Maintaining genetic variation in breeding populations of Radiata pine in New Zealand

Abstract Advanced generation selection (AS) for the future breeding population (BP), becam a focus of tree breeders‘ thinking in the mid 1970s., particularly with Pinus radiata in New Zealand (NZ). Multitrait selection among families was generally recommended, but this reduced genetic variation in the future breeding population. From Shaw and Hood‘s (1985) stochastic simulation, later confirmed by Rosvall, Lindgren and Mullin‘s (1998) stochastic simulation on Norway spruce, it was realised that selecting within families rather than among families of a new breeding population avoided any reduction of genetic variation in the BP. Heritabilities were low for seedling within-family selection but clonal replication within families should strongly increase heritabilities. Gains from cloned versus seedling populations of equal numbers of plants were also deterministically simulated (Shelbourne et al. 2007), and balanced (within-family) selection gains from the cloned populations were all higher than seedling equivalents at heritabilities of 0.5 and under. The late P.A. Jefferson‘s (2016) Breeding Management Plan (which will be soon superceded) contains a re description of New Zealand (NZ) radiata pine breeding. Selections were made in crosses from the earlier program and OP see and scion material were collected from all 360 selections. OP family tests of selections have been planted at 11 sites in NZ and 7 in New South Wales and Tasmania, and scions of their female parents have all been grafted at an archive. Crosses made in the archive are being cloned and the programme was committed to within-family selection to retain genetic variance for the future closed breeding population. Clonally-replicated testing paired with within-family selection is the solution for balancing long-term gain and diversity in BP and PP.