298 



BIOCHEMICAL SYSTEMATICS 



SO it was thought to be at the time, happens also to be exceptional in 

 another way. In 1914, Zade published a rather lengthy study on the 

 serology of some legumes and grasses, including a serological com- 

 parison of three species of clover, Trifolium repens, T. pratense and 

 T. hybridum. At that time T. hybridum was regarded as a hybrid of 

 T. repens and T. pratense [in Fernald's (1950) treatment it is im- 

 phed that this view no longer holds]. Zade concluded that serum 

 interactions supported the hybrid nature of T. hybridum. For example, 

 hybridum serum reacted more strongly with the putative parents 

 than did reciprocal tests with serum from the parents. Chester, who 

 discussed this work briefly in his 1937 review says, "Zade, with precip- 

 itin test, showed that Trifolium pratense and T. repens are related, 

 but serologically distinct, their hybrid, T. hybridum reacts so strongly 

 with both as to demonstrate its hybrid nature." If there had been no 

 a priori conclusion that T. hybridum was actually a hybrid, the sero- 

 logical data may have been interpreted as indicative that T. hybridum 

 was the closest of the three to some primitive Trifolium stock. There 

 is then the danger of circular reasoning in the interpretation of 

 such data. 



A similar situation is that of the disputed hybrid, Vicia 

 leganyi {''Lens esculenta X Vicia sativa" ). Its serological properties 

 were shown to be intermediate between the protein complexes of the 

 "parents" (Moritz and vom Berg, 1931). Though these data alone 

 would not serve to establish the hybrid identity of V. leganyi, Moritz 

 has developed supplementary techniques to disclose hybrids through 

 serological methods. Suppose, for example, that species A contains 

 antigen complement a + b and species B contains antigen comple- 

 ment b -h c. Thus, b represents the common antigenic substances. The 

 hybrid should, therefore, possess a complement a -f- b -|- c, and a hy- 

 brid antiserum, if adsorbed with serum type A and then serum type 

 B, should be completely neutrahzed. It should then give a negative 

 response to hybrid serum. Presumably then, if a residual activity 

 remained in the antiserum after adsorption with sera A and B, one of 

 three explanations might hold: 



(1) The plant was not a hybrid. 



(2) New "hybrid-type" antigenic substances were present. 



(3) Genetic heterozygosity in one or both parents led to 

 individual differences in antigenic complement. 



However, if serum of the "hybrid," completely neutrahzed 

 antisera of type A and type B, this result offers strong support for the 

 true hybrid nature of the plant in question. In the light of these 



