The presence of fat in the blood of diabetics was noted by the earliest students of the disease. It is sometimes sufficient in amount to produce a milky appearance of the serum, while the analyses of Simon revealed a quantity of 2 to 2.4 per cent., the normal being 1.6 to 1.9 per cent. The fat thus present is said to be sometimes sufficient to cause fat embolism in the capillaries of the lungs, and cases of this condition have been reported by Sanders and Hamilton,[23] Louis Starr,[24] and Rickards.[25] Ralfe ascribes the lactescent appearance of the blood to the action of the aceto-acetic acid, since acetic will give a milky appearance when agitated with a dilute and slightly alkaline mixture of fatty matter at 100°, and the injection of acids into the blood of animals leads to the increase of fatty matter in the blood and fatty infiltration of tissues. [Footnote 23: _Loc. cit._] [Footnote 24: _Loc. cit._] [Footnote 25: _Birmingham Med. Review_, Jan., 1882.] It must be admitted that the mode in which this lipæmic state of the blood is brought about is imperfectly understood, and whether it be by some chemical agency of the kind described by Ralfe, or by rapid absorption of the subcutaneous fat, or from an imperfect oxidation of absorbed fat, is undetermined. Possibly all may contribute. Albert G. Heyl[26] has described an altered appearance of the retinal vessels recognizable by the ophthalmoscope, which he ascribes to the fatty blood-plasma at the periphery of the blood-current, the normal plasma being invisible on account of its transparency. [Footnote 26: For a detailed description of this appearance, with a colored lithograph depicting it, see the author's work on _Bright's Disease and Diabetes_, p. 262.] The red blood-discs are diminished and their ratio to the white corpuscles altered. In a count by F. P. Henry, in Louis Starr's case, the number of red discs was 4,205,000 to a cubic millimeter, the normal being at least 5,000,000; the white were 50,000 to a cubic millimeter, or 1 white to 84 red, instead of 1 to 350 or 500. Changes in the Urine.--The most important changes in the urine are its increase in quantity and the presence of sugar. The variations in the former are extreme, being from an amount which but slightly exceeds the normal to as much as 50 pints (23.65 liters) in twenty-four hours, and even more. The quantity is of course limited by the fluid ingested, and although it may exceed this amount for a day or more, it cannot do so for any length of time. It is generally a little less. The more usual quantity in the twenty-four hours is from 70 to 100 ounces (210 to 300 cc.).