the zenith distance of the sun on which depends the thickness of the atmospheric stratum through which the rays pass, and the state of the atmosphere. The temperature of surrounding objects, of which the radiation may be positive or negative with regard to the body exposed to the solar rays and the losses of heat due to air currents, are also elements of disturbance which practically render impossible the appreciation of the sun's influence by ordinary thermometric observations. It is hardly necessary to remark that the old plan of comparing two thermometers, one in the sun, the other in the shade, conduces only to erroneous conclusions. Both Sir John Herschel and M. Pouillet conceived the idea of measuring the energy of the solar radiation by exposing to the sun's rays a given weight of water inclosed in an envelope having a certain determined surface. By noting the elevation of the temperature of the water at the end of a fixed time, and adding the losses due to the passage of the rays through the atmosphere and their dispersion, the above observers calculated appr oximately the total dynamic power developed by the sun. Herschel's actinometer consisted in a small open receptacle, attached to a fixed standard. Pouillet's pyrheliometer, which figures in nearly all treatises on physics, was of polished silver. The vessel exposed to the solar rays contained 100 grammes of water, was 100 millimeters in diameter, and 15 millimeters in thickness. It was covered on the exterior with lampblack. This instrument could not, however, be used in winter when the temperature fell below 32° Fall.—a serious defect, as the heat of the sun's rays is most intense near the solstice, owing to the diminution of the distance between the sun and the earth, and the limpidity of the atmosphere during cold clear weather. The loss of heat due to radiation in the pyrheliometer, the loss also owing to conducti bility, augmented by air currents, the absence of proper means of causing the liquid to circulate in the heating receptacle, the rudimentary method of orienting the instrument by the hand, and finally the influence of the observer's own radiations, and the disturbance due to his respiration, were all causes of error. To eliminate all these defects, and to obtain exact measurement of the intensity of solar heat, has been Captain John Ericsson's object in devising the solar calorimeter, represented in Fig. 1. The instrument is fixed on a movable table within a rotary observatory. It consists of a copper receptacle, filled with water, and covered with lampblack on the exterior exposed to the sun. A thermometer placed in the liquid indicates variations of temperature, and an agitator, moved by a belt passing over a little pulley, keeps the water in motion and insures its equable temperature. The water vessel is placed in the bottom of a chamber, the flaring mouth of which receives a plano-convex lens whereby the rays are concentrated upon the calorimeter. In this chamber as nearly a perfect vacuum as is possible is maintained, and a water jacket is provided so that the interior temperature may be constant. Fig. 2 represents an actinometer, also devised by Captain Ericsson, and intended to obtain the measure of the intensity of the solar radiation at a given moment. A hollow sphere of very thin copper is caused to rotate near the bottom of a cylindrical chamber, in which a vacuum exists as before. The sphere is blackened inside, and the sun's rays fall upon it through a thin plate of crystal which covers the upper part of the chamber. The lattei is jacketed, and the water in the annular space is kept in constant circulation. The sphere is rotated by a hand-wheel attached to its arbor, which last passes through a stuffing box in the side of the chamber. FIG. l.—ERICSSON'S CALORIMETER. The arbor is hollow, and a thermometer passes through; the bulb covered with lampblack being in the center of the copper sphere. The instrument, of course, rotates with the arbor. It will be seen from the engraving that the capacity of the reservoir is considerable relatively to that of the graduated tube; and hence the smallest variations of temperature are clearly and rapidly indicated. The instrument is supported by columns fixed to the table, which may be adjusted so that the axis of the chamber may point to the center of the solar disk. The water which circulates in the exterior envelope is kept at a constant temperature of 60° Fah. The point through which the thermometer passes is not tight, so that the air enters in the hollow sphere. The mode of establishing comparison between this instrument and the ordinary thermometer is as follows: “The convex surface of the square being equal to four great circles, and the surface occupied by the luminous beam being equal to one great circle, it is evident that the cooling surface of the sphere is quadruple the surface of the solar radiation in the experimental case. If the cylindrical chamber radiates freely toward the center, the temperature conserved by the sphere exposed to the cold radiation from all points will be only one-fourth of that which may be communicate"! by the solar beam. The chamber is kept at 60° Fah., and'the heat radiated by it toward the centre cannot exceed that limit. It may be demonstrated that the elevation of the temperature of the sphere represents only one-fourth of the calorific intensity of the rays which traverse the crystal plate. The temperature of the air inside the sphere is the same as that of the metal, and hence the thermometer within will show the degree with sufficient nearness. The indications of the thermometer then multiplied by 4 will give the true intensity of the solar rays at the surface of the earth, less the quantity of heat absorbed during passage through the crystal plate. The results of careful observations combine to show the exactness of t his conclusion. In Fig. 3, on next page, is represented Capt. Ericsson's new engine for the utilization of solar heat in the production of motive power. It is calculated that the heat radiated by the sun during nine hours per day, for all the latitudes comprised between the equator and the 45th parallel, corresponds per minute and per square foot of normal surface to the direction of the rays to 3'5 thermo units of 772 foot pounds. Hence, a surface of 100 square feet would give a power of 270,000 foot pounds, or from 8 to 9 horse-power. The engine illustrated is on the caloric system, and has run at 420 revolutions per minute with the sun near the zenith and during fine weather.