1833.) Mountains from the boiling point of water. 197 
The Rev. F. J. H. Woutuasrton was the first to introduce the ther- 
mometer practically as a substitute for the barometer in measuring 
heights. His plan was merely to render the thermometer more deli- 
cate by increasing the bulb, and allowing og mercury to enter the 
capillary tube only when it approached the boiling point, so that a few 
degrees occupied the whole scale, and by a sliding nonius each degree 
could be divided into 200 parts or more. But it is evident that to 
compete with the barometer in accuracy of indications, the scale must 
have a range of the same length as that of the barometer,—say 15 
inches, and the instrument would thus become fragile and unwieldy ; 
to obviate this inconvenience, he formed a reservoir above the capillary 
tube, containing a small supply of mercury, so that when the boiling 
temperature should be so reduced as to bring the reading point to the 
foot of his 6-inch scale, a portion of mercury was to be added to bring it 
to the top of the scale, by an operation so delicate and difficult that I 
may safely say, and from experience too, that few travellers would re- 
sort to it in the field, and fewer still succeed if they attempted it. In 1817, 
he exhibited his thermometer to the Royal Society, and in 1820, he ap- 
plied it to the measurement of Snowdon. On the latter occasion, he drew 
‘up a table of the value of the degrees between 214° and 202° in feet, 
founded on Doctor Urz’s empirical formula of tensions; but, as this 
range only extends to an altitude of 5405 feet, it is evidently quite in- 
sufficient for the traveller in India, who may ascend to 18,000 feet and 
still see Snowdons towering above his head. 
The error into which Wottasrton fell was an attempt at too great 
sensibility. His instrument is beautiful in a laboratory, where it will 
serve to shew minute variations in the index error, as it may be called, 
of a barometer in the course of years, as I have frequently proved. But 
for rough work out of doors, accuracy must in some measure be sacri- 
ficed to strength and portability, the points in which alone the thermo- 
meter can boast superiority over the barometer. Captain Hersert 
was so well aware of this, that he had provided himself from England 
with ordinary thermometers divided, from 180° upwards, to the tenths of 
degrees: half a division thus represented about 25 feet, which in 
most cases was ample, especially when the zero of elevation, or level of 
the sea, was 1000 miles distant. 
All who possess thermometers, therefore, divided to tenths of inches, 
may convert them into measurers of height, by attending only to a few 
trifling precautions in their use. 
1. The prime boiling point 212° should be accurately verified by com- 
parison with a good barometer, for the divisions of the instrument- 
makers are by no means to be trusted within the requisite limits. Thus, 
