H, A. Hazen—Reduction of Air-pressure to Sea-level. 459 
of variation, at any station, of Scions and temperatures at 
different temperatures and pressu 
It would seem that if there He taken a sufficient number 
of stations, near enough together, at different elevations, the 
above result might be ‘attained. 
ere are very great difficulties in sae gi satisfactory 
results at present, owing to the great distance between stations 
and the lack of reliable elevations. In “Table TV we find, in 
the case of reduction from a higher to a lower station, the law 
determined from a comparison of all the mean daily observa- 
tions at Valdobbia and Alessandria during 34 years; compari- 
sons are also given for the same time between Great St. Ber- 
nard and Alessandria. 
here is a remarkable uniformity in the temperature term 
variation at all pressures at both stations, but the pressure 
term variation is subject to breaks and rather regularly dimin- 
ishes as the temperature increases. 
e following formule have been constructed from the 
eae observations at several elevations in Italy and Switzer- 
and 
Great Saint Bernard sous Geneva : C = 9°68 — 0"'22 (30” —p) 
0” 
‘014 (t—32°) 
Monte Cavo and Rome: C wa 66 — 0""10(30" —p) 
— 0"°0052(t — 32°) 
Mondovi and ee PO Cao" 10—0"-06(30" —p) 
"0027 (t—32°) 
and from these the general formula, 
C = 0":0000000117A? + 07-0011068/— 0” Dees e ieee 
—0"-0000016 (t—32°)h + 070214 
A/4b(30’ —p—0"°0214) ++a°—a (2) 
2b 
h, in feet, = 
in this a = 0"-0011068 — 0”-000029 (30" — P) — 0"-0000016 (¢ — 32°) 
and b = 0"-000000011 
Table V gives difference of elevation by various formule. 
The results by Angot are better than by peice It will be 
hoticed that the sixth column has been computed from a 
formula into which the original observations at "Valdobbia 
have not entered and in so far goes to show the reliability of 
teas ethod. The formula applies best to elevations of about 
