nent illustrations of these. It is important to become familiar with the use of tables of food analysis, with clear knowledge of the essential points that should be looked for. The short table used above in the construction of a normal diet gives in round numbers the compositions of the commonest food-stuffs rated in proximate principles and in their two important elements of income. We may use this table again to estimate the nitrogen and carbon values of a hospital' ordinary' diet, e.g. :

Obviously, this sum-total must have been inadequate to sustain life in a large percentage of cases, especially if the patients were not confined to bed and entering upon convalescence, and as a matter of course it is supplemented by the liberal allowance of 'extras.' We may add for the sake of comparison the following figures:

All these diets are too low to sustain life at a normal level of energy for any prolonged period in the various classes of persons to whom they are administered. Flagging energy and loss of flesh are quickly produced in the first three examples, and even in the last three it has been found necessary to raise the diet or to lower the labour.'

The factors of nourishing value.-To properly estimate the

The veil that measurements in pounds, ounces, pints, and grains casts over a diet-table is best removed by converting them at once into grammes and cubic centimeters, and then proceeding to take percentages. It is easier to estimate diets by elements than by proximate principles, heed being taken that carbon has a higher value in fat than in carbohydrate. The real value of a diet often falls short of its paper value.

nourishing value of a food we must take into account, not merely its percentage of carbon and nitrogen, but also its palatability and its digestibility. A nourishing' food must not only be chemically adequate, it must also be palatable and digestible. It may possess any one, or even two, of these three essentials, and yet not be nourishing. Thus, dried peas and beans, or even horsechestnuts and acorns, are chemically adequate, but they are not palatable. Cheese and hard-boiled egg stand high as regards chemical value, but are very indigestible. Conversely, it may happen that a substance is particularly digestible, so that it has a nourishing value which its chemical composition would not have led us to expect; gelatin, for instance, in the form of calves'-foot jelly, is of indisputable practical value in the nourishment of the sick.

Stimulants. The stimulants in common use are alcohol, tea, coffee, and cocoa, the last-named deserving to rank among foods. Alcohol is a typical stimulant; it acts as a whip, causing a temporary acceleration of physiological activity. Such acceleration must subsequently be paid for, the extra expenditure brought about by alcohol entailing diminished capacity for further exertion. Alcohol is thus of service only for emergencies of short duration; it is eminently harmful when prolonged exertion and endurance are required. Like all rapid stimulants, alcohol is in large doses a direct depressant. Tea and coffee owe their stimulating property to the alkaloid caffein. They are more useful than alcohol because less liable to abuse, and less dangerous when taken in excess. Cocoa is a stimulant by virtue of theobromine, a food by virtue of the large amount of fat or oil that it contains.

The consideration of cost is of great practical importance and not without physiological interest in questions of diet; for the supply-price is a determining factor in the diet-scales unconsciously selected by large masses of men, and laid down by the governing authorities of prisons and of charitable institutions.

The points that the table given below teaches or illustrates are: (1) That nitrogen is much more costly than carbon. (2) That carbon in fat is more costly than carbon in starch or sugar. (3) That nitrogen in an animal proteid is more costly than nitrogen in a vegetable proteid.

The table also accounts for the fact that the poorer classes obtain their carbon by sugar rather than by butter, and that, even in the country, milk and eggs rank as luxuries. Beer is recognised to be an extravagance; rice, which is a staple food

T

among Oriental nations, and peas, which are a representative article in a vegetarian diet, are particularly economical. And in the prices of the three articles, bread, oatmeal, and potatoes, which are most largely consumed by the masses in England, Scotland, and Ireland respectively, we can recognise that the Scotchman gets most for his money, the Irishman least.

COMPARATIVE COST OF SOME COMMON FOOD-STUFFS.

According to the now discarded theory of Liebig, proteids were regarded as plastic, fats and carbohydrates as calorific. Bidder and Schmidt considered that only part of the proteid was plastic, and that the remainder was oxidised wastefully, or, as they termed it, that it underwent luxus consumption. These views are abandoned; the only sense in which a luxus consumption of proteid can be presumed to occur is by the short-cut' conversion alluded to on page 249; and we are about to learn that all food is calorific. The dominant doctrine since 1873 is that of Voit and of Fick, and we have followed it in the above exposition, more especially in that portion which is presented on pp. 256-260. But it should be stated that its fundamental principles have recently been called in question by Pflüger. Whereas, according to Voit, body-fat is produced from food-proteid, but not from carbohydrate, and according to Fick the work of the body is effected at the expense of non-nitrogenous food, Pflüger maintains that body-fat can be produced by food-carbohydrate, that Voit's analyses are not valid evidence of the production of fat from proteid, and that the work of the body is effected mainly at the expense of proteid food. That proteid alone can be the adequate source of heat plus work is clearly demonstrated in the case of dogs fed exclusively upon lean meat, but Pflüger goes further, and maintains that proteid is the chief driving food, even when accompanied by other kinds of food.

