It has been suggested that assistance could be rendered to forensic medicine by making use of this property of serums, to discover the origin of a certain blood. As is well known, it is often very important to decide whether a stain was caused by the blood of man or of another animal. Until quite recently it was not known how to distinguish human blood from that of other mammals. Experiments have been made to discover if the red corpuscles found in the blood stain could be dissolved by the serum of animals which had previously been injected with the blood of man. In a certain case the human origin of the stain in question was shown. But it was soon discovered that this method was not infallible. It is now found that the method of precipitates gives much more conclusive results. It is done in this way: Human blood is injected several times into any animal (rabbit, dog, sheep, horse). Some time afterwards the animal is bled, and a clear and limpid serum, quite devoid of corpuscles, is prepared. When there is added to this serum one or several drops of human serum, it forms immediately a precipitate which falls to the bottom. In this way it is discovered whether the prepared serum is sufficiently active. It then becomes possible to recognise even dried human blood. A little of such blood is dissolved in normal salt solution, and placed in a tube containing the serum of an animal prepared by means of the injections of human blood. If a precipitate forms in the liquid in a short time, the fact indicates that the stain is really human blood. This method is being practised in forensic medicine. This reaction is of great interest to us because it is of assistance in revealing the relationship between species. The serum of an animal prepared with the blood of the fowl gives a precipitate, not only with the serum of the fowl itself, but also with that of the pigeon; on the other hand, it remains undisturbed when the serum of mammals is added. The reaction indicates then that there is a sufficiently marked degree of relationship between the fowl and the pigeon. Here is another example: the serum of an animal prepared with the blood of an ox gives an abundant precipitate when there is added to it a little blood serum of the ox, but it does not produce this reaction with the serum of any of the other mammals, not even with that of the sheep, stag and deer.[59] The relationship between the _Bovidæ_ and these other ruminants is then not so close as that between the fowl and the pigeon. How does the serum of animals which has been injected with human blood behave? The serum capable of giving a precipitate with human serum does not produce the same reaction except with the serum of some monkeys (the small _Papio_).[60]