300 K. STERN 



optimum. Among the information needed for planning an optimal breeding 

 program, genetic variances and heritabilites of characters under selection 

 are of utmost importance. Estimates of these parameters serve as a base 

 for calculating genetic gains to be expected from different breeding 

 procedures where mass selection or related procedures are to be planned, 

 where the characters in question are of the typical quantitative type, 

 and where the population size is sufficiently large in each generation 

 of selection to eliminate random effects of sampling. It seems reasonable, 

 therefore, to try to extend the concept of genetic gain to the field of 

 resistance breeding, as has been done by Bingham, Squillace, and Wright 

 (I960) for resistance of western white pine to white pine blister rust, 

 or, at least, to investigate the particular problems to be solved before 

 this can be done. 



Breeding for resistance against fungal diseases is by far the most 

 important field of resistance breeding in forest trees. The following 

 discussion is thus restricted to specific problems of this kind of resis- 

 tance breeding; other problems arising from peculiar features of parasitic 

 organisms other than fungi, like behavioral performance of insects, have 

 been neglected. 



GENETIC BACKGROUND OF HOST -PARASITE SYSTEMS 



The evolution of parasitism has long been one of the most interesting 

 fields in evolutionary science. It offers examples of extreme specializa- 

 tion of organisms such as the adaption of metabolic cycles of parasites 

 to the performance of one or more host species. Host species, on the 

 other hand, are often able to evolve special mechanisms enabling them to 

 escape from being parasitized or to eliminate infections. 



It seems no longer justifiable to speak of evolution of virulence 

 of parasites without taking into account the simultaneously evolving 

 resistance in the host. Both processes affect each other. Geneticists 

 and evolutionists as well as breeders are stressing this mutual depen- 

 dence by speaking of co-evolution of parasitism or of evolution of host- 

 parasite systems. Genetic interactions between host and parasite are 

 often evident within short periods. When selecting for higher resistance 

 the breeder tries to change such a system by favoring the host. He might 

 either work on a host-parasite system that has reached an equilibrium 

 state or on a disequilibrated one. 



v Much of the theoretical and experimental work on resistance breeding 

 has been done on host-parasite systems characterized as gene-for-gene 

 systems (Flor, 1955). Mode (1958, 1961) was able to define some special 

 cases where stable equilibria can be reached if the host population 

 possesses a set of genes for resistance (R-genes) and the population of 

 the parasitic species is able to activate a set of genes for virulence 

 (V-genes) , each of them overcoming resistance reactions of one or 

 several R-genes . Person (1966, 1967) was able to verify one of the 

 stable equilibria predicted by Mode (1958) in a wheat :wheat-rust gene- 

 for-gene system. 



Gene-for-gene systems with many genes on both sides are not likely 

 in host-parasite systems where the host is a tree species and the parasite 

 a fungus. Generation intervals are too different, favoring the parasite 

 that is often able to produce more than a hundred generations while the 

 host is producing only one generation (Hattemer, 1967) . The same is 



