182 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1913. 



fourth as great displacements as do the fainter ones. Now, it appears 

 from various lines of reasoning that the fainter lines should be the 

 ones that are formed at the greatest depths, so that St. John is able 

 to arrange the iron spectrum with reference to faintness and with 

 reference to velocity of outflow in sun spots in a series which very 

 probably indicates a progressive depth of sounding below the sur- 

 face of the sun. Then corresponding to this iron scale, if he takes 

 the lines of the other chemical elements, comparing them line by line 

 as regards velocity of outflow with the velocity shown by his iron 

 scale, he may arrange all the chemical elements in terms of the iron 

 scale, in the order of their depths of occurrence below the sun's 

 surface. 



In this way he finds, as is indeed indicated by other lines of re- 

 search, that the heavy chemical elements will lie the lowest, and vice 

 versa. Corresponding to this arrangement it is natural to find that 

 the lines of calcium, sodium, magnesium, and hydrogen indicate a 

 flow of greater and greater velocity in the opposite direction from 

 those of iron, so that these elements are arranged above the upper- 

 most level of the iron lines in a progress outward from the general 

 solar surface. Thus, as shown in figure 1, we may have the arrange- 

 ment of the vapors as they exist in the sun, from the hydrogen at 

 the highest level down to the elements like lead, lanthanum, barium, 

 at relatively low levels. Such elements as uranium (and radium, 

 if it exists in the sun) are so very high in atomic weight that they 

 lie very deep down in the sun and do not give solar-spectrum lines 

 at all, so that we shall probably not obtain direct proof of the ex- 

 istence of radium in the sun on account of the low level at which 

 it must lie if present there. We have, to be sure, long known of 

 the existence in the sun of helium, which is a product of the disin- 

 tegration of radium. This may, perhaps, indicate that the parent 

 substance, radium, is also present in the sun, but of this there is no 

 certainty. 



3. MEASUEEMENTS OF SOLAR RADIATION 



In the Smithsonian Report for 1912 the writer gave an illustrated 

 account of the investigations of solar radiation by the Astrophysical 

 Observatory of the Smithsonian Institution. In Jul}', 1913, the results 

 of this long investigation were published with details in volume 3 of 

 the Annals of the Astrophysical Observatory. The most important 

 conclusions are as follows: 



1. The mean value of the solar constant of radiation for the epoch 

 1905-1912 is 1.932 calories per square centimeter per minute. 



2. An increase of 0.07 calory per square centimeter per minute in 

 the " solar constant '' accompanies an increase of 100 sun-spot numbers. 



3. An irregular variation frequently ranging over O.OT calory 

 per square centimeter per minute within an interval of 10 days is 



