Resumen. En la Estación Experimental de la Universidad de Nebraska, se llevó a cabo un estudio del progreso de selección masal en dos sub-poblaciones de la variedad de maíz Hays Golden. Los objetivos y resultados fueron: 1. Estimar la ganancia total en rendimiento después de ocho ciclos de selección masal en las dos sub-poblaciones. El octavo ciclo de selección en la sub-población "control" rindió 46% más que la variedad original. El mismo octavo ciclo de la sub-población "irradiada" fue 10% superior a Hays Golden. 2. Estudiar el comportamiento de los cruzamientos entre las generaciones de ciclos con la variedad original. Hubo aumentos progresivos de 6, 17, 18, Y 19% sobre la variedad original en el rendimiento de los ciclos 2, 4, 6, Y 8 respectivamente, en la sub-población "control". En la sub-población "irradiada" las correspondientes respuestas fueron 8, 4, 20 y 18%. 3. Estudiar el comportamiento y la respuesta heterótica cuando se intercruzaron los ciclos de ambas sub-poblaciones. Hubo un progreso continuo del rendimiento de 11, 13, 15, Y 22% sobre la variedad original para los cruzamientos de ciclos comparables en ambas subpoblaciones: Sin embargo, no hubo respuesta heterótica en esos cruzamientos. De los resultados se concluyó que la selección resultó en un aumento de homocigocidad de loci para rendimiento, que se tradujo en más productividad en los cruzamientos entre ciclos y entre éstos con la variedad original, aunque con un efecto heterótico bajo que sugiere que la selección practicada por separado en ambas sub-poblaciones (hasta el octavo ciclo) no resultó en diversidad genética. /abstract. The objectives of the present study were: (a) to estimate the total gain in grain yield through eight cycles of mass selection in two subpopulations of Hays Golden variety; (b) to compare the gains and behavior between these two selected subpopulations; (c) to study the behavior of the crosses of the cycled generations with the original variety and (d) to study the behavior of the heterotic response when the cycled generations in both derived populations are intercrossed. It was apparent that mass selection was successful in changing the frequency of favorable genes for yield in both subpopulations. This resulted in a substancial increase in their productivity; 4,42 lbs./ plot (46%) in the control mass cycle 8. and 3,86 lbs./plot (40%) in the irradiated mass cycle 8 population. As suggested by the highly significant upward linear trend, it was concluded that: (a) no substancial reduction in the additive portion of the genetic variance has occurred; b) continued progress in grain yield should be expected in further cycles of mass selection. A comparable overall performance between control mass cycles and irradiated mass cycles was observed, both as populations and in crosses with the original Hays Golden variety. A low heterotic response was obtained in the cycle intercrosses although an upper linear trend was shown for yielding ability. These results suggest that selection has resulted in an increase in homozigosity at loci affecting yield. This in turn, was translated into more productivity in the intercrosses between more advanced cycles and in a lower heterotic effect. It was concluded that separate selection in both subpopulatlons (up to the eighth cycle) resulted in no increased genetic -diversity. In general, with increases in yield there was an increase in ear height (r =.9139) in moisture of the grain at harvest (r =.4428), and in flowering date (r = .6817). As expected, moisture in the grain at harvest and flowering date were correlated (r = .3908). Stalk Iodging and dropped ears aIso increased with increases in yield; this was expected slnce no selection for these characters was performed and the heavier grain yields increased the degree on stress on the plants.