: 
; 
3 
4 
4 
: 
4 
eee ee ae 
on the Elasticity of certain Metals. | 350 
was assumed that this thermometer follows the gradually dimin- 
ishing temperature of the air to within a negligable = pena 
and the three thermometers were compared with it, under co 
ditions similar to those ee in the observations on clastcity 
All three thermometers were disposed at an equal height wi 
the thermometer of com eatin in the gradually emeundin 
temperature of the hollow space of the tin cylinder, and simul- 
taneous readings were made. They were also compared togeth- 
er in the ordinary manner in water of different temperatures. 
From these observations, which were conducted by Mr. Grotri- 
an, it resulted that the mean of the readings of the rah ther- 
mometers remained 2° ‘02 behind the temperature of the ther- 
mometer of comparison, when the velocity of the diminishing - 
temperature amounted to 1° in lL min. The correction of the 
readings due to the mass of quicksilver is therefore 
dt 
2°02 7, 
and that of the interior of the apparatus one part. If 1, de- 
notes the temperature of the room, the correction of the read- 
ing t is therefore —4$r(r—1,) 0,00016. 
This value cannot be far removed from the truth. It coincides 
with the difference which the determination of the boiling point 
afforded, according as the thermometers were placed in the 
steam only to the zero point of graduation, or a the boiling 
point, The uncertainty of this assumption is at all Lovet of 
small importance, since the whole correction for the highest 
tem wentiiee (90° ) amounts only to 0°°8. 
us, three —a* are to lied to the readings of 
the. duces e first, which relates to to the sey 
errors, is a function ape is temperature ; in co 
quence of the mass of the quicksilver, is a ye of "che 
pelosi of the diminution of the temperature ; the third, final- 
ne in consequence of the column of quicksilver projecting from 
€ apparatus, depends on the difference of the outer and inner 
