on the presentation of a Medal. 1 39 



I have explained in the case of 61 Cygni. The curves so projected 

 have been subjected to your inspection, and that inspection certainly 

 does leave a very strong impression of a real and tolerably well- as- 

 certained parallax having been detected in this star. But at the 

 same time an impression no less decided, owing to irregularities in 

 the march of the curve, when compared with the true parallactic 

 curve, is created, — that the errors of observation are far from being 

 eliminated, — that, on the contrary, they bear such a proportion to 

 the parallax itself as to leave room for some degree of hesitation, 

 and to justify an appeal to a longer series of observations, and to con- 

 current evidence from other quarters, before declaring any positive 

 opinion. The evidence of this kind, in short, is not equal to that af- 

 forded by the similar projection of Bessel's observations of either of his 

 two comparison stars. And to this it must be added, that only one 

 star of comparison existing in the line of a Lyrce, the possible effect 

 of temperature and annual instrumental variation is not eliminated 

 from the result in the way in which it is from the measures of 61 

 Cygni ; while all that great mutual support which the observations of 

 parallaxes of the two comparison stars afford each other in the latter 

 case, is altogether wanting in the former. These considerations, 

 without any under- estimation of the great importance and value of 

 M. Struve's researches yet formed essential drawbacks on the imme- 

 diate admission of his results. 



In a word, I conceive the question of discovery as between these 

 illustrious, but most generous and amicable rivals, may be thus fairly 

 stated. M. Struve's meridian observations in 1819-1821 seem to 

 have made the first impression on the general problem, but too 

 slight to authorize more than a hope that it would yield at no 

 distant day. His micrometric measures of a Lyrce commenced more 

 than a year earlier, and have extended altogether over a longer 

 period than M. Bessel's of 61 Cygni. From their commencement 

 they afford indications of parallax, and these indications accumulating 

 with time have amounted to a high degree of probability, and ren- 

 dered the supposition of parallax more admissible than that of instru- 

 mental or casual errors producing the same influence on the measures. 

 On the other hand, M. Bessel's measures commencing a year later, and 

 continued on the whole through somewhat less time, have exhibited 



