One of the most noticeable points in the pathology of the more excessive cases of polyuria is the disproportion which often exists between the amount of fluid ingested and the amount discharged, the latter often exceeding the former several times. The source of the excess of water has not been satisfactorily determined, but it is evident from a careful experiment of Watson, repeated by Dickenson, that the body has under some circumstances the power of appropriating water from the atmosphere instead of discharging aqueous vapor through the lungs and skin as usual. In the experiments referred to persons affected with extreme polyuria were weighed immediately after passing water, and again after as long an interval as they were able to restrain their thirst, of course being also without food and under observation, when it was found that the weight had been increased by a number of ounces. In Dickenson's case, weighing thirty pounds more or less, where the amount of urine excreted daily was from seven to nine liters, the gain in weight at several observations was as follows: in three hours, 15½ oz.; in five hours twenty minutes, 19¾ oz.; in three and a half hours, 3¾ oz. The DIAGNOSIS of this affection rests, in the first place, upon the determination of a permanent increase in the quantity of urine passed considerably above the normal, and, as has been already remarked, may require a measurement of the daily amount--a procedure which it is well to make a matter of routine in any cases where urinary trouble may be present. The increase being found, if it be very great it will only remain to determine whether sugar be present, which will be indicated by the specific gravity and the appropriate chemical tests. Traces of sugar are sometimes found in cases of polyuria which do not present the characteristics of saccharine diabetes, and can hardly be considered to materially affect the character of the disease. A specific gravity decidedly above normal, with an excessive quantity of urine, is not likely to belong to anything but diabetes mellitus, though the chemical tests should never be neglected. If, however, the polyuria be only moderate, it becomes necessary to exclude surgical affections of the urinary passages, especially an enlarged prostate, often attended with retention and distended bladder. Pyelitis and hydro-nephrosis may also give rise to the same condition of over-activity of the kidneys. The appropriate surgical examinations with the sound may be necessary, but the presence of pus, bacteria, and the epithelium of the urinary passages {33} in the surgical urine, as well as its frequent alkalinity, may direct a very strong suspicion before the sound is used. The age of the patient also will be of considerable weight in this connection.