Thus it is clear that lactic acid has the properties of a rheumatism-producing poison; but that in the human subject it is the veritable cause in every or any case of rheumatism, is only an inference. As yet, there are wanting many experiments to make the matter a demonstration. But this seems to me clearly proved; that if not lactic acid, yet some acid of analogous character is always the cause. It has yet to be learned by experiments what formic, acetic, lithic, and butyric acids will do under the same circumstances.

Again, if by after experiment it should be proved that various acids have the same influence, or if it should be proved that the lactic is the one simple representative of this class of special poisons, the experimental argument will have to move back another step towards the primary cause of the disease. The question will next be, Where in the body, and in what manner, is the poison generated?

We are not without a flickering light calling us towards the way in which to pursue this latter inquiry. To understand better the point of origin, is to understand better the process of oxygenation in animal bodies.

In some of my experiments, where animals were subjected many hours to an over-supply of oxygen, the general symptoms were those of an acute inflammatory attack. True it is, that as yet I have never obtained joint-affection by this means; but in one or two cases a vascular condition of the left endocardial surface, closely approaching to endocarditis in the first stage, was clearly

acid, or with a too concentrated dose. Its atphology was not, strictly speaking, the pathology of acute structural membranous inflammation, but rather conveyed the idea that the normal serous and plastic secretion had undergone direct coagulation from contact with the acid fluid. This view is supported in some measure by the facts that acute pain of the abdomen was never present, and that the absorption was least effective when the dose of acid was most concentrated.

present; and in the human subject, I believe endocardial mischief to be a more common complication than is generally conceived of pneumonia, which is the disease par excellence of hyperoxygenation.

It will be observed that, in the experiments with lactic acid, the right as well as the left side of the heart was sometimes affected. This was to be expected; for as the poison was thrown ready made into the body, it was taken up at once by the absorbing vessels, and carried into the right heart by the venous current, and afterwards by the pulmonary veins into the left heart, and so to the body at large through the arterial currents. The poison, therefore, of necessity passed through both cavities of the heart, and both were affected. But if by experiment, instead of injecting the acid ready made into the venous current, I could inject into that current an agent which, in the pulmonic circuit, and during the process of oxidation, should be transformed into lactic acid, then the inference is fair that the left side only of the heart would be affected.

In rheumatic endocarditis, occurring from the formation of the poison in the body itself, the left side only is affected as a general rule. Hence I infer that the chemical change whereby the materies morbi of acute rheumatism is produced, is completed in the pulmonic circuit; that in the respiratory act the acid quality of the poison is produced; that, thus formed, the poison is carried by the arterial circulation to be disposed of by decomposition or elimination, or both; and that it does not return as an acid by the veins, but simply as a product which admits of re-transformation in the pulmonic circuit into the acid state.

If this view be true, it is easy to connect further the

origin of the rheumatic disease with the digestive system, or with the arrest of special secretions. In either case,

the basis of the poison may be generated and carried into the returning venous current, thence into the pulmonic circuit, and finally, after oxidation, into the left heart and through the arterial system.

This, as far as inference as incidental to experiment is allowable, is the only explanation of the peculiar fact that in rheumatism the left side of the heart usually suffers alone, and always most severely.

Regarding the origin of lactic acid in rheumatism, a view has been advanced by Dr. Headland, to the effect that ordinarily the starch of the food is first converted into lactic acid, which afterwards is decomposed into carbonic acid and water, and so is excreted by the lungs; but that under conditions unfavourable to this series of changes, the lactic acid accumulates. This view, if it should eventually be proved that lactic acid is the unit cause of rheumatism, may offer an explanation of the argument now on hand; for on this hypothesis the starting point of the acid poison would be the pulmonic circuit.

The considerations here introduced lead, moreover, to two other deductions. The first is, that it is vain to look for the materies morbi of ordinary rheumatism in venous blood, but that it may be looked for with possible success in arterial blood. The second is, that the action of the poison in ordinary rheumatism is local, i.e. by direct application to the endocardial surface; for if the poison were conveyed by the blood to the membrane à posteriori, i. e. through the coronary circulation, both sides of the heart should suffer alike, which is contrary to the natural fact.

Returning to the experiments, and leaving the argument from inference altogether for matters of fact, there are certain truths, pathological and practical, taught by these experiments, which deserve consideration.

From the circumstance that the animals operated on either died, or were killed, at certain different and well marked phases of the disease, an opportunity was afforded of observing, perhaps for the first time, the course of endocarditis through each successive stage. I find from these observations that endocarditis has three well marked periods of progression.

1. The primary stage is one of congestion and oedema. In this stage (see Plate 1, p. 376), the endocardial surface is intensely vascular, approaching in colour to bright vermilion. The membrane has also a soft fleecy character, giving to it a velvety appearance. The curtains of the valves are in a swollen and vascular state, which my colleague Dr. Halford has very appropriately called an "oedematous condition". If the swollen valve be pricked with a needle, a clear lymph oozes from the puncture, and the valve collapses. In the heart depicted in Plate 1, the valve, several times larger than it is shown in the drawing, had lost its fulness before it reached the artist, in consequence of its having been punctured in this fashion. Occasionally during this stage, if the congestion and oedema is very great, there is a transudation of this lymph through the membrane on to its free or ventricular surface. When this occurs, there is laid at once the basis of a fibrinous deposit from the blood. In Experiment cccc, this occurrence was beautifully marked, and as a positive result its value. cannot be well over-estimated. For, as in the process here described, there is laid a basis for a fibrinous de

posit, so is there also a basis for an organised product; the connection between the deposited matter, and the matter effused on beneath the membrane, being direct.

While these changes are progressing in the curtains of the valves, around their margins, as shown in the drawing, small beads are abundantly developed. These primarily are ædematous points also: they yield a clear coagulable lymph when pricked or compressed, and shrink at once after the operation. In this stage the aortic valves are also vascular.

The general symptoms which mark this stage of the disease are characteristic. There are reduction of the animal temperature, distaste for food, shivering, and then continued feverish excitement, thirst, heat of skin, quick and short inspirations, and rapid sharp pulse. In this first stage, the heart's action is irritable; the first sound becomes muffled, and eventually is lost altogether. This condition marks the time at which the mitral or tricuspid valves are in the swollen or œdematous state; and this circumstance has special interest as proving the fact first experimentally shewn by Dr. Halford, that the first heart-sound is dependent on the tension of the auriculo-ventricular valves. During this stage also, the second sound is sharp and accented. Dr. Halford opines that this sharpness or accentuation of the sound is relative only, that is to say, is brought out more strongly from the loss of the first. To my own ear there is, however, always conveyed an intensity of the second sound; and considering that in all the cases where I could trace out the signs to their pathological origin, the aortic valves were red, and slightly thickened, the intensity here spoken of is accounted for. Lastly, this stage is marked by a short irritable cough, which