92 ANNUAL, EEPORT SMITHSONIAN INSTITUTION^ 1920. 



observations indicate. We have now only to suppose that there 

 are inequalities of radiation in different directions which may be due 

 to the absorption or scattering of the rays in the coronal regions 

 near the sun. These inequalities may persist with little alteration 

 for weeks. We, however, note them as variations of solar radia- 

 tion as they sweep by us in the course of the sun's rotation on its 

 axis. 



The honeycomb jyyranoTneter. — Mr. Aldrich constructed two copies 

 of a new instrument devised by Abbot and Aldrich for measuring 

 " nocturnal radiation." We call it provisionally the " honeycomb 

 pyranometer." In this instrument a long thin ribbon of " therlo " re- 

 sistance metal about one-half inch wide and one one-thousandth of an 

 inch thick is bent in such a way as to make up into 200 cells of trian- 

 gular cross section all included in a total cross-sectional area of about 

 1 inch square. The corners of the cells are electrically insulated with 

 baked shellac so that a current of electricity can be caused to flow 

 from end to end of the ribbon and thus all around each cell. Radia- 

 tion which enters the front of the cells from any source, if not ab- 

 sorbed there is reflected to and fro within the cells till it reaches their 

 rear ends. There its remnant emerges upon a silvered mirror in- 

 clined at a small angle so as to throw back the rays to make a second 

 course to and fro toward the front. Thus by repeated absorptions 

 the rays are at length almost wholly converted into heat. The de- 

 vice is, in short, a " black body." But unlike other " black-body " re- 

 ceivers, its central cells are protected from losses of heat to the sides 

 by reason of the nearly equally warmed cells surrounding them. 

 Thus the instrument is almost as sensitive as a flat blackened strip, 

 but possesses the valuable property of being fully absorbing, which 

 a strip does not. The temperature difference between the central 

 cells and the case of the instrument is indicated by thermoelectric ele- 

 ments. By passing a proper electric current through the "therlo" 

 ribbon the same temperature difference can be produced as by radia- 

 tion. The known energy of the electric current becomes the desired 

 measure of the energy of radiation, as in Angstrom's pyrheliometer. 

 Also the constant of the apparatus is calculable from the known di- 

 mensions of it. It is possible, too, to observe the solar radiation with 

 this instrument, and so to calibrate it. Measurements of this kind 

 check very closely with the computed values. 



Messrs. Aldrich and Abbot made a series of measurements with 

 the honeycomb pyranometer on various sources of radiation, includ- 

 ing comparisons with the ordinary pyranometer on incandescent 

 lamps of different kinds, and also observations on large hollow radia- 

 tors at different constant temperatures. Values of the constant of 

 the fourth power law of radiation differing by only 1 per cent from 

 the best accepted value were readily obtained in this latter work. 



