624 NORMAN E. BORLAUG 



[O 



developed by the Triassic period, some 230 to 180 million years 

 (Moore, 1949). The fossil evidence on the age of the rust fungi is even 

 more unreliable. Arthur (1929, p. 205) refers to a Miocene fossil (12 

 to 25 million years old) of Phelonites lignitum Fres . , a rust on seed of 

 Glyptostrobus , a coniferous genus once common, but not restricted to a 

 single species in southeast China. He considers Phelonites similar to 

 present-day Cronartium and postulates that the rust may have originated 

 in the early Mesozoic Era, but admits this is based on speculation rather 

 than on observational evidence. 



What is the explanation for the presence in the American species P. 

 montieola and P. strobus of the "fossil polygenic" general resistance 

 to C. ribicola, a pathogen only recently introduced into North America? 

 Presumably the center of origin of C. ribicola was in eastern Asia on 

 Asiatic soft pine species. The pathogen is endemic there today para- 

 sitizing, but causing little damage to a number of species including 

 P. sibirica, P. pumila Regel^ P. kovaiensis , P. armandii , P. parviflora 

 (and forms) and P. walliohiana. 



Similarly, in the mountains of the Central and Balkan European 

 countries, which may be a secondary zone of differentiation of the 

 pathogen, and which is also the area from which the pathogen was intro- 

 duced into North America, C. ribicola lives in a balanced biotic rela- 

 tionship with P. peuce Griseb. and P. cembra L. 



The circumstantial genetic evidence indicates it is highly probable 

 that at some unknown time in the geologic past the range of the ancestors 

 of modern P. montieola and P. strobus overlapped with those of one or 

 more of the Asiatic soft pine species on which C. ribicola was then 

 endemic (Moore, 1949;0gburn, 1970). Thus, the ancestors of present day 

 P. montieola and P. strobus likely came under infection and the selection 

 pressure of the pathogen and responded by evolving into species with 

 polygenic general resistance to C. ribicola. Before the Bering land 

 bridges disappeared between what is today the Asian continent and North 

 America, the predecessor of P. montieola had extended its range to the 

 latter continent. One can conjecture either that, for some inexplicable 

 reason, the pathogen was left behind in Asia (as was the case 400 years 

 ago when Zea mays and Puccinia sorghi were both introduced into Africa 

 from the Americas, while the other rust pathogen P. polysora was left 

 behind in America) , or that more probably the pathogen was subsequently 

 killed out on the American host at some later geologic period, perhaps 

 during one of the glacial periods of the Pleistocene or some earlier 

 ice age s , during which the host, P. monticola 3 was able to survive, while 

 the pathogen became extinct on the North American continent. 



Consequently , we may conclude that once the polygenes for general 

 resistance have evolved in an open-pollinated species they can persist 

 indefinitely in a population in the absence of the selection pressure 

 of the pathogen if they are not linked to other genes adversely affecting 

 the survival of the host species. They will, however, become dispersed 

 in the population and, until brought back together again under renewed 

 selection pressure of the pathogen, may sometimes give the impression 

 that the entire population is susceptible. 



It appears to me that the circumstantial evidence supporting the 

 above hypothesis implies that: (1) the pathogen C. ribicola is very 

 ancient; (2) P. montieola and P. strobus, or their ancestors, and C. 

 ribicola co-existed in a balanced host-pathogen relationship in the 



