The histological and serological changes in animals made experimentally hypersensitive to foreign proteins have been separately studied by many investigators. It does not fall within the scope of this report to review the extensive literature dealing with these subjects. It is our purpose to attempt to correlate the sequences in the development of definite tissue changes with concomitant serological studies of the levels of antigen and antibody. The demonstration by Rich and Gregory (1), that a single large injection of heterologous serum may produce marked vascular lesions in rabbits not sensitized previously, has provided an experimental technique adaptable to such studies.Hektoen and Cole (2), Cannon and Marshall (3), and others have pointed out the importance of using only highly purified and well characterized proteins in studies on the fundamental nature of immune reactions. The methods for the fractionation of plasma proteins recently developed in the Department of Physical Chemistry of the Harvard Medical School (4-10) have made available many of the proteins of both human and bovine blood plasma in states of greater purity and in larger amounts than could readily have been obtained by previous methods. The use of these materials has provided a new approach to studies of the reactions of hypersensitivity as well as to many other problems in biology and in clinical medicine. Research and Development. The molecular weight, as determined by osmometric methods and by analysis in the ultracentrifuge was 69,000. The albumin was electrophoreticaUy pure and impurities which could be detected by serologic methods were less than 0.05 per cent. The albumin was used in 25 per cent solution in water especially distilled for intravenous use, to which was added sodium bicarbonate (0.375 gln. per 100 cc. solution) and sodium chloride Martials

PLAZAS 21 TO 23In recent ycars studies of the clotting of blood have resulted in the chemical identification and isolation of some of the factors involved, and these studies together with clinical observations have contributcd to an as yet incomplcte description of the phenomenon. Investigations on both the morphological and chemical aspects of the problem have been greatly facilitated by the quantities of relatively pure materials made available through methods developed by Cohn and his associates (I--4).In contrast to the many physiological and chemical contributions to our understanding of the clotting process, there have been few recently that could be classed as morphological. Howell (5), in 1914, made use of the ultramicroscope to observe the conversion of fibrinogcn to fibrin. He described the end result as "a formation of a meshwork of beautiful needles," and suggested that the matter of greatest interest was to attempt to follow the actual process of formation of these needles. However, he commented that it was in the background of particles beyond the vision of the ultramicroscope that the "needles" or "crystals" of fibrin took their form.The resolving power of the electron microscope allows observations to be made in this previously inaccessible size range. In an earlier report (6) the authors described the fine structure of clots formed from purified bovine fibrinogcn and thrombin. The techniques used in that study seemed also to be applicable to the study of sequences in the conversion of fibrinogen to fibrin by the action of thrombin. Materials and MetkodsInasmuch as the present series of experiments required certain alterations in the previously described techniques, repetition and amplification of some of the details seem necessary.

Left auricular appendages from 183 patients, removed at operation for mitral stenosis, have been studied with regard to presence of Aschoff bodies and endocardial thrombi. An over-all incidence of Aschoff lesions of 40 per cent was found. There was a significant decrease in Aschoff lesions in the pIesence of thrombosis. The occurrence of Aschoff bodies in the appendages is correlated with the findings in the remainder of the heart in autopsied cases.T HE DEVELOPMENT of methods for the surgical correction of rheumatic deformities of the mitral valve has afforded an opportunity for the examination of the left auricular appendage which is removed at the time of operation. It is thus possible to study cardiac tissue in living patients and to correlate simultaneously clinical and pathologic observations.The microscopic anatomy of the auricle has been described by Gross.' The auricular appendage is similar to the auricle, differing chiefly in that the appendage has a thinner wall and presents many irregular pockets and trabeculations. Rheumatic endocarditis involving the mural endocardium of the left auricle is a well-recognized entity,1-4 but Aschoff lesions in the auricular appendage have not been described except ill connection with studies similar to the present one, although Von Glahn stated that superficial auriculitis which

Recently, the histologic and serologic responses in the rabbit to single, large intravenous injections of purified bovine serum proteins have been described (1). These responses were: (1) To bovine serum gamma globulin--The development in most animals of a transient intracapillary glomerulonephritis, together with focal lesions of the liver, heart, and joints, the acute phase occurring from 1 to 2 weeks following injection; (2) to crystallized bovine serum albumin--The development in only about half of the rabbits of an intimal and subintimal arteritis of the large pulmonary and systemic arteries, the acute lesions being seen 2 to 3 weeks after injection. When the antigen was gamma globulin, it regularly disappeared from the circulation within 8 to 14 days following injection, and homologous antibody appeared shortly thereafter. In contrast, the albumin disappeared from the circulation after 14 to 21 days and then only in some of the rabbits. In others, it was still detectable after 4 weeks.From this correlation between the times of disappearance of the antigens, of the emergence of their antibodies, and of the development of lesions, it was deduced that the latter were dependent upon a reaction between antigen and antibody. Furthermore, it was postulated that the difference in distribution of the lesions depended upon a difference in localization of the two proteins used as antigens, since their molecules differ in physical and chemical as well * This is one of a series of clinical, pathological, and immunological studies on the proteins of blood and tissues, using proteins prepared by methods developed in the Department of Physical

No abstract