SIR DAVID GILL EDDINGTON. 515 



Juno, -which was then favorably situated. He considered that a 

 single observer could by heliometer observations of a minor planet 

 obtain results comparable in accuracy with those derived from all 

 the transit of Venus observations together. Unfortunately, the 

 heliometer Avas delayed in arrival at Mauritius, and the first' half 

 of the opposition of Juno was lost. Observations in the latter half 

 were secured on 12 evenings and 11 mornings, but the parallax factor 

 was then small. The result, 8''.77±0''.011, though disappointing 

 owing to the causes mentioned, gave a clear indication of the value 

 of the method, and this pioneer effort served its purpose as a pre- 

 liminary to a more ambitious attempt. From that time onwards 

 Gill had a strong conviction of the value of the heliometer for work 

 of the higliest refinement, and he acquired his remarkable skill in 

 using it. The transit of Venus was observed by the party, but Gill 

 appears to have formed so low an opinion of the trustworthiness of 

 the measures that he took little interest in their subsequent use. 



In 1877 an exceptionally favorable approach of Mars to the earth 

 offered a good opportunity for la renewed attack on the problem of 

 the solar parallax. Gill, who had resigned his position at Dun 

 Echt, began to prepare for an expedition to Ascension Island for 

 this purpose. He fully expected that Mars would, owing to its large 

 disk, prove to be a less satisfactory subject for heliometer observa- 

 tion than a minor planet, which is practically indistinguishable in 

 appearance from the comparison stars; but the parallax factor was 

 so much more favorable than for any minor planet then known that 

 the opportunity was not to be missed. His anticipations proved 

 correct. The value of the solar parallax now found showed a great 

 improvement on any previous determination. Tlie result, 8".78, 

 with a probable error of d=0".012, marks a new stage of advance. 

 But Gill by this work became more than ever convinced that the 

 definitive determination of the constant must rest on minor planets. 



For his third and final attempt, in 1888-9, the minor planets 

 Iris, Victoria, and Sappho were chosen. Instead of measuring the 

 diurnal parallax, he proceeded this time by the combination of ob- 

 servations made at widely separated stations. This involved a great 

 scheme of cooperation in which many observatories and individuals 

 took some part. The actual heliometer measures of the planets were 

 made mainly by Gill and Finlay at the Cape, by Elkin and Hall at 

 Yale, and by Peter at Leipzig. Of the many other cooperators Dr. 

 Auwers in particular took a large and important share in the work. 

 Accurate places of the comparison stars were needed, and meridian 

 observations of these were made at a large number of places. In 

 the case of Victoria this was supplemented by a heliometer triangu- 

 lation in order to a^oid the various systematic errors that affect 

 meridian observations. The whole discussion, which forms two 



