ing care never to allow more than 40
calories per kilogram of body weight.

~Reappearance of Sugar.~--Should the urine again show the presence of
sugar, another period of fasting lasting for twenty-four hours, or
until the urine is again sugar free, must be instituted. After the
second fast the increase in the diet may be twice as rapid as used
after the first fast. However, it is not advisable to increase the
amount of carbohydrates to more than half of that determined by the
former tolerance for a period of two weeks, during which time the
urine has been entirely sugar free, then the increase is made more
slowly and the amount given should not exceed 5 grams a week.

~Weekly Fast Days.~--One day in seven should be set aside by the
diabetic patient for fasting, when the carbohydrate tolerance is less
than 20 grams. When, however, the tolerance is between 20 and 50 grams
of carbohydrates, the patient may take one-half of his daily allowance
of protein and fat and a certain amount of 5% vegetables as well upon
the weekly fast day. When the tolerance reaches between 50 and 100
grams per day, vegetables of a higher carbohydrate content may also be
included. If the carbohydrate tolerance should exceed 100 grams per
day, the carbohydrates upon the fast day may be simply half of the
amount allowed upon other days.

~The Giving of Alkalies.~--If acidosis is evident, as may be indicated
by an excretion of diacetic acid, oxybutyric acid, or acetone in the
urine, alkalies may be given. Bicarbonate of soda may be given in
doses of 2 grams every 3 hours, as suggested by Hill and Eckman,[143]
but this is not as a rule necessary, for, as Jacobi aptly remarks:
"Prevention is the treatment of acidosis in children, and those
susceptible to acidosis should not have fat." What he has said for
children holds good for adults. However, it is likewise true that fat
must constitute a large part of a diabetic diet and the only way to
prevent it poisoning, is to raise the fat in the diet gradually until
the tolerance is determined.

~Determining the Extent of Acidosis.~--It will be remembered that in
an earlier chapter it was found that the excretion of ammonia in the
urine to a certain extent indicated the extent of the acidosis in the
body, that is, if the ammonia output exceeds three or four grams a day
(twenty-four hours), the extent of the acidosis is considerable, while
if it falls below that amount it is not alarming. More exact metho