354 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 63 



another set. Eventually, one orbit determined by this means fairly 

 well matched their estimate, and it was this that was distributed to 

 the scientific community and to the press. 



Thereafter, the staff processed individually each of the observa- 

 tions, most of which were naked-eye or Moon watch sightings made in 

 the United States ; no photographic observations were available dur- 

 ing the first 2 weeks. The Observatory was primarily interested, at 

 this point, in making predictions of transits over the United States. 

 Many data were required to eliminate errors. 



On a large map of the United States the computers marked for 

 each sighting a spot to indicate where the observer was. Ideally, his 

 observation would have: (1) The time at which he made it; (2) the 

 azimuth, or the direction along the horizon; and (3) the elevation 

 above the horizon. Observations in a different form, in which data 

 were given with respect to the star background, had to be reduced to 

 readings of azimuth and elevation. 



From the position of the observer, the staff would draw a line in 

 the direction in which he saw the satellite, w^hich was the azimuth 

 that he observed. They computed from the orbital period the height 

 of the satellite above the surface of the earth. 



If the orbit is perfectly circular, then its height above the earth 

 is essentially independent of its position in the orbit. One takes the 

 elevation above the horizon at which the satellite was observed, and 

 combines this datum with the height estimated from the orbital period. 

 One can then calculate by simple trigonometry the distance of the 

 satellite from the observer, and on this azimuth line, mark a point for 

 the estimated position of the satellite. 



This was done for one evening's observations. Say the satellite 

 passed over New England. There would be perhaps half a dozen 

 observations. From each of these one derived a point representing 

 the position of the satellite projected onto the surface of the earth 

 at the time of the observation. There resulted half a dozen points, 

 more or less on a line. A straight line was drawn among these points 

 as well as possible. By noticing how the points fell in relation to the 

 line, one could go back and correct the estimate of the height of the 

 satellite and obtain a more consistant analysis of these particular 

 observations. The line on the surface of the earth represented a 

 trajectory of the satellite for that evening's pass. Then, from a 

 similar set of data for the following evening, one plotted another 

 line on the surface of the earth, representing the passage of the sat- 

 ellite for that evening. At this point, there was enough information 

 to compute with fair accuracy the period of the satellite — essentially 

 its velocity — and find the position of the line on the surface of the 

 earth for the following evening just by extrapolating the data. The 

 Observatory staff did this partly by using a theory that predicted the 



