564 HANS H. HATTEMER 



some way extreme. The risk of a certain family or variety remaining 

 basically untested is reduced by growing it under a series of environmental 

 conditions. It is difficult to conclude whether something like geographic 

 variation in virulence exists in the pathogen and whether the geographic 

 pattern varies rapidly in time without major shifts in the genotypic 

 structure of the host. If something of the sort exists, inoculation must 

 employ pathogen samples from several locations - as was done here at 

 Moscow. Or the host material must be exposed to the local pathogen at 

 several locations within a given area. In the latter case artificial 

 support of field infection might be difficult to accomplish. The method 

 has the advantage that different host vigor in the various plantations 

 (or sowings) is accounted for and more consistent estimates of general 

 combining ability might be made. Thus, interaction variances between 

 genotype and environment can be employed in estimating genetic gains that 

 may be realizable in the area. The distribution of such experiments 

 forming a series may greatly influence the validity of inferences made 

 on success and persistence of various operations of resistance breeding. 



"Absolute" resistance and its sooner or later breakdown are not 

 encountered if the variation of resistance is quantitative and host 

 material is exposed to other site conditions and pathogen genotypes. 

 It is just the selection response that may diminish. Field resistance of 

 half -sib families of Pinus silvestris against Lophodermium pinastvi 

 indicate that heritability for selecting trees on the basis of performance 

 of their open-pollinated progenies is much lower if interactions with 

 the environment are accounted for (Schmalenbeck Institute of Forest 

 Genetics, Germany, unpublished data) . 



SUGGESTIONS 



It is difficult to say anything under this heading without telling 

 commonplaces to the gentlemen here at Moscow who just cannot do every- 

 thing at the same time. The base material for breeding is most appropri 

 ately selected in areas where the pathogen wave went through and killed 

 many individuals. This was done here. Since the mode of variation of 

 resistance was not expected to be solely of the additive type, complete 

 mating designs were employed preferably. 



Species hybridization means to make a variety of new gene combina- 

 tions available. But unlike agricultural resistance breeding where so 

 often statements like "gene so and so from species so and so was incor- 

 porated in the new variety" are read, hybridization in forestry means an 

 array of new problems. First of all, we must foresee what makes these 

 other species resistant and how the genes responsible for resistance 

 vary in the first few generations of hybrids. The environments of these 

 new species may be different and the resistance behavior in hybrids of 

 varying status may be rather sensitive in the new environment. 



There should also be no fear of an intermittent generation of 

 inbreeding like crossing full sibs or making backcrosses. Inbreeding 

 is indicated in the selection for rare genes; but the mass-growing of 

 inbreds is forbidden because their reduced growth and vigor may increase 

 liability to infection. 



Testing has to be done under as widely different conditions as 

 required to sample all conditions where the varieties to be developed 

 shall be grown in the future. This may teach us something about the 



