s 



SCIENCE. 



[Vol. VII. , No. 152 



LETTERS TO THE EDITOR. 



♦** Correspondents are requested to be as brief as possible. The 

 u-riter's name is in all cases required as proof of good faith. 



The temperature of the moon. 



The interesting article by Mr. Ferrel in your issue 

 of Dec. 18 seems to require some words of comment 

 for the general reader, who may not otherwise notice 

 that the whole reflected heat (and light) of the moon 

 appears to be there omitted from consideration. 

 Moonlight is the ocular evidence that a part, at least, 

 of the moon's heat is lost by reflection, since what is 

 light to the eye is heat to the thermoscope. In fact, 

 what we see is but about one-third of what is re- 

 flected ; and all of this must be subtracted from what 

 the sun sends the moon before we have as a re- 

 mainder the amount radiated from its surface, which 

 is treated as the whole sum in the article under dis- 

 cussion. 



It is there assumed that the moon loses heat by 

 radiation only : but, even in this hvpothesis. the 

 highest temperature assigned to its sunlit surface is but 

 little above that of boiling water. Since, then, bodies 

 only begin to be visible by radiation at a red heat, it 

 follows from this hypothesis that the full moon 

 would always be black and iu visible, — an imagi- 

 nary moon, and not the moon which we see. Mr. 

 Ferrel is doubtless aware of this, and in his own 

 view may be supposed to be purely treating of a 

 hypothetical body ; but the ordinary reader is cer- 

 tainly apt not to understand the strictly limited prem- 

 ises with which he starts. 



In the exactness and competence of Mr. Ferrel's 

 mathematical treatment of this or any subject he 

 presents, all will agree ; but the more exact the 

 logical instrument, the more certain it is to deduce 

 limited conclusions from limited premises. 



Without entering on any discussion of Mr. Ferrel's 

 use of Dulong and Petit's formula, I may then say 

 that to those astronomers and physicists who are en- 

 gaged in the task of experimentally determining the 

 actual temperature of the lunar surface, the exist- 

 ence of this great amount of reflected heat is an 

 enormous difficulty, for it is not until this has been 

 differentiated from the radiated heat that the tem- 

 perature of the actual surface is either theoretically 

 or experimentally ascertainable. 



To the present Earl of Rosse belongs the credit of 

 making the first attempt to do this, and, in doing 

 so, to conquer those experimental difficulties which 

 lie even at the threshold, and which alone are exces- 

 sive ; for the total amount of all the heat of both 

 kinds is so minute as not to change the reading of a 

 thermometer directly exposed to the rays of the full 

 moon by nearly so much as the thousandth part of a 

 centigrade degree. 



The writer has now been engaged for a long time 

 in these researches, whose interest and importance 

 to us are not to be measured by the minute amount 

 of the heat in question. 



To prevent mistake, let it be stated that there never 

 has at any time been any doubt but that the lunar sur- 

 face radiates heat toward us, and there is scarcely 

 a doubt but that this radiated heat is greater than 

 the reflected. The question is, however, as to the 

 amounts, and as to whether the first kind passes 

 through our atmosphere as well as the second. 



This i.s not the place to discuss this somewhat 

 recondite point ; but the as yet unpublished Alle- 



gheny observations, now conducted through over 

 twenty lunations with the object of discriminating 

 the reflected from the radiated heat by the formation 

 of a lunar heat spectrum, show that a considerable 

 part of this radiated heat does pass through our 

 atmosphere along with that reflected. While the 

 writer differs from the conclusions of Lord Rosse as 

 to the temperature of the lunar surface, it seems due 

 to truth to say, that, in the particular just alluded to, 

 the interpretation of Lord Rosse is sustained more 

 fully than his own first one. 



Without anticipating the publication of these ex- 

 periments, the reader may care to learn of one obser- 

 vation made on the rare occasion when the full moon 

 is partially dark ; that is, during an eclipse. In the 

 lunar eclipse of Sept. 23, 1885, about eight-tenths of 

 the moon's diameter was covered by the umbra. The 

 night was beautifully clear at Allegheny, and obser- 

 vations were made with the bolometer on different 

 parts of the lunar image formed by a concave mirror 

 of twelve inches aperture, and ten feet three inches 

 in focal length, which was kindly loaned for the 

 occasion by Mr. J. A. Brashear. The image was a 

 little over an inch in diameter, and the bolometer 

 was limited by a diaphragm to an aperture of about 

 three-tenths of an inch ; so that any circular portion 

 of the moon's surface forming about one-eleventh of 

 the whole could be examined independently of the 

 rest. Previous observers have been obliged to utilize 

 all the lunar rays from a large concave mirror in 

 forming a very small image barely covering the 

 thermopile employed ; but, owing to the superior 

 delicacy of the bolometer, it has thus become pos- 

 sible to select small portions of a comparatively large 

 lunar image for separate study, and still have heat 

 enough for accurate measurement. 



Before the eclipse began, the exposure of the bo- 

 lometer to the central portion of the image produced 

 a galvanometer deflection of one hundred and eighty 

 divisions. The deflection on the east limb of the 

 moon was one hurjdred and sixty-four divisions : but, 

 as the eclipse advanced, the deflections here fell off 

 very rapidly, the diminution being noticed before 

 the penumbral shade became certainly visible to the 

 eye. The diminution of the effect on the centre and 

 west limb followed that on the east limb in time, as 

 these regions were progressively covered by the 

 shadow. On portions covered by the umbra the de- 

 flection was very small, varying from four divisions 

 soon after the beginning of immersion, to scarcely 

 more than a single division of the galvanometer 

 scale shortly before emersion from the umbra ; so 

 that the deflection was with difficulty detected. This 

 last minute effect might have been due to true radia- 

 tion from the darkened lunar surface, or possibly to 

 diffuse and irregularly reflected heat from the surface 

 of the mirror, though the method of exposure was 

 calculated to eliminate this source of error as far 

 as possible, — a doubt which must be resolved by 

 future experiment. 



As the middle of the eclipse approached, measures 

 made just outside the edge of the umbra indicated an 

 increasing transmissibility by glass for the feeble 

 radiant energy remaining. Thirty minutes before 

 the middle of the eclipse, the transmission by glass 

 for the lunar heat rays at this inner edge of the pe- 

 numbra was found to be thirty-two per cent, and 

 fifty-five minutes later it had increased to forty-eight 

 per cent. Although these latter deflections were 

 very small, the observations were apparently trust 



