If the blood, when extracted from the vein, is received into a vessel containing a solution of glauber’s salt, coagulation is prevented, as the fibrine remains dissolved, and by filtering the liquor so obtained, the serum and water pass off and the globules remain, mixed only with little of the salt. The globuline cannot, however, be separated from hematosine, except by acids, which, as described in the preparation of hematosine, then combine with the globuline. _Alteration of the Blood in disease._—The examination of the state of the blood in disease, although presenting important relations to pathology and to practice, has been conducted in a manner too disconnected and superficial to produce any satisfactory results. This branch of chemical pathology has, however, been taken up by the illustrious Andral and Gavaret, who have published the result of the analysis of the blood in three hundred and sixty cases of disease. Their researches have enabled them to recognize four classes of diseases, in which the composition of the blood is essentially altered, though in different ways. The first class presents as a constant alteration, an increase in the quantity of fibrine; it includes diseases remarkably different in their locality and form, but all belonging to the class of acute inflammations in some cases of morbid deposition, as in tubercle and cancer, a similar increase in the quantity of fibrine is found, but it may be doubted whether it be due to abnormal growth or to the inflammatory action which accompanies it. In the second class the fibrine remains stationary, or even diminishes in quantity, while the globules increase in proportion to the fibrine. The diseases which belong to this class, are, continued fevers without local inflammation, and some form of cerebral hemorrhages. _Cerebral Hemorrhages._—In the third class, the fibrine remaining unchanged, there is a remarkable diminution in the quantity of globules; of these diseases, chlorosis may be taken as an example, and in the fourth class it is no longer the fibrine or the globules which are the subject of the morbid change, but the quantity of the albumen in the serum is diminished; of this class of affections is Bright’s disease. It has been observed, that in cholera the blood becomes so thick as to arrest the circulation, and contains from thirty to forty-five per cent. of solid matter; it is then, also, less strongly alkaline than healthy blood; this is connected probably with the matters vomited and evacuated, which are strongly alkaline, and contain a quantity of albumen. The blood has been found, occasionally, in cases of Diabetes Mellitus, to contain traces of sugar. The great discordance of the results obtained, may result, perhaps, from the sugar contained in the blood only for a short time after meals, and then being rapidly evacuated by the kidneys. In the jaundice the green coloring matter of the bile has been found mixed with the blood.