Progeny testing is the most widely used method for intensive genetic evaluation of selected forest trees, but its major disadvantage is the length of time needed to get results. To circumvent this problem in selection for a multiproduct strain of slash pine (Pinuselliottii Engelm.), a system for short-term (3-year) progeny testing was developed. Oleoresin yields of selected second-generation 19-year-old parent trees were compared with yields of their 3-year-old offspring. Both parents and offspring showed little variation and a low offspring–parent correlation (r = 0.18), but other traits based on the same parent trees when they were 25 years old and their 3-year-old offspring showed relatively large amounts of variation and moderate to strong offspring-parent correlations: height, r =.56; volume, r = 0.43; turpentine, r = 0.71; ethanol–benzene extractives, r = 0.38; specific gravity, r = 0.43; moisture content, r = 0.34. Thus, indications are that additional gains could be made in the first generation by selecting among parent trees on a progeny performance basis in all traits except oleoresin yield. Original mass selection for oleoresin yield achieved gains of 100% improvement and apparently exploited almost all of the genetic variation in the base population. In second generation selection, progeny testing will be used primarily in screening for fusiform rust resistance. Between-family plus within-family selection will be used for other traits.