Anti-Human Serum 167 



cloudings m<ay also be observed in the dilutions of other mammalian 

 bloods, and such an antiserum will also cloud the dilution of lemur 

 blood. This is however a feeble reaction to which I have already 

 given the name of mammalian reaction, for I have not found it to occur 

 amongst non-mammalian bloods. This is very well seen by reference to 

 the preceding table, where only one avian blood, out of 320 tested, 

 is recorded as showing a faint clouding. On what such faint cloudings 

 may depend, seems fairly clear. To begin with, they are such great 

 exceptions that this alone suggests their being due to some error. At 

 times they may be due to clouded solutions, or to clouded antiserum, 

 at other times they may be due to the blood samples having been 

 brought together by collectors. In an earlier paper I first showed 

 that this reaction took place in blood mixtures. Consequently, if a 

 blood sample, say from a bird, were brought in contact with one from 

 a mammal, a positive reaction might take place with anti-avian as well 

 as with anti-mammalian serum upon these being added to a dilution 

 made from a sample of mixed bloods. Also, supposing that samples 

 of dried bloods are stored in contact with one another, and are allowed 

 to rub against each other, we may have the same result. This especially 

 might occur where blood had been allowed to clot upon the paper, for 

 blood-scales might break off and become attached to a paper saturated 

 with another specimen of blood. Finally, a certain mimber of these 

 results may be reasonably ascribed to . the flxct that in cutting out 

 squares of filter-paper in succession from a number of different samples, 

 a trace of a foreign blood may at times adhere to the scissors used 

 and thus find its way into a solution to which it does not belong. It is 

 remarkable however how few tests have given other than the result 

 which experience taught one to expect. It is nevertheless well to 

 remember these sources of error so as not to attach any special 

 importance to exceptional reactions. 



The anti-human sera used in the foregoing tests were usually 

 powerful, and therefore we find more " mammalian reactions " noted. 

 In contrast to this, we see on referring to the tests with a much 

 weaker antiserum (anti-monkey, p. 171) that the action of the antiserum 

 was practically limited to the Primates, for, even when we include the 

 faint cloudings, reactions only occurred in 1 "/„ of all the other mam- 

 malian bloods examined. 



