344 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 64 



entists that at the altitude of its perigee — approximately 220 kms. 

 above the earth — there was a good deal more atmospheric density than 

 had been anticipated. On November 6, the Observatory and the U.S. 

 Naval Research Laboratory jointly announced preliminary results 

 from the tracking of the Soviet satellite. Whereas pre-Sputnik esti- 

 mates had indicated a density of 10"^^ grams per cubic cm., analysis of 

 the orbit of Sputnik I now suggested a density of perhaps five times 

 that amount. These calculations had been made by Dr. Sterne, as- 

 sisted by Dr. J. S. Rinehart and Dr. G. F. Schilling. 



They had, then, the rather paradoxical situation that one of the 

 reasons for sending up a satellite was to determine atmospheric den- 

 sity, but that a fairly good estimate of the density was needed in order 

 to compute orbits and make predictions of satellite transits. 



A milestone in research and analysis of satellite data was reached in 

 May of 1958 at a meeting of the American Geophysical Union at the 

 National Academy of Sciences, Washington, D.C. There, some of 

 the results on Explorers I and III were summarized. Dr. Van 

 Allen presented his conclusions concerning the existence of a radia- 

 tion belt around the earth. Other scientists made preliminary esti- 

 mates of the concentration of meteors at the altitudes at which the 

 satellites were orbiting. And Drs. Schilling and Sterne offered a sum- 

 mary of tentative conclusions concerning the density of the upper 

 atmosphere as derived from satellite observations; table 3, which 

 appears in the Observatory's Special Report No. 12, dated April 30, 

 details their results. 



The authors noted that these estimates were made from observations 

 at different geographic latitudes and that the data were too few to pro- 

 vide an accurate mean value. They further cautioned that the esti- 

 mates were not strictly comparable because no allowances had been 

 made for seasonal, diurnal, and other sporadic variations of air den- 

 sity. These were now to become a major concern in the study of 

 atmospheric phenomena. 



Meanwhile, the Observatory had incorporated into its program for 

 the computation of orbits the changes of period caused by air drag. 

 They found, however, that they were still faced with rather serious 

 errors in predictions, although not nearly so bad as they had been 

 earlier. The problem was complicated by the fact that Sputnik II and 

 Explorer I were not spherical ; therefore, as their orientation changed 

 in space, the amount of surface against which atmospheric density 

 could act to decrease the altitude and increase the speed of the satellite 

 changed. 



Nevertheless, the variations in satellite drag from day to day did 

 not seem accountable by considerations of the presentation area. "When 

 the spherical satellite Vanguard I showed the same type of oscilla- 

 tions that had appeared in the orbits of Sputnik II and Explorer I, 



