Abstract

1. It has been shown that the assay of digitalis by emesis in pigeons gives very different results when different groups of pigeons are used. If the group contains only three birds the error may approach 300 per cent, and if it contains 25 birds, it may be 30 per cent.

2. Providing however that due allowance is made for the individual variation in different birds, by injecting a group of 25 with the same dose to determine the percentage in which emesis follows, digitalis may be assayed satisfactorily by means of a characteristic curve relating the percentage having emesis to the potency of the preparation.

3. The assay should be comparative, stating the strength of an unknown sample in terms of the strength of a standard. The accuracy is considerably increased if the strength of the standard is determined on the same group of birds as that used for the unknown.

4. The pigeon method so used, gives results resembling closely enough those obtained by the cat or the frog method.

5. The errors which follow from injecting different groups of a few animals each with different doses are fully discussed.