172 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 30 



the distance between the two nuclei in a diatomic molecule, of the 

 moment of inertia of the molecule, of its heat of dissociation, as Avell 

 as the discovery of isotopic forms of the atoms and the calculation 

 of their masses are all possible in favorable cases where the band 

 spectra are suitable for precise measurement. To interpret the value 

 of each of these results, which have actually been obtained for oxy- 

 gen molecules, would lead us far into fields remote from our topic. 

 To apply improved methods of observing the infra-red region of 

 molecular spectra is bound to aid in developing this rapidly expand- 

 ing department of spectroscopy. 



We are thus brought to the third field of large-scale spectroscopy, 

 namely, the study of the solar spectrum. Here we pause to pay 

 respect to the genius of Professor Rowland of Johns Hopkins Uni- 

 versity, who carried out a remarkable investigation of the visible and 

 ultra-violet parts of the solar spectrum, which has been a storehouse 

 of information for 30 years. Some 20,000 spectral lines were meas- 

 ured and described with an accuracy only recently surpassed. There 

 is no doubt that Rowland would have extended his researches into 

 the infra-red part of the spectrum if he had possessed photographic 

 plates which would respond to such radiation. 



For years it has been comparatively easy to photograph much of 

 the ultra-violet and all of the visible region of the spectrum, and the 

 method of photographic spectroscopy has proved superior to all other 

 ways of determining the exact positions and characters of spectral 

 lines. As we go beyond the visible red, however, the photographic 

 process becomes increasingly difficult and finally has to be replaced 

 by the thermometric method. 



The thermometric method, as its name implies, depends on meas- 

 urement of the heating power of radiation. For this purpose three 

 types of instrument are in use, two of which utilize electrical effects 

 and the other, a mechanical effect of the radiation. Instruments of 

 the first two kinds are called bolometers and thermocouples, while 

 the third is known as a radiometer — and has been seen by every one 

 in the windows of opticians and jewelers. 



It would be interesting to describe the working of these marvelous 

 devices and to show the results which they produce. It must suffice 

 to consider briefly the application of the bolometer to the study of the 

 solar spectrum. This instrument was invented by Professor Langley 

 at the Allegheny Observatory in 1881. Subsequently as Secretary of 

 the Smithsonian Institution he used it with the assistance of Dr. C. 

 G. Abbot, the present secretary, to explore in detail the infra-red 

 solar spectrum. In the 30 years following the publication of this 

 great work, Doctor Abbot has continued to improve the instrument 

 and to apply it in a scries of brilliant researches. During the past 

 year, with tlie assistance of Mr. Freeman, Doctor Abbot has published 



