E.. Loomis—Contributions to Meteorology. 8 
tion to sea-level from the barometric height 22° 3 oe to 24°2 
inches and from the temperature —10° to +65 order to 
smooth down the inequalities of the observed casi I took 
the mean between each three consecutive numbers correspond- 
ing to the same temperature, and substituted this result for the 
middle number. It is presumed that the results thus obtained 
represent pretty nearly the results which Me — peer 
rom observations extending over a long term of y 
The results thus described. are “Gehibnay in the “follonane 
Table, in which the height of the barometer on Mt. Washing- 
ton, from 22-7 to 24-2 inches is given at the top of the pte: 
and the mean temperature of the air column from —10° 
+65° is given on the left margin. Corresponding to each sa 
perature given in the table are four horizontal lines, the first 
of which (marked D), gives the reduction to sea- Jevel as com- 
puted from Dunwoody’s Tables; the second pig ess bat ins 
marked L), shows the reduction computed from Guyot’ 
bles founded on the formula of Laplace; the third oriageraat 
- (marked P), shows the reduction computed from William- 
s Tables, which are based on those of Plantamour; the 
ste horizontal line (marked O), shows ‘the reduction deduced 
from actual observations as above described.* 
n examination of this table atiows the following results: 
1. Dunwoody’s Tables accord very well with those derived from 
the formula of oh the differences ranging from +0011 
inch to —0-041 i 
2. The adferaees between the formulas of Laplace and 
Plantamour range from +0°080 inch to +0103 inch, the reduc- 
tion by nese ace being on an average 0°053 inch greater than 
by Planta 
3. The ieee deduced from the actual observations dif- 
fer very much from either of the values above computed; the 
differences from Laplace ranging from +0263 to —0°105 inch, 
These differences follow a remarkable law. Acociie to the 
formula of Laplace, when the pressure on Mt. Washington in- 
creases from twenty-three inches to twenty-four inches without 
any change of temperature, the reduction to sea-level is in- 
creased by gy part of its former value. Observations, how- 
sir, show that the actual increase in the amount of the reduc- 
tion is very small, being on an average only one-seventh as great 
* Since this article ~~ written I have been informed that the constant 6°36 
inches for reducing t. Washington observations to sea-level began to be used 
hs. lenge Ist, 1874, and that tar the two preceding years the constant 6°31 inches 
ie 
umn of observed reductions to sea-level, which is so small an i Pinas I have 
not considered it necessary to re-compute the entire series of observ 
