210 
MR. H. L. CALLENDAR ON THE PRACTICAL 
[To exclude errors of this nature, and to render the observations of different series 
strictly comparable, the value of 9 0 assumed must be the same throughout, since the 
value of t deduced varies directly as that of 6 0 assumed. All the observations 
entered on Plate 3, fig. 1, are therefore reduced, assuming the same value of d,„ 
namely, 272'90° C., which corresponds to a coefficient of dilatation for dry air of 
•0036642 per 1°C. This agrees very closely with Pi,egnault, and is deduced from the 
pair of observations (16), (17), Table A.-v. An inspection of the values of mk, cal¬ 
culated with this coefficient from the observations in ice and steam, and at the air 
temperature, from day to day, will show that it satisfies all the observations practically 
within the limits of error of the pressure measurements. The largest discrepancy 
between observations on the same day (Table A.-v., March 8, observations (25), (26) ) 
amounts to nearly 1 in 2000.] 
The variations of mk from day to day are sometimes larger, and more unaccount¬ 
able ; they are generally in the direction of absorption, but I have so far been unable 
to reduce them to rule. Tire difficulty is to a great extent avoided by observing mk 
frequently and using the appropriate value in the reduction of the day’s observations : 
this method must, in part at least, eliminate the error, to ivhatever cause it is due. 
Table A.-iv.—C omparison of Platinum and Air Thermometers. 
Time. 
Auxiliary 
thermo¬ 
meters. 
Resistances 
observed. 
HoSO, 
gauge. 
Mercury 
manometer. 
Barometer 
and 
attached 
thermo¬ 
meter. 
Number of 
observation. 
Results. 
Spiral. 
Connec¬ 
tion. 
X 
1 
m e 
Aug. 7 
R. 197 
20-437 
1-062 
Befot 
e filling ai 
r therm 
ometer 
(1) 
R 0 = 19-945 
Aug. 9 
11-12 A.M. 
A. 20-5 
R. 18 8 
M. 2P5 
43-435 
Mean of 
2-021 
several 
3-51 
obsei-v 
M e -M,= 
ations. 
+ 7-721 
30-085" 
70-2° F. 
(2) 
pt = 333-5° 
t = 350-5° 
d = 17-0° 
12.8 
R. 193 
36-64 
3-70 
48-09 
41-55 
t = 240-4° 
12.13 
A. 20-7 
36-51 
1-730 
2-60 
48-11 
41-52 
30-075" 
pt = 232-5° 
12.17 
M. 21-4 
36-42 
# # 
1-87 
48-10 
41-51 
70-7° F. 
fid 
12.22 
36-30 
3-32 
45-23 
38-07 
d - 7-9° 
3.30 
A. 20-6 
36-79 
3-40 
34-87 
28-56 
t = 243-7° 
3.41 
M. 20 9 
36-75 
1-7-56 
3-20 
34-86 
28-54 
30-052" 
/'.n 
pt = 236-6° 
3.48 
R. 19-3 
36-73 
1-05 
36-26 
30-42 
71-7° F. 
3.52 
3676 
1-25 
36-26 
30-41 
d= 7-1° 
