Rubber seedlings, the product of seed gardens, have been grown successfully for decades. Such populations are polycrosses or first generation synthetic varieties (SYN1). The genetics of SYN1 in rubber have not been investigated but some (rather speculative) calculations are possible of fragmentary data. It is likely: that parents are unequally represented among progeny; that crossing is very imperfect but that the deleterious effects of inbreeding are partly mitigated by seedling culling; that, nevertheless, SYN1 suffers from some inbreeding depression; that second generation seedlings (SYN2) offer interesting practical potential worthy of investigation; and that seed-garden design deserves far more attention that it has had. Suggestions for relevant experiments are made and it is pointed out that isozyme genes would offer a powerful method of analysing the genetical dynamics of seed gardens and their products.