function to fulfil, in the animal kingdom have common properties in their structure. In spite of our favourable experiences we have not ventured generally to recommend the use of animal organs. It is still very difficult to find a proper balance amongst the contrary results of many observers, and were we to change the type of substrate without sufficient experience, we should arrive at still more divergent results. This is the reason why it is particularly necessary to use organs of the same species as that to which the serum under investigation belongs, as well as those of a different type. Only when it is established that harmonious results are obtained ought we to be satisfied with non-specific organs, and always under the condition that no substrate is used which shows definite pathological alterations.
Means of Obtaining Blood Serum.—Three conditions have to be complied with. The serum must be as poor as possible in diffusible substances which react with ninhydrin, and this is attained by taking the blood in a fasting condition. In all cases in which the albuminous metabolism is very rapid, in cases of disease which are accompanied by decay of the tissues, as in the case of carcinoma, in cases of absorption of exudates and transudates, in all purulent processes, and lastly, in effusions of blood, the blood always contains a larger quantity of such compounds. The blood serum must further be absolutely free from
