474 
Dr. Brinkley's remarks on 
337 days of observation, reduced to January 1, 1815. The 
results for 1812 and 1813 are deduced immediately from 
Mr. Pond's reductions* of his observations published with 
the observations for 1814. The others I have calculated from 
the observations. 
Let us consider each of these five years separately. In 
the year 1812, the days of double observation were 69 ; but 
of these 23 were rejected or not used by Mr. Pond in his 
“ reductions ” above mentioned. There then remained 4 6 ; 
but we find that during the time these observations were 
made, the position of the telescope was changed five times. 
This therefore may account for part of the discordances 
which will be found in the measures of the intercepted arc, 
if the observations of each day are examined. It also renders 
the observations less fit for the investigation of parallax. I 
have however calculated from these observations, and find 
the constant of parallax = o",28. Mr. Pond, in his Table XI. 
makes the difference between the summer and winter obser- 
* These reductions are computed by Bradley’s refractions, and therefore I 
have computed the rest by the same refractions. By Bradley’s refraction, as is 
well known, a greater allowance is made for change of temperature than is now 
generally admitted. Bradley’s table supposes the refraction is increased 
part, by diminishing the temperature i®. The more exact quantity appears to be 
part. The difference is 24V5 part for l °- The effect of using part is to 
make the arc between y Draconis and a Lyrae appear larger in winter than in sum- 
mer, and therefore more in favour of parallax. Between July and November, an 
interval particularly considered hereafter, the difference is insensible. 
I may remark here, that Mr. Pond, Philosophical Transactions, 1817, p. 163, 
appears to have erred in estimating the effects of the French refractions in his re- 
sults for a Lyrae. He seems not to have considered that many of the southern stars 
he used for the index error were farther from the zenith than a. Lyrae ; hence, in- 
stead of diminishing the quantity he had found in favour of parallax, he should 
have increased it. 
