358 Proceedings of the Royal Society of Victoria. 



starting point, as well as a barograph, but this is rarely necessary. 

 Every opportunity that presents itself of comparing the aneroid 

 with a good mercurial barometer should be taken advantage of 

 as a check. 



Reducing the Observations and Computing the Altitudes 



Many formulae and tables have been published for this purpose, 

 but most of them are needlessly cumbrous. There is no need of 

 computing decimals of a foot when the instrument will not 

 indicate anything less than 10 feet, or 5 feet at the very closest, 

 and when continual atmospheric variations affect the result by 

 several feet in perhaps a few minutes of time. The same may 

 be said of such theoretical refinements as correction for the 

 variation of gravity due to latitude or altitude. 



Table I. 



The following table, founded on those of Laplace and Guyot, is 

 recommended for use in Victoria :■ — 



