Inequalities in size in populations have potentially important effects on fitness but have rarely been examined in natural populations. I measured size (number of culms) and fecundity (number of spikelets) in five populations of the grass Danthonia spicata from 1981 to 1985. The populations were in sites in a Pinus-Quercus-Populus forest in northern lower Michigan, USA comprising a secondary succession sequence. The sites had been burned in 1980, 1954, 1948, 1936, and 1911, respectively. Mean sizes and fecundities and the amount of hierarchy in size and fecundity, measured by the Gini coefficient, were compared between the YOUNG population, 1980 burn site, and the old populations, 1954, 1948, 1936, and 1911 burn sites. I found large differences in mean size and fecundity between the YOUNG and OLD populations with much larger individuals in the YOUNG population. No differences in size hierarchies were found in either the first year of measurement or after five years. The fecundity hierarchies showed no significant difference among the populations in the first year but after five years the YOUNG population showed a significant decrease in amount of inequality. The long-term patterns of size and fecundity hierarchies differed because fecundity was a cumulative trait while size was not. Size inequalities may not always be a good measure of fecundity inequalities. Short-term measures of inequality in perennials may not be a good indicator of long-term values. In contrast to greenhouse studies, habitat light levels did not affect size hierarchies although they did affect fecundity hierarchies.