ANIMAL HEAT

Physical data.-The thermometer is an instrument containing a fluid, usually mercury, which expands by heat. The expansion, and consequently the intensity of the heat that affects it, is measured in degrees; these are the conventional parts into which the total expansion between the freezing and boiling points of water is divided. The scales of degrees in common use are those of Fahrenheit and of Celsius. On the Fahrenheit scale the total expansion is divided into 180 parts or degrees, the freezing point is at 32°, the boiling point at 212°. On the Celsius or centigrade scale the total expansion is divided into 100°, the freezing point is at 0°, the boiling point at 100°.

The calorimeter.—Sources of error.-Heat-units.-Amount of heat is measured by the calorimeter, intensity of heat or temperature by the thermometer. Two vessels of water, containing respectively 1 and 2 litres, may be at the same temperature measured in degrees above zero; the first possesses, however, half as much heat as the second, or if the same two vessels of water possess the same amount of heat, then the water in the first vessel must be at a temperature double that of the second. A man weighing 70 kilos possesses twice as much heat as a boy of 35 kilos, the temperature of both being the same; a fever patient weighing 60 kilos, with a body-temperature of 42°, possesses the same amount of heat as a collapsed person weighing 70 kilos with a bodytemperature of 36°.

The unit of heat is the calorie, i.e. the amount of heat required to raise 1 gramme or cubic centimeter of water 1° centigrade; 10 calories are an amount of heat sufficient to raise 10 grammes of water 1°, or 1 gramme of water 10°, or 5 grammes of water 2°, &c.; thus the amount of water in grammes or cubic centimeters, multiplied by the temperature in degrees, gives the amount of heat in calories. The units of heat further employed to express physiological results are the kilo-calorie and the milli-calorie, the first being the unit generally used in numbers relating to the total heat-production of animals, the latter that used in delicate measurements made on excised muscles. The kilo-calorie is 1,000 calories, or the amount of heat required to raise one litre or kilo of water 1°; the milli-calorie is 1000 calorie, or the amount of heat required to raise 1 milligramme of water 1°.

The calorimeter, as used in physiology, is essentially a chamber within which an animal can be confined so as to impart all the heat it produces to a known volume of water contained in a surrounding chamber; an external chamber, packed with non-conducting material, limits as far as possible loss of heat by the water, and the temperature of the latter is read upon thermometers. Let us suppose that the water-chamber contains 10 litres of water, which are raised from 20° to 25°, viz. 5° in one hour, by a cat or dog confined in the inner

chamber. We know therefrom that the cat or dog has given off 50 calories.

Sources of error.-But if the theory is simple, the sources of experimental error are numerous and considerable. It is difficult to determine accurately the mean temperature of a large quantity of water; the material of which the apparatus is constructed absorbs heat; the non-conducting envelope is not a perfect insulator; the

temperature of the animal may have risen or fallen; a current of fresh air, which carries off heat and moisture, must be made to enter the chamber. All these circumstances must be taken into account and allowed for as accurately as possible, but it is obvious that a considerable margin of error remains unavoidable.

Mechanical equivalent of heat.-Energy cannot be destroyed; it can only be transferred from one place to another. Its various forms are heat, work, chemical action, and electrical action, and it has been experimentally determined how much of one form of energy is equivalent to how much of another form. Thus, it has been determined by Joule that 1 heat-unit is equivalent to 424 work-units. The heat-unit is the calorie; the work-unit is that amount of work required to raise 1 gramme to a height of 1 meter, and is called the grammeter; the kilo-grammeter and the milli-grammeter are its multiples by 1,000 and by roo. Therefore

1 calorie = 424 grammeters.

1 kilo-calorie = 424 kilo-grammeters.

1 milli-calorie = 424 milli-grammeters or 424 gramme-millimeters.