624 



SCIENCE 



[N. S. Vol. XXV. Xo. 642 



From various considerations one would ex- 

 pect to find the bands to shift to the long 

 wave-lengths with increase in molecular 

 weight. It was known that certain groups of 

 atoms cause certain absorption bands, but no 

 shift in the maximum of the band could be 

 detected when the number of groups of atoms 

 was increased in the molecule. The contra- 

 diction lies in the failure to make a distinc- 

 tion between the effect of the groups of atoms 

 which is to cause the absorption and the reflec- 

 tion bands, and the effect of joining these 

 groups of atoms to various elements (different 

 atomic weight) which have now been found 

 to determine the position of the bands. By 

 studying the transmission and reflection spec- 

 tra of a homologous series of compounds it 

 was found that the position of the character- 

 istic band shifts toward the long wave-lengths, 

 with increase in the molecular weight of the 

 metallic element to which the group of atoms 

 is united to form the compound. These bands 

 lie in the region of 4.6 /* and 6.5 ix (trans- 

 mission) and 8.7 /A to 9.1 /x (reflection). 



The silicates are exceptions to all the rules, 

 for there seems to be no regularity in the 

 reflection bands, indicating that the grouping 

 of the atoms of oxygen and of silicon is dif- 

 ferent in the different minerals studied. 



When energy is reflected from a plane 

 smooth surface it is commonly called ' reg- 

 ular ' (or less accurately ' specular ') reflec- 

 tion, while energy reflected from a rough sur- 

 face suffers ' diffuse ' reflection. The reflect- 

 ing power, B, of any substance is related to its 

 index of refraction, n, and its absorption co- 

 efficient, k, by the equation: 



S. 



(n — iy + h* 



For ' transparent media ' or * insulators ' the 

 absorption coefficient is almost zero and the 

 reflection power is a function of only the re- 

 fractive index. Here the reflecting power is 

 low, only 4 to 6 per cent., and decreases with 

 increase in wave-length. For metals, ' elec- 

 trical conductors,' the absorption coefficient 

 has become so large that nearly all the energy, 

 for all wave-lengths, is reflected (90 to 98 

 per cent.). 



For substances having selective absorption, 

 when the coefficient of absorption, k, attains 

 the high values the heat or light waves no 

 longer enter the substance, but are almost 

 totally reflected as in the case of metals— 

 whence the name, ' bands of metallic reflec- 

 tion.' If, then, the eye were sensitive to heat 

 waves, many substances would have a ' sur- 

 face color ' similar to that of gold and f ucli- 

 sine in the visible spectrum. Furthermore,, 

 substances having selective absorption (and 

 reflection) will have a low reflecting power in 

 the region where the absorption coefficient, k, 

 is small. In other words, the reflecting power 

 of plane surfaces (' regular reflection ') of 

 substances (' transparent media,' ' electrical 

 insulators ') having selective absorption, will 

 be low for all regions except where there are 

 bands of ' metallic reflection.' It is evident 

 that a rough surface of the same material will 

 behave similarly, i. e., it will likewise be 

 selectively reflecting. 



The speaker found that the silicates have a 

 low reflecting power, ' practically zero,' for 'the 

 region of the spectrum up to 8 ju, followed by 

 bands of metallic reflection from 8.5 //, to 10 /*. 



It was pointed out from the curves exliibited 

 that a surface such as the earth or the moon 

 if composed of silicates will have bands of 

 strong selective reflection. If, then, one were 

 to examine the heat spectrum of a planet 

 which shines by reflected light, and if its sur- 

 face is composed of silicates, one would expect 

 to find bands due to selective reflection. By 

 comparing these bands with those of known 

 substances, the composition might be deter- 

 mined. In the case of the moon this is prac- 

 tically impossible on account of the weakness 

 of the radiation to be measured. Atmospheric- 

 absorption, and the fact that, in the case of 

 the moon, the maximum of its proper radia- 

 tion lies in the region of the reflection bands 

 of the silicates, will interfere with the ob- 

 servations. But there is still another compli- 

 cation in that the lunar radiation curve can 

 not be smooth and continuous (as some- 

 writers seem to think) if the surface is com- 

 posed of silicates, because in the regions of 

 selective reflection the emitted energy will be- 



