950 
Proceedings of the Royal Society 
character. An explanation is doubtless found in the fact that the 
metallic element whose molecules acquired the greater activity owing 
to the increased temperature (consequently becoming more liable to 
attack from oxidation and corrosion), would, under the continuance 
of such circumstances, he likely to remain the positive metal. 
When, however, by the reduction of temperature, its molecular 
rigidity increased, compared with that of the other element, an 
interchange of electro-chemical position might he expected to occur. 
This cause appears sufficient at these high temperatures to produce 
complete interchanges and reversals between metals of such apparent 
general stability as platinum and copper or platinum and iron 
during the continuance of abnormal and unequal thermal conditions. 
The great difference of temperature which obtained in different 
parts of the same fusing salt would also cause it to act on the metals 
after the manner of two dissimilar solutions. Thermo-electric and 
inversion effects at these very high temperatures are probably also 
involved in the explanation. 
6. The Hygrometry of Ben Nevis. By H. N. Dickson. 
The observations tabulated in the present paper were taken at 
Ben Nevis Observatory, chiefly during August 1885. In a note 
communicated to the Society at the end of last session, a description 
was given of some observations of air-temperatures and humidities 
at Granton Marine Station, and it was stated that the work was to 
be continued at Ben Nevis. At present we confine ourselves to the 
hygrometrical work : it is hoped that the results of the temperature 
observations will be communicated to the Society early next session. 
The extreme dryness of the atmosphere which prevails at Ben 
Nevis during anticyclones, makes the Observatory an unusually 
favourable station for such investigations, and it is believed that the 
appended tables afford exceptionally good material for rediscussing 
the whole question of the hygrometry of the atmosphere. 
The observations here given were for the most part taken during 
an anticyclone which lasted from 18th to 28th August. They 
include all the regular hourly observations taken during that time, 
except when the air was saturated, besides a large number of extra 
observations, given at the end of the table of hourlies. Through 
