It will be convenient to summarise the results of the foregoing observations and the conclusions which appear justified by a comparison of results obtained by the two methods of standardisation under consideration, namely (1) cutaneous, on the human subject, and (2) intracutaneous (intradermal) on the guinea-pig.

1. In the absence of knowledge regarding the essential chemistry of tuberculin, a chemical method of standardisation is excluded.

2. The phenomena of allergic reactivity point to a biological basis for standardisation.

3. The allergic reactions have been studied in the sensitised animal and in the human tuberculised subject.

4. Tuberculins of unknown potency have been compared with a tuberculin of known potency which has been selected as standard.

5. Comparative observations have been made as between the standard and the unknown tuberculin, by (a) making use of different dilutions, and (6) assessing results in animals (sensitised) and in human (tuberculised) subjects.

6. In the sensitised guinea-pig the intracutaneous method has been used, and a definite procedure followed in the determination of results.

7. In the human tuberculised subject the cutaneous method has been used, and similar procedure followed for the determination of results. The method and results are illustrated in the text.

8. The results in the two sets of observations are definite and comparable. Both afford a basis for standardisation.

9. The average probable errors by the two methods are approximately equal, that is, the accuracy of the one method is approximately equal to that of the other.

10. If this be so, a strong plea may be advanced in favour of the human test. Shortly expressed: human (tuberculised) subjects are readily available for observation. The procedure involves little preparation, and the results are easily read with exactness. The tuberculin under test is to be used thereafter in relation to the human subject. This fact enhances the value of the test observations.

If it be objected that in intracutaneous injection the amount of tuberculin introduced is measured more precisely, it may fairly be maintained that the droplet application of tuberculin is limited to a sharply defined area of skin surface and, further, that the clear skin of the human subject allows of more accurate estimation of the diameters of the areas of reaction.

The present enquiry has shown that assays on a variety of human (tuberculised) subjects yield consistent results, and similarly, assays on various animals give consistent results. Yet the animal results are not always contistent with the human results. The explanation of the discrepancy is not very clear. It is not impossible that certain strains of tuberculin act in less degree on the human subject and in greater degree on the animal, and conversely. The occurrence of such differences might be misleading and even involve risk, if standardisation tests were limited to animals without control from observations on the human subject.

If we grant, as the records have shown, that the procedure in relation to the human subject is sound and is innocuous to the human subject of the test, much may be said for the simplicity of the method and for the clarity of results obtained. As the tuberculins under test are destined for use on human subjects—for diagnostic and therapeutic purposes—it would seem reasonable, and probably safer, to base the standardisation of tuberculin (for human purposes) on observations in relation to man.