Dreyer — On Astronomical Transit Observations. 



517 



Tabdxae Eekoes of E. A. 





Of the Sun. 



Of the 



Moon. 



I St Limb. 



2nd Limb. 



ist Limb. 



2nd Limb. 



Criswick — Duiikin, . 

 C.-ElHs, .... 

 C.-J, Carpenter, 



+ Os-052 

 + -103 

 + -150 



+ 0^-002 

 + -019 

 -0 -001 



+ s-034 

 + -112 

 + -132 



+ 0=-032 

 + -077 

 + -038 



There can hardly be any doubt that this personality principally 

 arises from the irradiation, as it is a well-known fact that the rliameter 

 of the sun, as well as that of the moon, is measured differently by 

 different observers, as also by different telescopes. But there is ano- 

 ther circumstance, the condition of the atmosphere and the quality of 

 the images (which depends thereon) whose great importance for obser- 

 vations of such very luminous objects must not be undervalued. The 

 observations of the sun made in Pulkowa by Wagner and Gylden 

 (especially by the former) show very distinctly that the apparent 

 diameter of the sun increases, as the images get worse,* and the circum- 

 stances under which the limbs of the sun and the moon are taken 

 differ therefore, in every respect, from those of star observations, so 

 that personality in the former cannot be treated from the same point 

 of view as in the latter. 



While the observer's individuality thus has a different influence in 

 observations of the different limbs of the sun and moon, we must 

 leave it to the future to decide whether observations of faint stars, 

 compared with those of brighter ones show a similar anomaly (that is, 

 a change in the common personal error) or not. It is impossible to 

 find anything respecting this question by examining any of the 

 published observations of personal equations, as these only extend 

 over three or four of the first classes of magnitude, where, no doubt, 

 a distinct variation of the personal errors never will be found. 



The matter is certainly of importance. If we suppose that an 

 observer's personal error is not subject to very sudden variations (as 

 we have seen is not the case with tolerably experienced observers), 

 it will for common astronomical observations be of no importance 

 whatever, how large his error is, if he only determines the correc- 

 tion of the clock himself. But the case becomes quite otherwise, if 

 his error is different for stars of a different magnitude. If he, for 

 instance, observes a zone of small stars of 8 to 9-10 mag., and uses 



* Vierteljalirssclu-ift der Astron. Gesellschaft, viii., pp. 48-55. 



