descendens was supported by Kussmaul,[31] in the record of a case where disease of the bladder was complicated with pelvic inflammation, atheromatous degeneration of the arteries, and consequent fatty degeneration of the sciatic nerves, causing direct paraplegia. We return to the centripetal theory of urinary paralysis with Leyden's own cases, published in 1865; of three patients with urinary paraplegia, two died, and the existence of a secondary (centripetal) myelitis seems to have been established, and by all analogy it must have existed in the third case, which recovered. The only puzzle and doubt that ensued was caused by the fact that there was an absence of neuritis in the different nerves themselves; though it seemed plain that the starting point of the myelitis was at the entrance of these nerves into the cord. This mystery seemed to be cleared up by the important experiments of Tiesler, ("Ueber Neuritis" Konigsberg, 1860) a pupil of Leyden's. This observer excited local traumatic inflammation in the sciatic nerve of rabbits and dogs; the rabbit became paraplegic and died three days afterward. At the site of the artificial irritation there was a localized formation of pus, and there was a second similar formation within the vertebral canal at the point where the posterior roots of the sciatic enter the cord; but there was no neuritis of the intervening portion of the nerve. Upon this and similar evidence is based the modern doctrine of a neuritis migrans, with centripetal tendencies, upon which it is supposed that a very large proportion, at least, of the urinary, dysenteric, and uterine paraplegias, miscalled "reflex," depend; and it is clear that the application of the word "reflex" in such a case is a grave abuse, tending to produce such confusion of thought and error in practice. In relation to the subject of our own inquiry--neuralgia--it is obviously of the highest consequence to investigate the question whether peripheral irritations, analogous to those which produce urinary paraplegia, are at all frequently the cause of the changes in the posterior roots which produce true neuralgia; for of course an inflammation may be the beginning of an atrophy which may presently exhibit no distinction whatever from one of which the origin was altogether non-inflammatory. I think that there is strong reason for thinking that this is not at all frequently the case. In the first place, all the evidence that exists respecting these centripetal inflammations of the cord is opposed to the idea that, save in the rarest instances, the inflammatory process limits itself to one small segment of the cord. Secondly, the description of the pains that have usually accompanied such inflammations of the cord is considerably different from the strictly localized, frankly intermittent character of a true neuralgia; in fact, all we know of the history of myelitis (except when complicated with a large amount of meningitis) forbids us