That the temperature on the surface of Mars is comparable with that of the earth was long ago inferred from the behaviour of the polar caps, under the supposition that the latter are composed of ice or snow; and this inference, whatever the validity of the premise from which it was drawn, has been corroborated by some very remarkable analyses of the planet's thermal radiation that have lately been made. The determination of the temperature of a planet through measurement of the heat it sends us presupposes a refinement of technique which, until very recently, seemed quite unattainable, for small as is the amount of heat received, it consists of two parts, that which came originally from the sun and is reflected by the planet, and that which origi nates in the planet's own warmth. The second of these parts, called the planet's intrinsic radiation, is alone the index of the planetary temperature, and in order to be utilized as such it must either be isolated from the first part, or sufficiently separated from it to permit certain characteristics of thermal radiation to be recog nized. The most practicable procedure that has until now been developed is to compare the total heat received from the planet with that part of it which falls within certain limited regions of the spectrum. The spectral limitations are secured by passing the beam through various substances, notably water, of known trans missive properties. The logic of the method will be suggested by the consideration that, since the bulk of the intrinsic radiation of a body well below the temperature of incandescence, such as a planet, lies far in the infra-red, a great falling off in the amount Of heat received when this region is excluded by the
for example, of a water cell, will indicate a substantial contribu tion by the planet itself, and therefore a relatively high tempera ture. The actual analysis, while it is not in its present approximate form very complex, involves many factors, and cannot be pre sented here. Of surpassing interest is the delicate electrical thermometer with which the heat measurements are made. The sensitive element is a thermocouple of such small dimensions that the surface of the planet can be explored and the radiation of its several parts measured. These first successful measurements of planetary radiation from Mars were made in 1922 by W. W. Coblentz and C. 0. Lampland at the Lowell Observatory, but those undertaken at the Martian apparitions of 1924 and 1926 by these observers, and by E. Pettit and S. B. Nicholson of Mount Wilson, are of such superior quality that only they will be con sidered. Temperatures of several parts of the disc, depending on these measurements, are given in the table to the nearest 10° :— Considering the difficulty of the observations, the independent estimates by the Lowell and Mount Wilson observers are remark ably accordant, especially when it is remembered that the quantity determined is the absolute temperature (or temperature measured from the absolute zero point, —273° C) and not the difference between this and the comparatively close number which has arbitrarily been chosen as the zero of the Centigrade scale. The fall in temperature from the centre to the edge of the disc is beautifully shown ; and the polar temperatures as well as those at sunrise and sunset seem painfully low to dwellers on the earth. Two outstanding points of difference are presented by the esti mates made at the two observatories. The Lowell observers find the Martian afternoons to be warmer than the mornings, as is indi cated by higher temperatures at sunset than at sunrise. The Mount Wilson results on the other hand show equal temperatures for morning and afternoon. Our terrestrial experience is that it is usually warmer in the afternoon than in the morning, a fact which is due to the considerable water vapour content of the earth's atmosphere, and the Mount Wilson observers explain the sym metry of their morning and afternoon temperature curve to be a consequence of the known scarcity of water vapour in the atmosphere of Mars. The second discrepancy relates to the temperature of the south polar cap, which alone was visible at these oppositions. The Mount Wilson observations place the temperature at — 70° C, those made at Lowell indicate approxi mately o° C, the temperature of melting ice. This is a matter that bears critically on the question of the constitution of the polar caps. An interesting indication of the Lowell observations is that the dark parts of the planet are warmer than the bright ones, the estimated difference being 10° or 55° C, a fact which would appear to be better in accord with the supposition that they are merely areas of comparatively high optical absorption than that they are regions of vegetation. Optical absorption by a dark sur
face results in a rise of temperature, while expanses covered with vegetation are cooler than desert areas under like conditions.
The estimates of temperature provided by these observations have been generally accepted with confidence, and there seems to be no ground for questioning their substantial accuracy in so far as the central part of the planet's disc is concerned. The cal culations through which they were derived do not, however, take into account the properties of the Martian atmosphere, though the possible effects of an atmosphere were made the subject of con jecture by both groups of investigators. Whatever the magnitude of these effects, they must of necessity increase as the edge of the planet is approached, because of the greater amount of atmos phere intervening between us and the planet's surface at points remote from the centre. Furthermore the optical properties of the atmosphere are quite certainly affected by the lower tem perature of the terminator, which at the most favourable time for general observation is close to the edge. That the atmos phere of Mars is not altogether negligible seems a fair inference from photographic evidence already discussed, but we are not in possession of sufficient knowledge of its optical properties to permit a reliable estimate to be made of its effect on the planet's radiation. For such reasons the temperatures derived for the polar caps and other parts of the edge are not regarded with the same confidence as those obtained for the centre. With these and other considerations in mind, Messrs. Coblentz, Lampland and Menzel have summed up their conclusions in the following words : "We can say with some assurance that our measurements and deductions indicate that the temperature of the surface of Mars, under a noonday sun, rises considerably above the freezing point of water-a question that had been under discussion for years" (Publ. Astron. Soc. Pacific, vol. 39, p. ioo; 1927).
the opposition of Mars which oc curred in August 1877 the planet was unusually near the earth. Asaph Hall, then in charge of the 26" telescope at the Naval Observatory in Washington, took advantage of this favourable circumstance to make a careful search for a visible satellite of the planet, and on the night of Aug. i I found a faint object near the planet. Cloudy weather intervened, and the object was not again seen until the i6th, when it was found to be moving with the planet, leaving no doubt as to its being a satellite. On the night following an inner satellite much nearer the planet was observed. This discovery, apart from its intrinsic interest, is also noteworthy as the first of a series of discoveries of satellites of the outer planets. Hall named the outer satellite Deimos and the inner Phobos, from the horses that drew the chariot of the god Mars. A remarkable feature of the orbit of Phobos is that it is so near the planet as to perform a revolution in less than one third that of the diurnal rotation of Mars. The result is that to an inhabitant of Mars this satellite would rise in the west and set in the east, making two apparent diurnal revolutions every day. The period of Deimos is only six hours greater than that of a Martian day; consequently its apparent motion around the planet would be so slow that more than two days elapse between rising and setting, and again between setting and rising. Lowell estimates the diameter of Deimos to be about io m. and that of Phobos slightly more.
Long and careful series of observations have been made upon these bodies by other observers. At the very favourable oppo sitions of 1892 and 1894, observations were made by Hermann Struve at Poulkova, who later subjected all the data up to 1909 to a very careful discussion (see bibliography). He showed that the inclination of the planes of the orbits to the equator of the planet is quite small, thus making it certain that these two planes can never wander far from each other. The relations of the several planes can be best conceived by considering the points at which lines perpendicular to them, or their poles, meet the celestial sphere. By theory, the pole of the orbital plane of each satellite revolves round the pole of a certain fixed plane, differ ing less from the plane of the equator of Mars the nearer the satellite is to Mars. The observations of Lowell and Slipher, and of Trumpler, agree in placing the axis of Mars in position R.A. 316° .o; Dec. +54° .5 (Equinox of 1925). The tilt of the Martian equator to the Martian ecliptic calculated from this position is 23°.4.