126 DETERMINATION OF ALTITUDES BY BAROMETER. 
Whenever water-courses gave practical checks upon the relative altitudes of different stations, 
as was often the case, the profile bore the most careful study in a perfectly satisfactory manner, 
and confirmed, without exception, the use of the mean daily air temperature, instead of the 
observed, in the computation. 
HEIGHT OF FORT READING; EXPLANATION OF TABLES OF BAROMETRIC OBSERVATIONS, ETC., IN 
APPENDIX D. 
The great number of observations taken at Fort Reading has enabled me to compute its alti¬ 
tude in two different ways ; one of which is independent of the tables of horary and abnormal 
correction. We had very careful hourly observations taken from July 22 to July 26, and from 
November 1*7 to November 22, inclusive. It is well known that a mean of the observations at 
7 a. m., 2 p. m., and 9 p. m., differs very little from the mean barometric reading for the 
whole twenty-four hours ; and, as five of the days on which the above mentioned observations 
were taken were in the dry, and six in the rainy season, it is probable that a mean of their 
mean readings thus found may approximate to that of the year. It is 29.506 inches. The 
corresponding mean air temperature, found by taking a mean of the observations at 7 a. m., 
12 m., and 10 p. m., for the eleven days, is 62°.3 Fah. The altitude given by these data is 
518 feet. 
By the second method, I applied the tables as in other cases, and found the corrected mean of 
all our observations at the fort, which were about 600 in number, and, with the corresponding 
mean air temperature, computed the altitude. It is 544 feet, differing only 26 feet from the 
other. I have adopted the first result on the profiles, as it is obtained from observations taken 
with very great care, while many of the others are less reliable. 
The tables in Appendix D exhibit the original data for the construction of our profiles, 
and the altitudes deduced from the observations. It must be remembered that all the altitudes 
are referred to the level of mean tide at Benicia, as the barometric reading at the sea level 
north of that place is not known. The column headed “ corrected barometric reading” gives 
the height of the mercurial column, after the corrections for reduction to 32° Fahr., for instru¬ 
mental error, and for horary and abnormal oscillation, have been applied. The very slight 
difference between these corrected readings, when there are several taken at the same camp, 
confirms the accuracy of the horary and abnormal tables. 
