For the purpose of determining the physiological role which each of the vitamins plays, it is necessary that we should learn as precisely as possible what happens to the cells composing the different tissues when a break is induced in metabolism through specific starvation for a single one of these, when no other nutritive principle is involved in bringing about metabolic disaster. Research in this direction, which must be achieved through the cooperation of the biochemist and the histologist with knowledge of pathology, may be expected to prove highly fruitful. As yet, however, little effort has been directed to this undertaking. Many attractive chemical problems of a specific nature have held the attention of the chemists in this new field. These could be studied through observing the effects on growth, fertility, nervous symptoms, longevity, or the development of ophthalmia, scurvy or beri-beri. The criteria for the interpretation of the experimental results were such as not to require the services of a pathologist. The pathologist was, however, unable satisfactorily to comprehend the theory underlying the preparation of diets for inducing specific conditions profitable for study, and could not, even if he desired to do so, accomplish anything in this field without the aid of the experienced physiological chemist. With efficient cooperation between specialists in these lines we may confidently expect new and surprising relations to be brought out in future.

Wason (28) has recently studied the pathology of ophthalmia of dietary origin, but was not able to secure any fundamental data concerning the etiology of this condition. She observed in the eyes of rats in which the disease was induced by selective fasting for fat-soluble A, hyalinization or necrosis of the outer layer of corneal epithelium, exudation of serum and cells into epithelium and stroma, and a proliferation of blood vessels and fibroblasts. In advanced cases invariably the anterior and occasionally the posterior chamber were invaded. "The type and virulence of the organisms of secondary infection determine, in part at least, the course of the disease."

Cramer (29) reported an interesting study of the relation of fasting for fat-soluble A to changes in the histological structure in certain adipose tissue, which he designates as "glandular." Several histologists have recognized that adipose tissue in certain parts of the body, especially the subpleural tissue, the neck and interscapular region, the axillae and around the kidneys, presents a pink and lobulated appearance resembling a fresh pancreas, and differs markedly, therefore, from ordinary adipose tissue and can be recognized macroscopically. This tissue is histogenetically distinct from ordinary adipose tissue. It has been described under a variety of names such as "primitive fat organ," "fat gland," "hibernating gland," "brown fat" and "interscapular fat." This fat has in the embryo a characteristic gland-like structure, and in certain species, as the white rat, domestic mouse and the hibernating animals, retains this structure throughout life. In most species, according to Cramer, soon after birth the tissue acquires the appearance of ordinary adipose tissue.

This "glandular" adipose tissue is very vascular and is distinct from the ordinary form in its function. Many of its cells are polygonal, present a reticulated appearance, and contain numerous small globules of lipins surrounding a nucleus, in contrast to ordinary fat cells which are oval, and generally contain one large fat globule engorging the cell and pushing the nucleus and protoplasm to one side. The "glandular" adipose tissue appears to contain a larger proportion of anisotropic or doubly refracting lipins than ordinary fatty tissue. The cells of the former are much richer in protoplasm than the common type of adipose cells.

During fasting the "glandular" type tends to persist and shows no obvious diminution whereas the ordinary type of adipose tissue is rapidly and greatly diminished. In cretins there are circumscribed swellings of the "glandular" adipose tissue in the neck and axillae, and there appears to be an hypertrophy of this tissue when the thyroid gland atrophies, suggesting a functional relationship between them. Cramer points out also that there is evidence of such a relation between the "glandular" adipose and the adrenals. There appears to be a particularly close relationship between the lipins of this tissue and those of the adrenal cortex.

When rats are fed a diet of purified food substances, free from all vitamins, this "glandular" tissue loses all its fat and lipins and presents the appearance of a glandular organ. Under the same conditions there is a great reduction of the lipins of the cortex of the adrenals, but this never becomes entirely free from them unless death occurs. Under the influence of such a nutritive disturbance the ordinary adipose tissues lose all their fat and other lipins and revert to the connective tissue type of cells. Cramer suggests that the "glandular" adipose tissue is a reservoir of fat-soluble A, since the disappearance of the characteristic lipins from it and the adrenal cortex is followed by death.