Field experiments were conducted at the University of Arkansas Agricultural Experiment Station  (Fayetteville) in 1997 and 1998 to determine if modified production environments, such as ultra-low plant densities, would affect soybean seed yield, viability, and vigor. Maturity group (MG) V soybean genotypes as main-plots and ultra-low plant populations as sub-plots were established using a Randomized Complete Block (RCB) split-plot design with three replications under irrigated and dryland conditions. Data for plant  height,  number of  plants m^-2, number of pods m^-2, number of seeds per pod, 100-seed weight, and seed yield m^-2 were collected at harvest. Standard germination tests and cold germination tests were conducted after harvest. Data were analyzed using the Statistical Analysis System (SAS). Seed yield per plant decreased as plant population increased regardless of irrigation treatment or genotype selection. In both years, soybean seed yield under irrigation was maintained above 1900 kg ha^-1 for plant population as low as 40,000 plants ha^-1. Irrigation improved soybean seed yield by an average of 74% regardless of year, genotype, or row spacing and ultra-low plant population combinations. The highest plant population, 160,000 plant ha^-1, with 0.5 m row spacing achieved the highest seed yield (2475 kg ha^-1) but was not significantly different from the seed yield of 2311 kg ha^-1 and 2212 kg ha^-1 for the plant populations of 80, 000 plants ha^-1 and 40,000 plants ha^-1, respectively, with the same row spacing (0.5 m). Although irrigated plots showed significantly higher  seed  viability  and  vigor  than dry land plots in 1998, inconsistency in seed quality was observed among ultra-low plant populations between the two years. Genotype, year and/or irrigation variables had more influence on soybean seed viability and vigor than the ultra-low plant populations. The yield compensation ability of full-season soybean (MG V and later maturing) genotypes from the mid-south should be considered as a major component in decision-making processes, especially, for replanting or for reducing seed cost of high priced soybeans such as Round-up Ready genotypes.