June 20, 1913] 



SCIENCE 



953 



Ten times these values would be a reflection 

 of more than the whole and " a low estimate " ! 



To make my meaning entirely clear, let it 

 be noted that the moon occupies on the av- 

 erage about 1/97,300 part of the hemisphere 

 of the sky, and could send to the earth no 

 more than this fraction of sunlight if it had 

 an albedo of unity, or if it were a perfect 

 diffuse reflector. With an albedo of 17.4 per 

 cent., light from the full moon should be 

 0.174/97,300 = 1/559,300. This is the frac- 

 tion for which Coblentz adopts in round num- 

 bers 1/500,000. A smooth sphere having the 

 property of specular reflection would yield a 

 small star-like image of the sun of great bril- 

 liancy, the rest of the surface remaining dark. 

 Nothing of the sort occurs, and the reflection 

 is mainly diffuse; but the notable increase of 

 brightness at, or near, full moon, and the 

 somewhat greater brightness of the limb, as 

 compared with the center, signifies that there 

 is enough crystalline material in the rocky 

 surface of the moon, and especially upon the 

 cliffs which are presented favorably for ob- 

 servation along the lunar limb, to give an 

 appreciable percentage of scattered specular 

 reflections from innumerable crystalline facets. 

 The distribution of such reflection may dift'er 

 enough from that for a matte surface to ac- 

 count for the peculiarities of lunar reflection. 

 Infra-red rays, on the whole, are less reflected 

 than visible rays by the moon. 



Dr. Coblentz finds for his hypothetical 

 quartz moon an emissive ]X)wer of 0.1 for a 

 limited section of the spectrum near 9 fi,; 

 though his Fig. 5" founded on the observa- 

 tions of Rosenthal, makes this fraction nearer 

 0.25. We may remark in passing that there 

 are no common minerals with a relative emis- 

 sivity as low as 0.1, even if we confine atten- 

 tion to this limited region of the spectrum, 

 and that the most notable depression in the 

 lunar spectrum at this point is also shown in 

 the solar spectrum and is probably atmos- 

 pheric. I speak of a " depression " in the 

 Zollner, 0.1195; W. H. Pickering, 0.0909. Com- 

 pare Zijllner, ' ' Berechnung der waliren und schein- 

 baren Albedo des Mondes, " op. cit., pp. 161, 162. 



^'Op. cit., p. 317. 



emission-curve, but the depression is only 

 a minor feature in what is otherwise a 

 maximum. 



Describing his Fig. 7," Coblentz says: 

 In Pig. 7, curves b, c, d show several of Lang- 

 ley's observed lunar radiation curves, and as a 

 whole there is a close parallelism between the 

 theoretical and the observed curve, especially at 

 10.7 /i, where we have to consider only atmospheric 

 absorption." 



There is a mistake here. Curve d. Fig. 7, 

 is transferred from curve c. Fig. 6, which is 

 derived in turn from the hypothetical emis- 

 sion curve with superposed atmospheric ab- 

 sorption. The supposed " close parallelism " 

 vanishes when this mistake is corrected. The 

 principal feature of the lunar curves is that 

 they show a region of maximum radiation 

 between 8 and 10 /a (highest point at about 

 8.3^), where the hypothetical emission curve 

 has a minimum. The mistake is indeed cor- 

 rected a little farther on where we read (p. 

 318): 



The computed emission curve is the most intense 

 at 10.2 II, while the observed curve is the most 

 intense at 8.3 n. 



But here another error is introduced, for 

 we are informed that 



this is to be expected if the observed energy curve 

 is the composite of the selectively emitted energy 

 of the moon and the selectively reflected energy of 

 the sun. The selectively reflected energy of the 

 sun would, to a certain extent, fill up the minima 

 in the lunar emission curve, and as far as our 

 present knowledge goes would explain the observed 

 curves b, o, d [o?], (Fig. 7), which lack a minimum 

 at 8.5 II. As a whole, from whatever standpoint 

 we view this matter, we come to the same con- 

 clusion, viz.: that in the region from 8 to 10 |it the 

 energy emitted from the moon consists of its own 

 proper radiation and of reflected energy from the 

 sun. 



The explanation, unfortunately for this 

 writer, does not explain, since, as I shall show, 

 the reflected radiation can not possibly exceed 

 about 1/3,000 of the emitted. 



At this point in his argument, Dr. Coblentz 



" Op. cit., p. 319. 

 "Op. cit., p. 318. 



