Until recently the development of edema, such as is commonly associated with kidney disease, was ascribed as a rule to changes in the permeability of the blood vessels of the body. One hypothesis, for example, postulated an increased permeability of the capillaries secondary to toxic injury, expressing itself in the glomerular capillaries by albuminuria and anatomic renal disease, and in the cutaneous and serous capillaries by edema. The fluid thus segregated is rich in protein. Indeed, it is much like an inflammatory exudate. Experimentally an edema of this type can be produced by damage to the capillaries through use of a variety of irritants—so-called capillary poisons. An analogy between this sort of experimental edema and the dropsy of Bright's disease seems to have been established.
Recent years have furnished the demonstration that edema may arise without any renal, cardiac or capillary damage as the significant factors. Diets poor in protein