373 



He next compares the refraction above deduced, with the results 

 of other astronomers. Piazzi, having an instrument which turns in 

 azimuth, has deduced the actual refractions at all distances from 

 the zenith, by means of numerous observations on Procyon, a Lyrae, 

 and Aldebaran, at various altitudes, from 38 to 89 zenith distance, 

 in addition to several circumpolar stars. Piazzi's result is, that the 

 mean refraction at 45 is 57"'3, which is less by eight tenths of a 

 second than that of the author ; but by the present French tables it 

 is stated to be 58"'2, which, on the contrary, is rather greater. But 

 beside the difference in the quantity of mean refraction at 45, Piazzi 

 observes that the law assigned by Bradley does not obtain ; for though 

 the actual refractions, so far as 80 from the zenith, are, in fact, greater 

 than was supposed by Bradley, the refractions within the remaining 

 10 of the horizon are less than he supposed them to be. 



In the series of observations given by the author, a similar want 

 of conformity to Bradley's law is observable ; and he observes, that 

 the change of difference, from greater to less, takes place at 80 

 zenith distance, which is the same point of the heavens assigned by 

 Piazzi. 



Mathematicians, who have endeavoured to reconcile the known 

 laws of refraction through different media, with the actual quantity 

 deduced from observation, have proved that the refractions vary nearly 

 as the tangents of zenith distance ; but in order to reconcile this rule 

 with the fact at low altitudes, they have found it necessary to intro- 

 duce a correction of the zenith distance, and have invented a formula, 

 consisting of a tangent of the zenith distance, diminished by some 

 multiple of the refraction. The magnitude of this multiple has been 

 estimated differently by different authors. By Simpson it is rated 

 at 2-75 ; by Dr. Bradley 3 ; by Bouguer 3'23 ; by Cassini 3'226. 

 Mr. Groombridge computes that this multiple should be as much as 

 3-3625. 



In addition to the above endeavours to determine the mean re- 

 fraction, and its variations at different altitudes, the author also con- 

 siders the corrections which should be made for the states of the ba- 

 rometer and thermometer, and explains the means by which he de- 

 duced those that he has adopted, in order that any error therein may 

 be more easily detected. 



Extract of a Letter from the Rev. John Brinkley, D.D. F.R.S. An- 

 drew's Professor of Astronomy in the University of Dublin, to the 

 Rev. Nevil Maskelyne, D.D. F.R.S. Astronomer Royal, on the an- 

 nual Parallax of a Lyra. Read April 12, 1810. [Phil. Trans. 1810, 

 p. 204.] 



The principal object of Dr. Maskelyne in making this communi- 

 cation, is to inform the Society of a discovery, made by Dr. Brinkley, 

 of the parallax of the annual orbit, which he has ascertained by ob- 

 servations on a Lyrse. 



