The quantity of sugar varies greatly in different cases and at different times in the same case. The maximum quantity reported by Dickinson was 50 ounces, or 1500 grammes, in twenty-four hours. The proportion may reach as much as 15 per cent., but the more usual amounts are from 1 to 8 per cent., or from 5 to 50 grains (.324 to 3.24 grams) to the fluidounce, or from 300 to 4000 grains (19.44 to 260 grams) in the twenty-four hours. It is important to know that intercurrent febrile disease may produce a decided diminution in the daily quantity of urine, and of the sugar contained in it. A similar decrease, and even disappearance, is said to take place sometimes toward the fatal termination of a case. {208} The effect of exercise upon the sugar secretion is not uniform. Bouchardat and Kuelz have noted a diminution, and even disappearance, of sugar from urine as its result, and it is reasonable to suppose that judicious exercise is at least without harmful effect, while it is certain too that muscular exercise, if excessive, will increase glycosuria. Changes in diet of course modify the secretion of sugar, starches and saccharine foods increasing it, while nitrogenous and oily foods diminish it. So, too, the urine secreted on rising in the morning has almost always less sugar in it than that passed on retiring; and it is not rare to find no sugar in urine passed on rising, when that passed on retiring at night may contain a small amount of sugar--from ¼ to 1 per cent. On the other hand, I have found a small amount of sugar in the morning urine when the evening urine contained none. Anxiety and excitement both increase the proportion of sugar. Inosite, or muscle-sugar, is sometimes associated in urine with diabetic sugar, and occasionally replaces it. So, too, in experiments upon animals puncture of the fourth ventricle is sometimes followed by inosuria instead of glycosuria, and in corresponding organic disease of the brain the same thing is observed. The substitution of grape-sugar by inosite in the course of diabetes is considered by Laboulbène[27] a favorable change. [Footnote 27: "Note sur l'Inosurie, succédant au diabète glycosurique, et paraissant avoir une action favorable," _L'Union Médicale_, Oct. 14, 1883.] As would be expected, the specific gravity of saccharine urine is usually high--most frequently from 1025 to 1040--and Bouchardat noted a specific gravity of 1074 in one instance. On the other hand, I have found sugar easily detectable in urine with a specific gravity as low as 1010. Pavy records an instance of the same specific gravity, and Dickinson one in which the specific gravity was as low as 1008. It is to be remembered that the sugar is rapidly destroyed when fermentation sets in. A coincident diminution in the urea and other solids of the urine will reduce the specific gravity of a saccharine urine otherwise heavier.