Since experimental poisonings with almost all the auto-toxins show their selective affinity for the nervous system it is necessary to discuss more fully these effects and see how they contribute to the production of disease. In common with other parts of the body the nervous system is attacked not only by poisons brought to it by the blood and lymph streams from remote parts of the body, but it may be also affected by the products of its own inherent metabolism. Auto-toxins affect the nervous system in two ways, the first as furnishing toxic stimuli which act as involuntary impulses causing muscular spasms, vaso-motor changes and the like; the second as causing actual change in nerve tissue and degeneration of neurons. The first of these actions is very evident from the experimental data given in Part II, the second is shown by analogies with well known poisons and by the association of auto-toxins with diseased nerve tissue.

Nissl (36) has studied the nervous system as influenced by poisonous doses of morphin, strychnin, lead, mercury, alcohol and tetanin, and has shown that definite structural changes are caused in the large motor cells of the ventral horns of the cord in the rabbit. The tetanus toxin caused in two hours a swelling of the nucleolus and of the tigroid masses which were much paler than normal although their arrangement was not changed. Five days afterwards the nucleolus was very much swollen and there was observed a breaking up and intermingling of thé tigroid bodies, and in extremely severe cases of poisoning the nucleolus becomes deformed and crenated. In experimental phosphorous poisoning there was complete obscuring of cell structure the process going as far as atrophy, disintegration or even disappearance of the neurons. Tetanin has a predilection for the motor cells of the fifth nerve nucleus, the toxin of diphtheria for the pneumogastric, and from the investigations assembled in Part II it is evident that auto-toxins too have selective affinities. That the integrity of nerve tissue is dependent upon free circulation has been shown by the experiments of Ehrlich and Brieger, in which compression of the aorta of a rabbit, produced in one hour, necrosis of the gray matter and cells of the spinal cord.

The power of alcohol to cause degeneration and inflammation in the brain has been indisputably demonstrated by recent investigations, and Kremiansky (37) has produced internal convexity pachymeningitis in dogs by the experimental induction of chronic alcohol poisoning.

As yet we have no experimental poisonings with known auto-toxins which compare with the researches of Nissl in thoroughness, for thus far there have been no microscopic examinations of nervous tissues of the animal subjects of experiment.

It has been shown that in any considerable degen erative process of the nervous system, cholin may be detected in the blood and cerebro-spinal fluid. This substance, the properties of which have been discussed in Part II, is constantly being given off from the brain in exceedingly minute amounts, being formed by the hydration of the lecithin in myelin, the reaction of which has already been given.

In a state of health no appreciable amounts of cholin can be found in the fresh brain, but in paresis, beri beri, and degenerations of the central and peripheral nerves it gradually accumulates and from lumbar puncture of the living or directly after death sufficient cholin may be obtained to give distinct chemical and physiological reactions. In this way we see that cholin can not only originate in the intestine but also in nervous tissue by its decomposition. If by future experiment it is proved that cholin is the cause and not the result of nerve degeneration, we have established the intrinsic origin of nervous disease, a thing already made to seem probable by analogy, but lacking experimental proof, without which no scientific truth may become accepted.

The probability of the intrinsic origin of nerve degenerations is materially strengthened by recent investigations (60) in which it is found that acute delirious mania is almost invariably associated with fatty degeneration of the liver, resulting in imperfect oxidization of end-products, retaining poisons within the system, inducing the processes which form diamins, - thereby causing the nerve cells to be bathed in a poison impregnated fluid, thus irritating and destroying their cytoplasm.

Migrim is caused by a number of the auto-toxins. Uric acid, paraxanthin and other of the purins, cholin, neurin, butyric acid, carbon monoxid and dioxid, hydrogen sulphid, and the bile salts have all been proven causes by actual experiment. Cholin commonly acts as a cause when eggs are eaten, from the hydration of lecithin, and cases are common in which the omission of eggs from the dietary is followed by relief of chronic neuralgia and headache. Migrim may be accompanied by a marked rise of temperature from the irritation of the fever centre by the poison, and it is very often preceded or followed by severe gastro-intestinal disturbances the underlying cause of the whole.