be very properly analysed the diet on which this was possible, and found it to con-
tain quantities of nitrogen and carbon which (when allowance is made for difference
in weight) were practically identical with Letheby's estimate for ordinary labour
in England, and of the same value as the hard labour diet of English convicts. On
the basis of this experience his three scales of diet were formulated :—

     No. 1 Scale.—The labour scale; is simply the Bombay House of Correc-
         tion scale raised or lowered proportionally to the weight of the indivi-
        duals to be fed.

     No. 2 Scale.—The light labour scale; is scale No. 1 reduced in the pro-
        portion that the light labour diet of English convicts is lower than
        English convict hard labour diet.

     No. 3 Scale.—The bare sustenance scale; is scale No. 1 reduced in the
        proportion that a mean between Edwin Smith and Letheby's esti-
        mates for bare sustenance is lower than English convict hard labour

     Dr. Lyon, therefore, devised his diets by working back from an analysed diet
which had proved itself sufficient for all physiological purposes, i.e., a diet from
which enough of the different proximate principles was absorbed to supply the
needs of the body. The ultimate object of the present investigation is to devise
diet scales in which the several ingredients are so combined that the maximum
absorption is obtained with a minimum of waste—a point not taken notice of in
either Lyon's diets or any other diet scales that have been framed for Indian jails.

     The reason why dietaries for prisoners in India have all been worked out from
the chemical analyses of the foods in use appears to be that, from the investigations
that have been carried out in Europe and America on the percentage absorption
of the proximate principles, a fairly fixed proportion of these was found to be
absorbed, i.e., above 90 per cent. in a mixed animal diet and above 85 per cent.
in a so-called vegetable diet. So that, by assuming these percentages to hold good
for Indian food-stuffs, the framers of the different diets estimated that between 85
and 90 per cent. of the proximate principles found by chemical analysis in the diets
would be absorbed. This we shall show is not by any means the case; and, further,
we shall give abundant evidence that the actual amount of absorption does not
vary directly with the quantity of the proximate principles contained in the diet;
this at least we can strongly assert is so in the case of most important of the
proximate principles—protein. It was a very natural assumption to make and, so
far as we know, the fallacy was not discovered until actual examination* of the
excreta of prisoners on a known jail diet proved that a comparatively poor absorp-
tion of protein takes place when the actual amount of protein offered in the diet is
taken into consideration.

* Scientific Memoirs No. 34—Standards of the Constituents of the Blood and Urine and the Bearing of
the Metabolism of the Bengali on the Problems of Nutrition.