Qualitative Tests 69 



along one side of the upper surface of the base. When a number of 

 blood samples, my usual limit being 80, were to be tested by different 

 antisera, the racks were placed in rows of suitable length upon the 

 table. Where 80 bloods were being tested, the racks were ordered as 

 follows : 



Backs 

 Row 1. A B C D 



2. E F G H 



3. I J K L 



4. M N O P 



the number of rows corresponding to the number of antisera to be 

 used in testing the series of 80. In rows of 20 bloods the number of 

 antisera used might number 12; in sets of 80, the number of antisera 

 used was usually 4 to 6. 



Each blood, whether on filter-paper or in bottles, received a number 

 on arrival. This number is given in the following tables (in the second 

 column) in small figures, and corresponds to the order in which the 

 bloods were tested. My plan was to test every blood received by means 

 of every antiserum produced. In this way a "network" was, so to 

 speak, formed through which all the bloods were passed in testing. In 

 some cases this scheme was departed from. For instance there appeared 

 to be no object in continuing to test avian bloods with mammalian 

 antisera after ten or more such antisera had continually given negative 

 results. Moreover, in some cases the supply of antiserum or of 

 blood to be tested ceased, and the series of tests was necessarily 

 interrupted. In other cases again the general results obtained with 

 one antiserum sufficed, so that it appeared only necessary to test a given 

 group with a particular antiserum, this for instance being the case with 

 the tests made with anti-chimpanzee and anti-ourang serum upon the 

 bloods of Primates, etc. The bloods received were not tested in their 

 zoological order, but in the order in which they were received. For 

 example : 



696. Sus scrofa: Pig, 



697. Ateles vellerosus: Spider Monkey, 



698. Balaenoptera rostrata: Rorqual, 



699. Ornithorhynchus paradoxus: Platypus, 



700. Podiceps : Diver, 



were actually tested in succession. Similarly the tests with antisera 

 were made in the order of production of the antisera, which depended 

 entirely upon the time when the bloods for injecting rabbits were 

 received. For reasons of convenience the order of these antisera, as 



