258 



BIOCHEMICAL 5YSTEMATICS 



criteria alone that they are as close to the Loganiaceae as they are to 

 the Gentianaceae. Though the chemical correlations are interesting, 

 they do not provide sufficient evidence to adduce family relationships. 

 In the family Asclepiadaceae there occur a number of glyco- 

 sides, often described as "bitter principles." Korte and Korte (1955a) 

 compared the properties of several of these substances and found 

 that, apparently a number of substances described independently and 

 given different names were identical to one or the other of two com- 

 pounds of widespread occurrence in one of the two sub-families 

 (Cynanchoideae). The two compounds are kondurangin, a glycoside- 

 yielding glucose, thevatose, and cymarose, and vincetoxin, a glycoside- 

 yielding glucose, thevatose, cymarose, and diginose. The structui'es of 

 the two aglycones have not yet been fully estabhshed. These sugar 

 combinations were otherwise known only from the cardiac poisons, 

 and therefore these bitter principles, though not yet characterized, 

 appear to be closely related to the cardiac poisons. As a result of the 

 efimination of some of the chemical synonymy, a quite interesting 

 pattern of distribution of the compounds is exposed. For example, all 

 members of the sub-family Cynanchoideae examined contained either 

 vincetoxin or kondurangin. The sub-family Periplocoideae do not con- 

 tain either of the previously described bitter principles but rather a 

 heart poison known as cardenohdglycoside (cardenohd, p. 257). The 

 only genera in the Cynanchoideae to contain cardinolidglycoside are 

 Xysmalobium, Gomphocarpus, and Calotropis, and they show no 

 evidence of being misplaced in the sub-family. The two sub-families 

 are well marked and are without obvious transitional forms; the 

 Cynanchoideae occur in both hemispheres while the Periplocoideae 

 are absent from North and South America. It is noteworthy that 

 the related family Apocynaceae typically produces cardenohdglycoside. 

 Korte and Korte concluded that the Apocynaceae gave rise to the 

 Asclepiadaceaes and within the latter family the Periplocoideae are 

 the most primitive sub-family. In this latter case, the morphological 

 and biochemical transition has not been strictly parallel so that cer- 

 tain genera of the Cynanchoideae retain the synthesis of cardenohd- 

 glycoside. Korte and Korte conclude that the glycosides, vincetoxin 

 and kondulangin, are truly "characteristic constituents" of the sub- 

 family Cynanchoideae, and their conclusion seems to be well sub- 

 stantiated by the evidence at hand. 



The family Apocynaceae has been studied in considerable de- 



3 In another paper this statement is reversed; "Die Apocynaceae, die sich wahr- 

 scheinhch phylogenetisch von den Asclepiadaceae herleiten lassen, . . ." It is not possible 

 to determine whether this was an unintentional reversal or not since there was no further 

 comment in the body of the text. 



