102 



RECORD AND REDUCTION 



the first part comprising the results in Baffin Bay, from September, 1857, to 

 August, 1858, inclusive, and the second part, the results at Port Kennedy, from 

 September, 1858, to August. L859, inclusive. 



For greater convenience the results by the aneroid have been reduced to the 

 results by the mercurial barometer, by the application of the correction — 0"\221. 



The readings for the hours 4, 8, 12. A. M. and P. M., for the first eight months 

 between Sept. and April, were taken from the preceding abstract of the mercurial 

 barometer (the readings in Sept. February, and April from the table containing 

 the referred means). All tabular numbers for the same eight months, at the hours 

 2. G, 10, A. M. and P. M., are derived from the readings of the aneroid barometer 

 by interpolation by means of differences; thus to obtain the reading at 10 A. M., 

 in September, we have — 



Aneroid reading at 10 A.M. 0.013 greater than at 8 A.M. Mercurial baro- 

 meter reading at 8 A. M. = 20.715, hence at 10 A.M. = 29.728; again, aneroid 

 at 10 A.M. 0.003 smaller than at noon. Mercurial barometer at noon 29.727, 

 hence at 10 A. M. = 29.724, and the resulting mean from the comparison of the 

 preceding and following hour becomes 29.72G as given in the table. 



The annual mean for the hours 2, 6, 10, A. M. and P. M. is obtained in a similar 

 manner; thus, for 10 A. M. we have: From 8 months, Sept. 'to April, mean at 

 10 A. M., the reading 0.020 greater than at 8 A. M. or = 29.731 + 0.020; it is 

 also 0.00G greater than at noon or = 29.743 + 0.00G; the mean of the two 

 values is 29.750 as given in the table. 



Diurnal variation of the atmospheric pressure during the year from September, 

 1857, to August, 1858, in mean latitude 72°.5 N., and mean longitude G5°.S W.; 

 nearly in the centre of Baffin Bay. 29 inches is to be added to the tabular 

 numbers. 



The table of bi-hourly means for the second group was obtained from the general 



recapitulation of result, by subtracting 0.221 from each mean to reduce it to the 



ling of the standard marine barometer, and by referring the incomplete means at 



the hours 2, G, 8, A. M. and P. M., to their corresponding value for a complete series 



of 12 values by a process similar to that explained in case of the preceding table. 



