from those of an embolic process. The necrotic area which is characteristic of embolism is never observed in poliomyelitis.
3. Experimental embolism in the spinal cord does not substantiate Marie's hypothesis. Occasionally lesions restricted to the gray matter have been produced; whereas, usually the white matter is extensively implicated.
4. In ulcerative endocarditis, caisson disease, and other maladies in which demonstrable embolic processes do occur in the spinal cord, no preferential selection of the gray matter can be seen.
Proof has not yet been given that the changes in Heine-Medin's disease are referable to embolic or thrombotic processes; and, further, the pathology of the spinal cord comprises no analogous conditions to lend support to this view.
I have also emphasized facts—apart from the nature of the process—which seem to me to contraindicate any preponderating influence of the central artery:
1. In recent cases the most marked changes usually appear within, but never coincide with, the area of distribution of the central artery. Beyond this particular vascular zone, in the posterior horns, in the white matter, and in the pia, round-cell infiltration generally occurs. Clearly, since the central artery sends no branches either to the posterior horns or to the pia, the central artery cannot be implicated in lesions in these situations. Lesions of the white matter have been alleged to be distributed exactly within the area of supply of the central artery, but I believe otherwise. Kady's experiments show that in a transverse section of the spinal cord, three vascular zones can be distinguished: a peripheral zone, which is supplied exclusively by the peripheral artery; a central, limited to the gray matter, supplied by the central artery; and an intermediate, which is supplied from each side in such a way that the capillaries of both systems intermingle. If the central artery determined the local distribution of the disease, one would expect that the lesions would radiate from the gray, far into the white, matter; and that they would appear wedge-shaped on cross section. Such lesions, however, have never been observed. The morbid changes commonly cease precisely at the line of demarcation between the gray and white matter; and when, as occasionally happens, the disease encroaches further, it extends
