Just as we found that the character of the fat of the coldblooded animals is adapted to the maintenance of a fluid or plastic consistency at the low temperature to which it is exposed, so to a less degree the character of the fat of warmblooded animals appears to vary with its position in the body and with the temperature to which the body is subjected during the time that the fat is in process of formation. Thus Hen-riques and Hansen concluded from experiments with pigs that the thick layer of subcutaneous fat on the back, where it was not thoroughly warmed by the blood and therefore had an average temperature considerably below that of the interior of the body, was richer in unsaturated compounds (olein, etc.) and had a lower melting point than the fat of the body as a whole; while the fat from animals which had been grown in a warm room, or which had been heavily jacketed so that the skin was not exposed to cold air, contained near the skin fat of more nearly the same composition as in the interior of the body.

Moulton and Trowbridge have observed that the fat in beef animals becomes richer in olein and therefore softer with age, with fatness, and with nearness to the surface of the body.

Usually, however, the nature of the fat found in the body is more or less characteristic of each species or group of closely related species. Herbivora contain as a rule harder fats than carnivora, land animals have harder fats than marine animals, and all warm-blooded animals have fats which are decidedly harder than those found in fishes. The fats of different mammals were investigated by Schulze and Reineke, whose results * showed little variation from an average of carbon, 76.5 per cent; hydrogen, 12 per cent; oxygen, 11.5 per cent, as may be seen from the following:

The foregoing statements refer to the fat of the adipose tissues. In the fat extracted from the liver, kidney, and heart, Hartley* finds fatty acids of the series CnH2n-4O2, CnH2n-6O2, and possibly CnH2n-8O2.

* Armsby's Principles of Animal Nutrition, page 61.

† Benedict and Osterberg {American Journal of Physiology, Vol. 4, page 69) found in 8 samples of human fat an average of 76.08 per cent carbon and 11.78 per cent hydrogen.

A possible explanation of this difference between the fat of the adipose tissues and of the actively functioning organs is to be found in the greater reactivity of the unsaturated acid radicles. The saturated fatty acid radicles are relatively stable and inert; and when the glycerides of such acids are deposited in the inactive adipose tissues, the fats may remain unaltered for a long time and accumulate in considerable quantities. The unsaturated fatty acid radicles are less stable and more readily acted upon and broken up. This is consistent with the fact that we find them more abundantly in fats of the organs in which metabolism is more active and has led to the view that the desaturation of fatty acid radicles by the active organs of the body may be an important preliminary to the metabolism of the fat. On the other hand, the formation of unsaturated fatty acid radicles such as oleic and linoleic does not, according to our present knowledge, seem essential to the "ß-oxidation theory" which is now generally held as most probably representing the main course of fatty acid metabolism (Chapter V (The Fate Of The Foodstuffs In Metabolism Carbohydrates. Oxidation Of Carbohydrate)). It is therefore entirely possible that the highly unsaturated fatty acids found, for example, in the liver, may be present as constituents of the protoplasm of these cells, essential to the properties which enable them to carry out some of their functions but not necessarily connected with the metabolism of fat itself.

Butter fat differs from body fat in containing fatty acids of lower molecular weight (particularly butyric acid, which is fairly characteristic of butter), and so shows a higher percentage of oxygen and lower percentages of carbon and hydrogen. The most abundant acids of butter fat are, however, palmitic, oleic, and myristic, and the ultimate composition is not very greatly different from that of body fats. A sample of butter fat analyzed by Browne * showed 75.17 per cent carbon, 11.72 per cent hydrogen, and 13.11 per cent oxygen.

* Journal of Physiology, Vol. 36, page 17.