Laboratory-derived composition and pyrolysis data are essential inputs for the modeling of fire behavior. Recently, fire research has focused on live fuels including living wood and leaves, which exhibit sharp differences in moisture levels and chemical composition as compared to dead fuels. These leaf components have fuel properties that function in the spread of wildfire and must be considered in order to produce nuanced predictive models that reflect real life conditions. The goal of this study was to assemble a suite of methods that would achieve summative mass closure for analysis of live leaves from 12 tree and plant species from the southeastern United States (consisting of broadleaves, conifers, grasses, and palmettos). Most of the procedures used were adapted from standard methods commonly used for biomass analysis at the Forest Products Laboratory (FPL), National Renewable Energy Lab (NREL), and others. A mass closure (aka mass balance) of 95 to 100% was achieved for 10 of the 12 species the other 2 were both 91%. This required measuring of 12 parameters which are lipids, nonstructural sugars, protein, pectin, hemicellulose, cellulose, starch, phenol, structural lignin, silicates, and minerals. When burned, these components span a wide range of pyrolysis temperatures from 70 to 600 degrees C, posing a challenge for pyrolysis measurements. Observable differences in leaf composition were noted within groups of plant types as well as between them-with grasses being most similar and palmettos being most dissimilar. A rigorous statistical analysis was out of the scope of this study (involving analysis of 12 matrices of 12 plants); instead, validated standard analytical protocols (with known % error) were used in most cases. These error percentages (and sources) are reported with data presented in this study. The overriding goal of reaching summative mass closure (for all plant species) was however achieved, serving as a validation of the reported methodology for use in pyrolysis modeling. To predict the ultimate analysis data for leaf elemental composition and the heat of combustion, the empirical formula for each components was identified to provide for summation over all components. A formula for heat of combustion based on the oxygen consumption principle was found to be adequate to within 4% error.