RESISTANCE TO RUSTS IN HARD PINES 449 



In the most extensive of these studies (Kinloch and Stonecypher, 

 1969), controlled- and open-pollinated families from a large number of 

 randomly chosen parent trees growing in a natural stand were outpl anted 

 in an area of moderate to high hazard from fusiform rust. The same 

 families were replicated on different but proximate sites that varied 

 considerably in their edaphic properties and in the amount of rust that 

 subsequently developed on the trees. After a natural epidemic, family 

 means ranged from 0.5 to 10.8 galls per tree on the site with the most 

 rust, and from 0.0 to 2.2 on the site with the least. Estimates of 

 herit ability, on a family basis, were high (0.65 to 0.85) and remarkably 

 consistent for both controlled- and wind-pollinated families on all sites, 



This study also showed the influence of site in predisposing trees 

 to increased susceptibility. The sevenfold range in the amount of rust 

 observed on different sites was most strongly associated with factors in 

 the edaphic environment, particularly the previous cultural history of 

 the sites. Thus, trees on sites that had been forested and relatively 

 undisturbed before planting had the least amount of rust; those on sites 

 that had been intensively cultivated up to or within a few years of 

 planting had the most. This evidence, though circumstantial, was consis- 

 tent with independent observations of the effects of cultivation, ferti- 

 lization (Gilmore and Livingston, 1958), planting on old fields (Siggers 

 and Lindgren, 1947), and other site-disturbing treatments (Miller, in 

 press) that increase the intensity of fusiform rust infection. These 

 results have important and obvious implications for future plantation and 

 second growth management which increasingly will tend to alter natural 

 forest sites by similar cultural practices. These cultural treatments 

 will put an even higher premium on obtaining stable genetic resistance. 

 Fortunately, families highly resistant and susceptible to rust consis- 

 tently maintained their relative rankings in susceptibility on all sites- 

 even though the absolute amount of rust may have increased greatly from 

 one site to another. Furthermore, relatively highly resistant families 

 were much less affected by site influence than susceptible ones. 



In summary, results of studies on fusiform rust show that ample 

 variation in resistance exists in wild host populations and that resis- 

 tance is highly heritable and apparently stable over a range of natural 

 environments. This pattern of variation may well characterize that of 

 other pine hosts to their endemic rusts. 



ARTIFICIAL INOCULATION 



Although the practical value of parent material selected for tree 

 improvement ultimately depends on field performance of progeny under 

 normal conditions, more rapid and uniform tests for screening progenies 

 and for basic research presumably can be accomplished by artificial 

 inoculation. Ideally, such tests would demand exposing host material of 

 appropriate age and stage of development to inoculum of known genetic 

 identity, in sufficient and standard amounts to enable prediction of 

 progeny performance under conditions of acute hazard in the field. 

 These requirements have yet to be fulfilled adequately. 



Recurrent objections to artificial inoculation, as it has been 

 practiced (e.g., Jewell, 1960; Kinloch and Kelman, 1965), are that over- 

 whelming doses of inoculum applied under environmental conditions 

 optimal for the pathogen may only screen for immunity, masking relative 

 levels of resistance that might still be effective under field conditions 



