2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IIO 



for this work and many comparisons were made against various silver- 

 disk instruments.^ This campaign established a standard scale of 

 radiation which we called the "Smithsonian revised pyrheliometry of 

 1913." The constants of all silver-disk pyrheliometers have been 

 based on this scale. In the years 191 5, 1916, and 1920 further com- 

 parisons were made against standard water-flow No. 3 on Mount 

 Wilson, Calif., these results confirming the adopted scale of 1913.^ 



In 1932 a marked improvement was made in the standard water- 

 flow pyrheliometer. This was suggested by V. M. Shulgin * and con- 

 sisted in the substitution of two identical absorbing chambers instead 

 of one. The advantages of this change and others of a minor nature 

 are discussed in our paper "An Improved Water-flow Pyrheliometer 

 and the Standard Scale of Solar Radiation" (Smithsonian Misc. 

 Coll., vol. 87, No. 15, 1932). The new pyrheliometer proved free 

 from the worst difficulty we had experienced with the earlier instru- 

 ment, namely, irregular drift of the galvanometer zero. The results 

 now obtained were more concordant and more satisfactory than ever 

 before. Thirty-seven comparisons between the new water-flow No. 5 

 and our silver-disk pyrheliometer S.I. No. 5bis showed the scale of 

 our revised pyrheliometry of 1913 to be too high by 2.5 percent. In 

 1934 we repeated this work on Mount Wilson.^ Forty-two compari- 

 sons showed the 191 3 scale to be 2.3 percent too high. 



Since 1934, 13 years have elapsed with no further comparisons 

 against a water-flow standard. In August 1947 opportunity came to 

 make further comparisons at Mount Wilson. In preparation for this, 

 standard water-flow pyrheliometer No. 5 was altered as follows: 

 New thermoelements of copper-constantan were substituted for the 

 former nickel-platinum junctions. These and also the special glass 

 housings for the thermoelements were made by L. B. Clark of this 

 Institution in such form that the whole assembly could be waxed in 

 place without the use of rubber tubing. On arrival at Mount Wilson, 

 however, it was found that seams had opened up in the wax, owing 

 probably to changes in temperature and jolting in transit from Wash- 

 ington. After considerable difficulty the wax was remelted and the 

 whole made watertight. 



All the precautions which we took in 1932 and 1934 to insure 

 greater accuracy were again taken in the present comparisons. In 

 addition, the following steps were taken : 



2 Ann. Astrophys. Obs., vol. 3, pp. 52-72, 1913. 

 2 Ann. Astrophys. Obs., vol. 4, pp. 92-97, 1922. 



4 Monthly Weather Rev., August 1927, p. 361. 



5 Smithsonian Misc. Coll., vol. 92, No. 13, 1934. 



