The various complications that may arise, as well as the general results of purpura, must be treated symptomatically. For the mucous membranes astringent washes should be used, and in favorable situations the tampon may sometimes be employed with profit. In purpura rheumatica the arthritic pains will be alleviated by anodyne liniments and plasters, and the often accompanying abdominal pains and colic by anodynes internally administered. Hæmatemesis, hæmaturia, etc. must be treated upon general principles. The results of profuse hemorrhage must be combated with stimulants. Transfusion of blood has been proposed and practised for the extreme anæmia that sometimes occurs, but without encouraging results. If necessary, the bowels may be kept free by mild aperients. In severe cases rest in bed should be rigidly enforced until after the establishment of convalescence. Quinia, iron, and nux vomica are indicated above all other remedies for the anæmia resulting from an attack of purpura.
Diabetes mellitus is a term applied to a group of symptoms more or less complex, of which the most conspicuous is an increased flow of saccharine urine--whence the symptomatic title. It is associated with a derangement of the sugar-assimilating office of the liver, as the result of which an abnormally large quantity of glucose is passed into the hepatic vein and thence into the systemic blood, from which it is secreted by the kidneys. The condition is sometimes associated with alterations in the nervous system, at others with changes in the liver or pancreas, while at others, still, it is impossible to discover any structural alterations accompanying it.
[Illustration: FIG. 1. To show the position of the punctures required to produce glycosuria, the lobes of the cerebellum are separated. Below are seen the restiform bodies, the divergence of which circumscribes the apex of the calamus scriptorius and the fourth ventricle. The puncture _p'_ produces glycosuria; the puncture _p_, glycosuria with polyuria; and a puncture a little higher up than _p_, albuminuria.]
PATHOLOGY AND PATHOGENESIS.--Notwithstanding that this disease has been recognized for two centuries and a half, that abundant opportunity has been furnished for its post-mortem investigation, and that experimental physiology has contributed much information bearing upon the subject, its pathology is still undetermined. Experiment has, however, rendered it very likely that all cases of essential glycosuria--that is, all cases in which saccharine urine is not the direct result of over-ingestion of sugar or sugar-producing food--are accompanied by a hyperæmia of the liver. This hyperæmia, with its consequent glycosuria, can be induced by puncturing or irritating the so-called diabetic area[1] in the medulla oblongata. This area corresponds with the vaso-motor centre, and with the roots of the pneumogastric or vagus nerve in the floor of the fourth ventricle; whence it was at first inferred that this nerve is the excitor nerve of glycosuria. It was soon ascertained, however, that when the pneumogastric was cut, glycosuria ensued only when the central end was stimulated, while { 196} stimulation of the peripheral portion was without effect. Whence it became evident that this nerve is not the excitor, but the sensory nerve concerned in glycogenesis.