136 
LITTLEHALES. 
break-circuit chronometer in the electrical circuits which 
actuate the magnets for opening and closing the valve at 
the orifice. The quantities of mercury that were run out at 
the various places of observation were saved and afterwards 
weighed where a weighing balance of precision could be used, 
but the results were not such as to encourage the investi¬ 
gator to hope that the refinements of measurement requisite 
for the determination of the relative force of gravity at the 
various stations could be attained in this manner. 
Nearly twenty years ago Mascart made trials to determine 
the variation of the force of gravity from place to place by 
a method which should prove almost as practicable at sea 
as on land. He used a syphon barometer whose short arm 
was closed, and contained a certain quantity of gas. At a 
given temperature this gas had a certain volume and exerted 
a certain pressure to contribute toward balancing the mer¬ 
curial column: the greater the force of gravity, the shorter 
the barometric height corresponding to the pressure. The 
apparatus is reported to have been tested up to 69° of north 
latitude, but the correction for temperature introduced such 
difficulties that the required exactness could not be attained. 
Recently Prof. H. Mohn has published a method which, 
having already afforded him some degree of practical satis¬ 
faction on land, awaits trial at sea (Das Hypsometer als 
Luftdruckmesser und seine Anwendung zur Bestimmung der 
Schwerekorrektion, Christiania, 1899). He employed, side by 
side, a boiling-point apparatus and a mercurial barometer, 
and obtained a direct function of the value of the force of 
gravity by comparing the true atmospheric pressure, as in¬ 
dicated by the temperature of unconfined steam, with the 
pressure indicated by the height of the barometric column. 
To attain the necessary degree of exactness in determining 
boiling points to meet the requirements of this method, it 
has been necessary to construct a thermometer capable of in¬ 
dicating temperatures within one-five-hundredth part of a 
degree on the centigrade scale. Dr. Guillaume, of the In¬ 
ternational Bureau of Weights and Measures at Paris, appears 
