BY H. G. CHAPMAN. 



551 



in six doses. The tests were carried out by mixing 0"1 c.c. anti- 

 serum and 05 c.c. extract of the seeds to be tested. Both heated 

 and unheated extracts of the seeds were employed. The controls 

 are omitted from the table. 



Table xix. 



Quantitative differences in the amounts of the precipitates are 

 not recorded in the table. In the first place, it may be noted 

 that the seeds of the Natural Order of the Leguminosse are 

 marked off clearly from the seeds of the other well marked 

 Natural Orders. In the second place, certain of the sends of the 

 Let'uminosje do not react with the antiserum for Acacia pyc- 

 nantha. These results have been repeated with other aniisera 

 with similar results. It is probable that the precipitin test will 

 be equally useful in the vegetable kingdom as in the animal 

 kingdom. It po.ssesses also the possibility of supplying data* for 

 a correct appreciation of the value of morphological characters in 

 the differentiation of species. 



*An extended research on this aspect is being carried out in collaboration 

 with Dr. Petrie. 



