132 THE ROYAI, SOCIETY OF CANADA
liberating sufficient glucose to maintain the blood sugar above the
convulsion level. One of the animals in both groups died from shock
following an injection of adrenalin and the results are not included.
The following percentages of glycogen were found:
A. Insulin and adrenalin, 9.7 and 10.2 per cent.
B. Adrenalin alone, 4.3 and 8.0 per cent.
Nov. 22nd (1923).-Six well-fed rabbits were injected at about
half hour intervals with 5 u. of insulin (per kg.) followed as soon as
possible by 0.5 c.c. - adrenalin chloride. The first injections were
made at 10 a.m. and the last about 4.30 p.m. The total amount of
adrenalin chloride injected into each animal was 5.5. c.c. Since
convulsions occurred in two of the animals treated with insulin the
amount of this actually injected at each animal varied somewhat.
Rabbit 2.-2260 gms. received 100 u. insulin and did not show
typical symptoms. Liver contained 22.40 per cent. glycogen.
Rabbit 3.-2320 gms. received 93 u. insulin and showed symptoms
in which the convulsions were mild, although coma was pronounced
towards the end. Killed at 4.31. Liver contained 18.7 per cent. of
Rabbit 4S.-1560 gms. received no insulin; killed at 4.34. Liver
contained 8.7 per cent. glycogen.
Rabbit 5.-1820 gms. received no insulin; killed at 4.38. Liver
contained 5.9 per cent. glycogen.
Rabbit 6.-1690 gms. received no insulin; killed at 4.45. Liver
contained 3.7 per cent. glycogen.
The foregoing results are gathered in tabular form in Table I.
It will be seen that there was on an average twice as much
glycogen in the livers of the animals treated with insulin as in those
without. Perhaps a fairer comparison is among the animals of each
day, since feeding conditions and duration of adrenalin action were
then alike. When this is done there only are two cases (April 12th
and 19th) in which the insulin-treated animals did not show more gly-
cogen than those not so treated.
Notwithstanding every precaution to have the animals of the
two groups treated alike, except that those of one group receive
insulin, there is always a considerable chance of error. This is owing
to the fact that one can never predict what the percentage of glycogen
in the liver will be even when the feeding conditions are strictly alike.
It was with the object of minimizing this source of error as much as
possible that the animals in the experiments of Mar. 29th, April 12th
and April 19th were starved for some time prior to being put on a