128 CLINICAL BACTERIOLOGY AND H^MATOLOGY 



unsuitable materials, especially the amboceptor. Another 

 advantage is that it is possible to perform a rapid quantita- 

 tive estimation, by determining the highest dilution of antigen 

 which will just give a positive reaction. We will suppose that 

 there is no haemolysis with antigen diluted I in 10. Prepare 

 a second antigen, diluting it i in 30, and repeat the test. If 

 there is still no haemolysis, try at i in 50, and, if necessary, 

 higher. There are other methods of performing a quantita- 

 tive test, but this is the simplest. A series of quantitative 

 estimations is of enormous value, as it enables us to gauge 

 the effect of treatment. 



SECOND MODIFICATION OF THE TEST. 



Good anti-human amboceptor is not at present easily 

 obtained, and for the benefit of those who may be able to get 

 sheep's corpuscles I may point out that the process described 

 above can be carried out in exactly the same way, using 

 washed sheep corpuscles and anti-sheep serum, which is an 

 ordinary article of commerce. The process is exactly the 

 same, but a word or two is necessary as to the materials. The 

 sheep's blood must be obtained fresh from the slaughter- 

 house. The best method is to prepare a flask containing a 

 few glass beads or fragments of glass tubing, plugging it 

 with cotton-wool, and sterilizing by hot air. The butcher is 

 to be instructed to receive the blood into this flask, and to 

 rotate the latter g'ently for a few minutes. It will then be 

 defibrinated, and will no longer clot. When received in the 

 laboratory, a few cubic centimetres are to be washed in sterile 

 normal saline solution three times, centrifugalizing down each 

 time. Washed corpuscles will keep about a week in sterile 

 normal saline solution to 300 parts of which i part of formalin 

 has been added. 



The anti-sheep serum must be diluted before use, as it is 

 usually too strong when sent out. The degree of dilution 

 varies greatly; I have had samples which have been quite 

 inert and others that have acted when diluted 1,200 times, and, 

 as a rule, it works well at a dilution of i in 500 or 600. The 

 method to be adopted is as follows : Prepare a series of dilu- 

 tions of fresh normal serum with normal saline solution in 

 the proportion of i to 4, as before. Prepare also dilutions of 

 the anti-sheep serum of i in 200, i in 400, i in 600, i in 800, 



