466 ILffiMOLYSINS AND ALLIED BODIES 



Thus by injecting the blood of a pigeon into a hen, or that of a 

 rabbit into a guinea-pig, no precipitin is developed. The prae- 

 cipitin produced by injecting the milk of one animal into another 

 of a different species is also specific in its nature. Thus if cow's 

 milk be injected into a rabbit, the rabbit's serum will precipitate 

 the casein of cow's milk, but not that of a goat or of human 

 milk. 



The production of precipitins has received a practical applica- 

 tion which may be of considerable medico -legal value. This 

 applies to the detection of human blood which, when present in 

 small amount, cannot be distinguished, with certainty, from that 

 of any other animal, by any of the older tests. The new test we 

 owe to A. Wassermann ( 7 ). It depends on the fact that the serum 

 of a rabbit, subcutaneously inoculated several times with human 

 serum, will give a precipitate in any solution containing the pro- 

 teids of human blood serum, or of some of the higher apes, but 

 not of other animals. The exact technique of the method is as 

 follows : The blood stain or crust is macerated in isotonic salt 

 solution, and filtered. The clear nitrate is divided into two 

 parts, each of which is placed in a small test-tube. To one of 

 these is added the precipitating rabbit's serum ; the other is left 

 standing to serve as a control ; into a third tube is placed some of 

 the precipitating serum alone ; and into a fourth tube a mixture of 

 the precipitating serum and some indifferent serum (of an ox) in 

 physiological saline. These four tubes are placed for about an 

 hour in an incubator at body temperature. If the stain be due 

 to human blood, a precipitate or cloudiness will develop in the 

 first tube, but all the others will remain clear. The amount of 

 precipitate produced by mixing the human precipitin with the 

 blood serum of an ape is most marked with those apes higher 

 up in the scale of the Primates. Indeed, as Nuttall ( 14 ) and 

 others have shown, the biological relationship of one animal to 

 another can be very clearly defined by this reaction. It may be 

 well to mention that, although the use of Wassermann' s reaction 

 for the detection of human blood is undoubtedly of great medico- 

 legal value, it will require some further elaboration before it can 

 be used as an absolute test ; there being, apparently, a tendency 

 in some cases for anti-prsecipitins to be developed, and these of 

 course mask the reaction. The method has a further practical 

 application. If an extract of horse flesh be inoculated into a 

 rabbit, this rabbit's serum will produce a precipitate in extracts 



