BIOCHEMICAL STUDIES OF HYBRIDS 307 



resented by the phenolic complements of apple and pear leaves, and 

 the results reported are not at all surprising. 



In general, hybrids tend to accumulate the compounds pecuHar 

 to both parents, but exceptions should be expected. For example, 

 Kawatani and Asakina (1959) found that a hybrid of Papaver orientale 

 and P. somniferum contained the alkaloids typical of both species ex- 

 cept for oripavine, an alkaloid reportedly in P. orientale. A serious 

 limitation of this study, however, is that the alkaloids of the actual 

 parents were not examined but rather were assumed on the basis of 

 previous reports. It is particularly important to establish that a cer- 

 tain compound is actually present in the parent (and consistently 

 present in the species in question) when considering the failure of the 

 substance to appear in the hybrid, for such instances, if authentic, 

 may prove to be of special theoretical significance in connection with 

 metabolic interrelationships. 



There is no reason to expect that the pattern of inheritance 

 of two biochemical characters must be similar. It is quite easy to 

 imagine a character which becomes intermediate in the hybrid while 

 yet another exceeds the parental extreme. Similarly, biochemical and 

 morphological characters need show no relationship. In this connec- 

 tion Dillemann's (1953) discussion of the inheritance of cyanogenetic 

 substances in natural hybrids of Linaria vulgaris and L. striata (a 

 cyanogenetic species) is an oversimplification. Dillemann finds that 

 the hybrid is intermediate in cyanogen content and points out that in 

 morphological characters the hybrid is always intermediate. He 

 then adds: 



If in interspecific crosses of Linaria intermediate characters are the 

 rule in the hybrids, then it indicates that the factors which govern the 

 characters are not dominant. Under such conditions, it is altogether 

 normal that the same would be true for the factor of cyanogenesis. 



Why one would expect the inheritance of a biochemical 

 character to be intermediate just because most morphological charac- 

 ters were intermediate in the hybrid is not clear. Sometimes bio- 

 chemical components are reduced in the hybrids, and sometimes they 

 are present in increased amounts. However, there is no likelihood that 

 any dependency upon the manner of inheritance of morphological 

 characters exists in such cases. 



Tsitsin and Lubimova (1959) described the appearance of 

 some new "hybrid" characters in inter-generic hybrids of Triticum 

 and Agropyron. These characters were indicated as multipistillate 

 florets which produced double and triple kernals and cases of stamens 

 becoming transformed into pistils. The authors come to a rather sur- 

 prising conclusion from these observations: 



