be indulged in with impunity. Thus, I have known the use of a few saccharine pears undo, in a few hours, all that I had been labouring for months to accomplish; and the disease, with all its horrors, has become re-established in an aggravated form.' It would be superfluous to adduce further evidence in confirmation of the testimony of Dr. Prout on this all-important practical question, inasmuch as it is in perfect accordance with the common experience of those whose attention is much directed to the treatment of diabetes.

The dietetic treatment introduced by Dr. Rollo consists in restricting the patient to a purely animal food (all vegetable matter being excluded) comprising meat-fat beef, mutton, pork, and game fish, oysters, light boiled eggs; and a mixture of milk, and beef or mutton decoction, and water for drink. The meat to be cooked, but no seasonings allowed except a little salt. Dr. Rollo strictly enjoined the importance of restricting the animal food taken to a certain limited quantity. With regard to meat, however, he recommended that it should be as fat as possible. He also pointed out the necessity of continuing this regimen for a considerable period after the disease had been removed, in order to subdue the marked tendency it manifests to return for

a long period afterwards, if the patient is guilty of indiscretions in his diet, or of irregularities in his mode of living. It is necessary to add that Dr. Rollo recommended the use of alkalis in addition to the regimen prescribed.

In making a change from a purely animal regimen, Dr. Rollo recognised the necessity of excluding such vegetable substances as are known to contain starch; he considered it the safest course to pursue to allow brocoli, spinach, cauliflower, cabbage, and lettuce. These,' he observed, 'do not seem to furnish sugar when prudently used in the diabetic stomach, after a proper adoption of animal diet.'

6

Dr. Prout strenuously advocated the essential, practical importance of Dr. Rollo's treatment, but modified his system so far as to recommend the green portions of vegetables (on account of their not containing starch or sugar) to be taken from the commencement of the treatment in addition to animal food, of the exclusive use of which he did not approve. In preference to other kinds of animal food Dr. Prout recommended beef and mutton, plainly cooked, especially mutton chop or beef steak lightly done, taken twice daily. The other meals to consist of such simple dishes as can be prepared from eggs, milk, butter, and the like animal substances.

Dr. Prout, too, strongly insisted that the quantity of food taken daily, and at each meal, is a point of equally as great importance, in the treatment of diabetes, as its quality; and, in illustration of this, points out that the frequent occurrence of cases of sudden death in the disease has been clearly traced to errors in the quantity as well as in the quality of the food taken; the patient having been usually cut off after a hearty meal, as it is vulgarly termed.' Consequently he recommended a strict regimen, adjusted to the peculiarities of each case, to be rigidly adhered to, and a meal taken every four, five or six hours; and that all fluids should, as far as possible, be abstained from at each meal of solid food, and for an hour or two afterwards.

Since the time of Rollo no other important modifications deserving of special notice have been made in his treatment of diabetes, with the exception that Bouchardat has interdicted the use of milk altogether; but he has recommended that cream should be taken; and it appears that his opinion on this subject has been adopted by some more recent writers, and, amongst others, by no less authorities than the late Professor Niemeyer of Tubingen and Dr. Pavy.

1

1 Éléments de Patholog. et de Thérapeutiq., tome ii. p. 886. Paris, 1866. French edition.

A serious objection to Dr. Rollo's regimen is the difficulty encountered in subjecting patients to it for a sufficient length of time. In some individuals, indeed, the loathing or disgust felt for an exclusive meat diet, under any form, is so great as to present an insuperable obstacle to putting the prescribed treatment in force; while in other instances again, although it may be borne with impunity for a certain period, yet, sooner or later, a dislike to it becomes engendered, and this increasing in intensity renders further perseverance utterly insupportable; so that the regimen must be abandoned, notwithstanding all that may be attempted by cookery or variety to render it more palatable. No one regretted more keenly this serious obstacle to the treatment than Dr. Rollo himself, who discovered that his patients, however well informed as to the serious consequences of indulgence, frequently violated the strict rules laid down for their observance, and concealed their clandestine transgressions to themselves.

It is unquestionable that the regimen just described has hitherto been the only treatment found capable of exercising any control over diabetes, and that when rigorously pursued it has been of very great utility. In some cases, indeed, it has undoubtedly cured the disease; while in other and more

numerous instances it has, without effecting a cure, removed its worst symptoms, and held it in check; so that patients have continued to enjoy for several years, moderately good health, who otherwise would probably have died in a few months after extreme suffering.

On making a careful investigation of the cases in which a diet consisting essentially of meat has been either completely or partially successful, it will be found that the instances in which complete success was obtained, as in the case of Captain Meredith recorded by Dr. Rollo, the disease had not advanced beyond its first stage when the whole, or nearly the whole, of the urine sugar is formed from the starchy or saccharine principles of the food: none, or very little of it being derived from the albuminous alimentary compounds.

But those cases of partial success, on the other hand, and they have been the most numerous, in which the disease was not removed, but only held in check and its symptoms subdued by the treatment, the urine still continuing to be more or less saccharine, are instances in which the affection had passed into the second, or confirmed, stage, in which there is a considerable saccharine transformation of the albuminous food. Indeed, it would appear that such