Ai'PENDix. UPON BOILING-POINT OBSERVATIONS. 417 



D.— UPON BOILING-POINT OBSERVATIONS. 



Henderson's boiling-point apparatus, and nine boiling-point thermometers, 

 were taken lest accidents might occur to the barometers. The thermometers 

 were self -registering ; they were constructed in two series, in order that the 

 scales might not be too finely divided ; and they were verified in the custom- 

 ary manner. 



As the barometers were not injured, it was unnecessary to depend for 

 altitude upon the indications afforded by the boiling-point of water. Obser- 

 vations of the boiling-point were, however, made at seventeen different stations 

 from 8400 to 19,500 feet above the sea, for the purpose of comparison against 

 the. harometers. These observations have been worked out for altitude by Mr. 

 W. Ellis, F.R.A.S., but are not quoted in this volume. As a general rule, 

 three or more thermometers were boiled at each station, and in the calcula- 

 tions for altitude the means have been used of the observations which were 

 made on each occasion. 



It was found, in all cases, that the mean boiling-point at every station 

 was higher than the corresponding barometer would lead one to expect,^ from 

 the tables which are generally accepted as authoritative, namely, Eegnault's, 

 as given in the Smithsonian Tables, Meteorological and Physical, third edition, 

 Washington, 1859. The divergencies became more pronounced the higher 

 we ascended, and this leads to the opinion that the tables are not perfect. 



I venture to take the opportunity to make a few general remarks upon 

 the method of attempting to determine altitudes by observations of the 

 boiling-point of water ; but I offer them with great diffidence, as they will 

 be found somewhat opposed to the expressed opinions of eminent travellers, 

 and of persons in authority. 



Of late years, the practice of deducing altitudes from observations of the 

 l)oiling-point of water has to a considerable extent superseded the older 

 method of obtaining them by observation of the mercurial barometer ; and 

 it has been assumed (I think, erroneously) that the former method is but 

 slightly inferior to the latter in accuracy, and is superior to it in facility. 

 I think that any person who will, even at the level of the sea, take the 

 trouble to examine the matter for himself, will speedily be convinced that 

 the boiling-point method is one which must always be liable to considerable 

 errors iipon the mercurial barometer ; that mountain-travellers who have had 

 experience of the two methods — more especially those who have experimented 

 upon the summits of lofty peaks — will concur with me in saying that the 

 occasions are very rare indeed upon which observations of the mercurial baro- 

 meter cannot be made, even upon those on which high wind and severe cold are 

 experienced at inconvenient situations ; and that, upon the other hand, occa- 



1 The heights deduced from the boiling-point observations are consequently lower 

 than the truth. 



