278 PROCEEDINGS OP THE AMERICAN ACADEMY 



and M respectively indicate Gunstock, Kearsarge (S.), Passaconna- 

 way, Whiteface, and Moosilauk, as in Table I. 



II. Harvard College Observatory. 



The station occupied was on the southern corner of the western bal- 

 cony of the Sears Tower of the Harvard College Observatory. Its 

 height is about 30 meters above mean tide level. Three points were 

 selected very nearly in line which are designated below as A, B, and 

 C. A is the summit of Fenno's Peak, the second highest of the Blue 

 Hills of Milton, and nearly the most distant terrestrial object visible 

 from the Observatory. Its height is about 160 m., and its distance as- 

 sumed to be 18.865 km. £ is the top of the chimney of the house 

 No. 20 Terrace Avenue, Roxbury District, Boston. Its distance is 

 assumed to be 7.042 km. C is the top of the roof at the base of the 

 lightning-rod of the gasholder on Everett Place, Cambridge. Its as- 

 sumed distance is 0.700 km. B is about 6' below A, and G about 

 17' below B. Each set consisted of nine settings in the order ABC 

 A B O A B G, the readings being made to tenths of one division of the 

 micrometer screw. The mean of the three readings of A was then 

 subtracted from the corresponding means of B and G. The various sets 

 are enumerated in Table V. The successive columns give a number 

 for reference, the date, and the hour and tenth in Greenwich mean 

 time. The day is here assumed to begin at Greenwich noon, or at 

 about seven o'clock in the morning of Cambridge mean time. The only 

 occasion on which the day changes during a series of observations is 

 between Nos. 120 and 121. The barometric pressure in inches after 

 correction for the temperature of the mercury, and the temperature in 

 degrees Fahrenheit, are given in the next columns. These are followed 

 by the intervals ^(7 and BC expressed in seconds of arc. One divis- 

 ion of the micrometer is assumed to be equal to 14". 2, in accordance 

 with a determination made at that time. As the observations are dif- 

 fei'ential, the uncertainty in the value of this constant will be inappre- 

 ciable in the result The following column gives the zenith distance 

 of the sun in degrees. The next two columns give the residuals in 

 seconds, found by subtracting from the observed value of ^ C its mean 

 value, 1392", and also the same residuals after applying the correction 

 for temperature and pressure to be described below. The last col- 

 umn but one gives the probable error of a single setting, expressed in 

 seconds of arc, derived from the accordance of the individual readings. 

 The last column gives the initial of the observer. E. indicates the 

 observations of Mr. Edmands, and P. my own observations. 



