ment as a quantitative test, the greatest exactitude is, of course, required in its preparation.

A test liquid may be conveniently prepared, where a small quantity only is required, by pounding together five grains of sulphate of copper and ten grains of tartrate (neutral) of potash, and dissolving in two drachms of liquor potassæ. A clear deep blue liquid is formed, which answers quite as well for the detection of sugar as any other form of copper solution.

It is necessary to be aware of the fact that all the copper solutions are liable, after having been kept for some time, and especially when kept exposed to the light, to allow a slight reduction to occur on full boiling without any sugar being present. Some change must evidently have taken place in the fluid itself. Probably the amount of free alkali has been diminished by being converted into a carbonate, for the addition of free alkali restores the liquid to a proper state. It is only a slight deposit that occurs from this circumstance, one that could not be mistaken for the reaction of strictly diabetic urine, although it might lead to the inference that a trace of sugar existed. The state of the test can very easily be ascertained by boiling a little of it in a test tube alone. As a matter of precaution, this, after the fluid has been made beyond a few weeks, should be occasionally done, in order to see that it is in a proper state. Should any change be noticed on its being boiled, the addition of a fragment of caustic potash will render it as fit as ever again for use.

Sugar is not quite the only organic principle by which the copper test is reduced. Its reaction, therefore, cannot be taken as affording an absolutely positive indication that sugar is present. It is stated that

glycerine, tannine, cellulose, leucine, uric acid, and chloroform, are capable of exerting more or less reducing action upon it. Chloroform certainly produces a strong effect. Uric acid I have found occasion a slight deposit of suboxide. Cotton (cellulose) I have also seen give rise to just a trace of precipitate, but with glycerine I have not noticed that any reaction whatever has occurred. Glycerine, indeed, has appeared to me to be an agent that might be not inappropriately used to take the place of the tartrate of potash in the preparation of the solution.

Although from what has just been said the reaction of the copper test is not to be looked upon as affording unequivocal evidence of the presence of sugar, still in actual practice it is found to be one of much value, and one with a little precaution that is scarcely likely to mislead. Indeed, certainly where small quantities of sugar are in question, my experience would incline me to place more reliance upon the behaviour of the copper test than on that of any other. There is no test, it must be said, that can be spoken of as affording an absolutely indisputable indication, for even the fermentation test, as will be seen further on, has its fallacies as well as the others.

Another circumstance that it is necessary to be aware of in the employment of the copper tests is the fact that under certain conditions there may be no fall of suboxide, notwithstanding that sugar may be indisputably present. Ammoniacal salts have the property of occasioning this result. Ammonia and its salts, it is found, possess a solvent power over the suboxide of copper, and hence when they happen to be present no precipitate is produced by the action of sugar. sugar. The

CONDITIONS INTERFERING WITH REACTION OF COPPER TESTS. 17

combined ammonia contained in urine is often sufficient to obscure at first the presence of sugar when the quantity does not happen to be large. I have often noticed on examining urine-I do not mean diabetic urine, for this, unless albumen be present, always gives a neat reaction at once-that no change has at first resulted, whilst after boiling for some little time a copious amount of suboxide has fallen. Ammonia has been recognised escaping from the test tube, the potash contained in the test liquid having set it free for it to be driven off by the heat employed. After a little boiling the test has changed to a yellow or brownish-yellow colour, but without any deposit being produced, there being still sufficient ammonia present to hold the suboxide in solution. By further boiling the remaining ammonia has been expelled, and then the

precipitate has appeared.

The presence of albuminous matter also interferes with the reaction of the copper test; and it is probably to the generation of ammonia, as a product of decomposition produced by the influence of the alkali contained in the test, that the effect is attributable. I have known the existence of albumen conceal the presence of sugar in urine where even a considerable quantity has existed. Should a negative result be obtained on testing a liquid containing albumen, the albumen must be separated and the liquid again examined before it can be pronounced to be free from sugar.

In the case of urine the separation of albumen may be effected with the greatest facility, all that is required being to submit the specimen to boiling and filtration. With blood, however, and other animal

products, it does not happen that all the albumen can be got rid of so easily. Unless, however, it has been entirely separated no reliance can be placed upon the result.

To prepare a specimen of blood for testing, one of the following processes may be adopted.

It is on account of its alkalinity that boiling does not suffice to precipitate all the albuminous matter from blood and leave a suitable liquid for testing. By neutralizing its alkalinity and then boiling, a clear and colourless liquid is obtained on filtration. For this purpose acetic acid may be cautiously added until the exact point of neutralization is obtained. Care is required not to exceed the point, for an excess, even a slight excess, of the acid operates in the same way as the original alkalinity. Being a solvent of albuminous matter, the presence of acetic acid in a free state gives rise to a condition identical in effect with that which its employment was intended to remove. On account of this circumstance the neutralization process is found a little troublesome in practice.

A more easily applied process is one which consists in boiling the blood with sulphate of soda. A small quantity, say about half an ounce, of blood is placed in a small porcelain capsule, and treated with about its own bulk of crystallized sulphate of soda reduced to small fragments or to a powder. Heat being then applied, and the contents of the capsule briskly boiled, it will be found that a perfectly limpid colourless liquid is readily obtainable by filtration. The sulphate of soda, with the employment of heat, causes the precipitation of all the albuminous and colouring matters, and the only circumstance to attend to is that

enough is used to secure that this is completely effected. The presence of sulphate of soda in the filtered liquid in no way interferes with the subsequent application of the copper test.

The use of animal charcoal has also been recommended for preparing a specimen of blood for testing. It is well known to enjoy an extensive absorbent power; but, according to Bernard, glucose forms a principle that it does not take up. Hence by its means, according to the same authority, albumen and colouring matter may be removed from blood; albumen and uric acid, from urine; and caseine and fatty matter, from milk; leaving whatever sugar may be present to pass through the filter and be contained in the filtrate. I have several times tried this process, but my experience does not lead me to recommend it for use.

For examining a solid material, such as the substance of the liver, &c., a suitable liquid may be obtained for testing, either by making a plain decoction of it; or else, by pounding it in a mortar with nearly an equal bulk of sulphate of soda, and then heating and filtering in the same manner as was recommended for blood.

The copper test is a very delicate one, and will display the presence of minute quantities of sugar in urine without any preliminary preparation. Where, however, an exceedingly minute quantity is being looked for the urine may be concentrated by evaporation, and treated with an excess of acetate of lead to get rid of colouring and other solid matters in the form of precipitate. The lead contained in the filtrate, belonging to the excess of acetate employed, is best