448 BOHUN B. KINLOCH, JR. 



studies (Wells, 1966; Kraus , 1967) included seed from a few distant 

 sources. They showed that the local variation occurred within the 

 broader pattern of geographic variation described above. 



Reasons for these patterns of variation are not known, though two 

 hypotheses have been advanced. Wells and Wakeley (1966) proposed that 

 varying degrees of selection pressure imposed by the rust in different 

 parts of its host's range could have generated corresponding variation in 

 frequencies of genes for resistance. Presumably, this hypothesis would 

 demand that the most effective selection for resistance evolve near the 

 gene center of the pathogen and host. They countered their own argument 

 by pointing out that populations with the greatest resistance now exist 

 where the incidence of the disease and conditions for it are minimal -- 

 namely, at the extremities of the species range--and that there is no 

 evidence that the situation was ever otherwise. 



The argument for selection pressure cannot be discounted on a local 

 level, however. Since very young trees may be intrinsically more suscep- 

 tible than older ones, and certainly more liable to die once infected 

 (Siggers, 1949), and since the frequency and severity of rust epidemics 

 fluctuate widely in different years and regions, different stands may be 

 subjected to varying degrees of selection pressure from the pathogen 

 within a single generation. Because individual trees within a stand vary 

 in susceptibility, it is reasonable to conjecture that a young stand 

 subjected to a severe epidemic would have gene frequencies quite different 

 from those of a stand that escaped infection until relatively mature. The 

 genetic bottleneck that second growth stands regenerated by one or few 

 seed trees commonly go through in the South after logging or an agricul- 

 tural field is abandoned could produce the same effect (Kinloch and 

 Stonecypher, 1969). 



A more plausible explanation for the geographic pattern of variation 

 observed is introgression of shortleaf genes into loblolly (Wells and 

 Wakeley, 1966) . Evidence for this hypothesis is the known resistance 

 shortleaf imparts to its hybrids, and the fact that where the two species 

 overlap or exist in mixture (especially west of the Mississippi River), 

 intermediate forms occur (Wells and Wakeley, 1966; Kinloch and Stonecypher 

 1969; Zobel, 19S3)--both in morphological traits and in protein profiles 

 (Hare and Switzer, 1969) . 



Racial variation in resistance to rust apparently does not occur in 

 slash pine (Snyder, Wakeley, and Wells, 1967; Gansel and Squillace, 1965). 



INDIVIDUAL VARIATION 



The patterns of geographic variation demonstrated are interesting 

 from the standpoint of evolutionary changes in the genetic structure and 

 gene frequencies in host populations. But practical utilization of 

 variation in resistance for tree improvement will depend more on variation 

 found among individual trees. That a large amount of such variation exists 

 has been amply confirmed in progeny tests of both slash (Barber, 1964; 

 La Farge and Kraus, 1967) and loblolly (Woesner, 1965; Barber, 1966; 

 Kinloch and Stonecypher, 1969) pines. 



