For a solid organ, such as the liver, the method of procedure to prepare for testing, may be; either, to make a plain decoction of it, or else to pound it in a mortar with nearly an equal bulk of the sulphate of soda, and then heat and filter in the same manner as with blood.

Animal charcoal has been more lately recommended for separating sugar from some other materials. The absorbent power of animal charcoal is a well-known property; but, according to Bernard,* glucose is a substance that it does not take up. Hence (still following the same authority), although it removes albumen and colouring matter from blood, albumen and uric acid from urine, and even caseine and fatty matter from milk, yet, sugar present in either case will be left and pass with the liquid through the filter. I have several times tried this process upon blood, but have no hesitation in giving a decided preference to the old plan with the sulphate of soda.

Although a point of doubt might exist in the minds of some concerning the reliance to be placed upon the indications afforded by the tests I have mentioned, yet the fermentation test has been hitherto generally looked upon. as perfectly characterising by its reaction the presence of sugar. It seems, however, from the researches of a French chemist, M. Berthelot, who has recently devoted much attention to the chemistry of the sugars, and whose statements are entitled to the greatest consideration, that there are other substances besides sugar-viz., glycerine, mannite, dulcine, and sorbine, which are capable of undergoing the alcoholic fermentation with yeast. Alcoholic fermentation, therefore, can no longer be regarded as affording an infallible indication of sugar.

For the application of the fermentation test the process is exceedingly simple. A contrivance such as is repre* Leçons de Physiologie, Cours du semestre d'hiver 1854-55,' p. 45.

sented at Plate, fig. 3, will be found the most convenient. The urine or liquid to be examined is mixed with yeast and placed in the test tube, which ought to be filled to the top. The bent tube is then introduced, and the cork tightly adapted. The end of the bent tube should nearly reach the bottom of the test tube, so that as the gas, resulting from fermentation, is generated and rises in the test tube, it may cause, by its pressure, an expulsion of the liquid, which is to be received into a glass suitably placed for the purpose of collecting it. The test tube is immersed in a vessel of tepid water, and placed in a warm situation, as before a fire; or else, the water should be renewed as occasion may require. By this application of warmth, when grape sugar is present, fermentation in a few minutes commences briskly to take place, and the generated gas, rising in the test tube, displaces the liquid, which becomes transferred to the glass.

For ordinary purposes the evolution of gas may be taken as a proof of fermentation; but, should it be considered to require confirmation, the actual production of alcohol and carbonic acid may, in the following manner, be easily established.

The test tube, which has been gradually emptied of its contents and is supposed to be filled with carbonic acid gas, is inverted and the cork removed under water. A small fragment of potash being introduced, the thumb is applied to the extremity to close it, and agitation freely performed for a minute. The carbonic acid is absorbed by the potash and a vacuum formed, which is shown by again inverting the test tube, and, with its mouth under water, removing the thumb. The water instantly ascends and occupies the space of the carbonic acid that has been absorbed.

To recognise the alcohol, the liquid which has been dis

Or the varieties of sugar, glucose or grape sugar is that which is encountered in investigations upon the subject of diabetes. As a preliminary step, it is of the first importance that the means adopted for recognising our agent should be agreed upon as satisfactory. Happily, we do not here, at the onset, encounter any difficulty. Such strikingly characteristic, and easily recognisable properties are possessed by glucose, that it is almost as readily susceptible of detection as any inorganic material; and this, even, when present to an exceedingly minute extent. Numerous tests have been from time to time recommended, but, it is only to those, that have been employed in my investigations, that I consider it necessary in the following pages to refer. These are the alkali, the cupro-potassic, and the fermentation tests.

physiological investigations, however, it may be necessary to seek for cane sugar, which is not sensitive to either form of the copper test. To render it so, it must first be converted into glucose; and this may be done by boiling the specimen supposed to contain cane sugar for a short time with a small quantity of sulphuric acid. The acid employed being neutralized, the copper test will indicate if any glucose have been formed.

For the quantitative determination of sugar, the process I always adopt is with the cupro-potassic solution; and this, my experience enables me to speak of as being susceptible of very great delicacy. We do not separate and weigh the sugar, as the chemist does with an inorganic material; but, on the other hand, estimate its amount by its extent of reducing or deoxidizing effect on a copper solution of standard strength. The copper solution I employ is that composed of the tartrate of potash, potash, and sulphate of copper, the precise composition of which is to be found at page 5. This liquid is of such a strength that 100 minims of it are decolorized by half a grain of grape sugar-in other words, half a grain is the exact amount of sugar required to convert the whole of the oxide of copper into the state of sub-oxide that is contained in 100 minims of the liquid. The sugar from which this determination of strength was drawn, was specially sought after to procure as pure a specimen as possible. It was obtained from Mr. Morson, and was thoroughly deprived of water by drying in a steam oven before being used.

In describing how the quantitative analysis is performed, I will give an illustration of the method of procedure I adopt in the case of a specimen of diabetic urine.

In an ordinary case of diabetes the urine is too concentrated with sugar to operate upon alone, so it is diluted with four times (this I have found about the most con

venient extent of dilution) its bulk of water. For the process of dilution, I use a tall, narrow, graduated glass, such as is represented at Plate, fig. 4. It is divided into 100 measures; and I therefore fill up to 80 with distilled water, and then to 100 with the urine. The mixture is emptied into a vessel, and poured backwards and forwards once or twice, so that the two may become thoroughly mixed together.

One hundred minims of blue liquid are now taken and placed in a small porcelain capsule, with a fragment of solid caustic potash about double the size of a pea. This addition of potash causes the precipitate to fall in a denser form, and thus leaves the colour of the liquid more easily seen. The capsule is placed over the flame of a spirit lamp on a retort-stand, or what is better, on a piece of iron-wire gauze, fitting on the top of an earthenware cylinder, as at Plate, fig. 1. The cylinder protects the flame of the spirit lamp from draught, and the gauze distributes the heat and renders the boiling much more steady. The liquid in the capsule is to be made to boil gently, and then, the diluted urine is dropped into it slowly from a graduated tube, until the blue colour is entirely removed. The contents of the capsule having been kept steadily boiling all the time, as soon as the blue colour has disappeared, the amount of diluted urine employed is to be read off from the graduated scale of the tube.

Usually when the operation is slowly and properly performed, the red oxide produced, collects at the bottom of the capsule, leaving the liquid above clear, the colour of which may be gradually observed to fade, and finally disappear. At other times, the reduced oxide is more or less distributed through the liquid, but even then, it is not difficult, after a little experience, to decide when the point required, is attained. All trace of blue or green should be just,