136 



RECORD OP' SCIENCE FOR 1887 AND 1888. 



a Bootis and ex Auriga', <lo not eoufirin the large values found by 

 O. Struve, -fO". 51(5 and 4-0".305; but in the ease of Ibe former star 

 tbere is a close agreement with Hall, who got +0".102 ±0".030, and 

 there seems to be but little doubt that the Pulkowa value is largely 

 in error. 



Of the four stars where the parallactic displacement has been inap- 

 preciable, Arcturus, with its large proper motion of over 2", second 

 only to that of a Centauri in all of the 200 brightest stars down to the 

 fourth magnitude, is especially note-worthy. The minuteness of the 

 parallax is beyond doubt, depending, as it does, on five pairs of com- 

 parison stars, all in reasonable agreement, and it can not be considered 

 as seriousl}^ at variance with the results previously obtained by Peters 

 and Johnson, -f0".127 .4:0".073 and -f0".138 ±0".052 respectively, 

 when their liability to systematic error is taken into account. The Yale 

 result for ex Lyrte does not fall in well with those hitherto deduced for 

 this star. If we commence at the epoch of W. Struve and neglect the 

 earlier attempts to find the absolute parallax, we have the following list 

 of values: _ 



from whicli a parallax of about -fO'MT would seem well assured. The 

 pair of comparison stars used by Elkin are very symmetrical, and so 

 large a value would seem incompatible with the heliometer measures. 



ritotograpMc (Ictermination of stellar ^xcraUax. — Professor Pritchard, 

 of the Oxford University Observatory, pursuing his experimental work 

 upon the determination of stellar parallax by the help of photography, 

 has ])ub]ished the following interesting results. The last three are pro- 

 visional, having been determined from only six mouths' observations: 



<;i' Cygui 



(>V^ Cygui . ... 



// Cassiopeai 

 Polaris ... 



a Cassiopea^ 



/J Ca.s>siopea'. 



y Cassiope;e 



Parallax. 



< 



-fO. 4289 

 . 4353 

 . 035f) 

 . 052 

 .072 

 .187 

 050 



Probable 

 error. 



J-O. 0180 

 . 0152 

 . 0250 

 . 0314 

 .042 

 , 039 

 .047 



