Two competing models explain patterns of species performance relative to resource gradients: The fundamental niche differentiation model and the shifting competitive hierarchy model.  The performance of three historically, co-occurring canopy tree species, tuliptree (Liriodendron tulipifera), chestnut oak (Quercus prinus), and American chestnut (Castanea dentata), were used to test the two competing models.  Our null hypothesis is that seedling performance of American chestnut, blight-resistant American-Chinese hybrids (BC₂-F₃, BC₁-F₃), chestnut oak, and tuliptree do not differ as a function of resource gradients.   To test the seedlings’ response to different light intensities, shade cloth (50% and 80%) was used in a greenhouse environment to mimic light levels in experimental forest gaps.  Seeds were planted in April 2007 in one gallon pots containing soil collected from ridges or slopes.   Seedling height, diameter, leaf number and chlorophyll levels were measured in September 2007 and 2008.  Seedling growth data was analyzed by performing a three-way ANOVA test for each dependent variable in which light intensity, soil type, and species were independent fixed variables. 

Results/Conclusions

Light had a significant effect on height, diameter and leaf number (p < 0.0001) with 50% light resulting in greater growth.  The interaction of light and soil had a significant effect on the height and diameter of seedlings, most clearly shown by diameter (p <0.05).  Overall, seedlings performed best in 50% light, regardless of soil type.  However, in 20% light, seedlings had greater growth rates in slope soil.   Pure chestnuts and hybrids were not significantly different from each other although they were significantly different from chestnut oak in all treatments and from tuliptree under 20% light in ridge soil.  Pure and hybrid chestnuts were the tallest seedlings after two growing seasons in optimal conditions (µ = 75 cm).  Results suggest that American chestnut and hybrids do not differ in terms of their ecological response to environmental conditions.  In addition, they outperform their two potential competitors, tuliptree and chestnut oak in low light levels on ridge soil.  All species preformed best under the same environmental conditions, lending support to the shifting competitive hierarchy model. This experiment has implications for the restoration of the American chestnut in terms of predicting seedling performance in a heterogeneous environment.