Although switchgrass (Panicum virgatum) is expected to be used as a biofuel crop on marginal land, there has been little to no extensive research to evaluate its performance on marginal land.  The objectives of this study were to determine the influence of soil type on bioenergy traits including lodging, tiller number, plant height, and anthracnose (Colletotrichum navitas) and rust (Puccinia sp) diseases to determine optimum selection procedures, broad sense heritability, and gain from selection. Five clones from 10 different populations, representing both upland and lowland cultivars, were vegetatively propagated to obtain six replications which were planted in a randomized complete block design at each of three locations in NJ in June 2009.  The three sites included two marginal sites - Somerset County, NJ (class IV Kleinsville shale) and Jackson, NJ (class V Evesboro sand), and a prime farmland site in Freehold (Adelphia), NJ (class II Freehold sandy loam).  Differences were observed between switchgrass clones for lodging, tiller number, plant height, and anthracnose and rust resistance.  Clones exhibited less lodging at the two marginal sites which could be due to the shorter plant heights observed at the marginal sites.  Clones showed similar relative levels of anthracnose and rust symptoms across all three locations.  Overall, clones at the prime site (Freehold) showed the highest number of tillers while clones at Jackson had the lowest tiller numbers.  A significant location by clone interaction was observed for all traits evaluated; however, clones with good performance tended to rate well at other locations and vice versa indicating that selection for biomass traits can be made on prime soil with the expectation that selected individuals will also perform well on marginal lands.  Data will continue to be taken in 2011 and heritability estimates will be calculated.