Yellow birch seeds were found to vary in size directly as the strobilus on which they are borne and to increase in size within the strobilus directly as their point of origin is removed from either end. Viability was found to vary from 0% to 34% of the seeds on an individual tree. The best method of determining viability is examination of the embryo following dissection of the seed. A valid test of viability is given by 2,3,5-triphenyltetrazolium chloride, but this is not recommended because of the small size of the embryos. It is shown that the seed coats contain a water soluble substance that inhibits growth of the embryos. This substance loses its inhibiting properties when exposed to extended periods of light, but the injury to the embryo is permanent. It is suggested that trees producing few viable seeds in a seed crop may do so because this species is not functionally monoecious and that pollination difficulties would account for a small number of viable seeds on partly isolated trees.