A study of collagen was initiated several years ago in relation to work on the pathogenesis of rheumatic fever since it is generally believed that the primary injury in this disease is to collagen, the ground substance, or both (1-5). The chemistry and structure of collagen have been studied extensively (6--19) but little is known of its biological properties. 1 The studies to be reported below relate primarily to the antigenicity of collagen. Review of Previous WorkNageotte (20,21) showed that when the tail tendons of the rat are placed in dilute acetic acid (1:25,000 dilution in distilled water) the collagen fibers slowly swell and over a period of time go into solution. He also demonstrated that fibers reconstituted from this acid solution by the addition of sodium chloride retain the tinctorial properties of native collagen. Wyckoff and Corey (22) demonstrated that the x-ray diffraction pattern of this reconstituted material was like that of the native tendon. Schmitt, Hall, and Jakus (9) and others (23, 24) have found that fibers reformed from dilute acid solutions either by neutralization or the addition of salt retain the characteristic striations of native collagen as observed in electron micrographs.It is generally agreed that gelatin, a soluble derivative of collagen, is non-antigenic (25, 26). Loiseleur and Urbain (27) reported complement-fixing antibodies in the sera of rabbits immunized with an acid solution of rat tail tendons prepared as described by Nageotte. Hopps (28) demonstrated antibodies to catgut and sheep intestine; however, there were cross-reactions with sheep serum. With highly purified preparations of sheep collagen obtained from Dr. F. O. Schmitt he reports agglutination with anticatgut rabbit serum in a dilution of 1:80; however, another sample of the highest

No abstract

Electron microscope studies of collagen have dealt, for the most part, with the morphology of the mature or completely formed fibers.' It has been shown that fibers from a variety of sources have the same characteristic periodic structure which consists of regularly spaced bands of greater density to electrons, arranged crosswise to the long axis of the fibers. The combination of a dense (A band) and a light region (B band), which constitutes a single macroperiod, measures about 640 A (1, 2). Metal shadowing reveals that, in the dried fiber, the dense bands represent elevations in the contour and therefore regions containing greater amounts of non-volatile material, whereas the fight bands coincide with depressions (3). Studies at higher resolution reveal further complexity. The more opaque band is found to have two component striae and the more transparent band or depression, a single striation. The number of these intraperiod striations is usually greater in fibers "stained" with phosphotungstic acid but it may be that these are phases or regions within the organization that specifically fix the tungsten compound (4). In any case the over-all picture is one of extremely precise and complicated organization and this poses such questions of morphogenesis as (a) what is the form of the fundamental, possibly macromolecular, units which make up the fibers, and (b) how do these units combine to give the now familiar, striated fiber.Some information relative to these problems was obtained from a study of the natural formation of collagen as it occurs in cultures of collagen-containing tissues (5). It was observed that the smaller natural fibers start out with bands of more or less uniform size, spaced at 210 to 270 A, and that only as these fibers grow to greater diameters do their striae acquire the inequalities which account for the macroperiod of 640 A. However, the conditions of this study did not lend themselves to a demonstration of the means by which the fibers grew in size. While there was some indication that the narrower fibers fused to