It might be expected, perhaps, from the foregoing, that I' = I; this
would take place if the excess of temperature of the metal, measured
by the contraction, were rigorously proportional to the heating of the
liquid, for then the two quantities would be null at the same time.
Careful experiment proves that this is not the case. The sulphate of
copper gives compressing deposits on a thermometer which is undoubtedly
cooling; chloride of zinc of a density 200 can give expanding
deposits on a thermometer which is heating. There is, therefore, no
proportionality; but it must be remarked that the temperature of the
metal which is deposited does not depend only on the quantities of heat
disengaged in an interval of molecular thickness which is infinitely
small compared with the thickness of the layer, of which the variations
of temperature are registered by the thermometer. There is nothing
surprising, therefore, that the two variations of temperature,
according exactly with one another, do not follow identically the same
laws.--_Comptes Rendus._

* * * * *




ANALYSES OF RICE SOILS FROM BURMAH.

By R. ROMANIS, D.Sc., Chemical Examiner, British Burmah.


The analyses of rice soils was undertaken at the instance of the Revenue
Settlement Survey, who wanted to know if the chemical composition of