Biomass particle size impacts handling, storage, conversion, and dust control systems. Size reduction mechanical energy was directly measured for switchgrass (Panicum virgatum L), wheat straw (Triticum aestivum L), and corn stover (Zea mays L.) in an instrumented hammer mill. Direct energy inputs were determined for hammer mill operating speeds from 2000 to 3600 rpm for 3.2 mm integral classifying screen and mass input rate of 2.5 kg/min with 90 degrees- and 30 degrees-hammers. Overall accuracy of specific energy measurement was calculated as +/-0.072 MJ/Mg. Particle size distributions created by hammer mill were determined for mill operating factors using ISO sieve sizes from 4.75 to 0.02 mm in conjunction with Ro-Tap (R) sieve analyzer. A wide range of analytical descriptors were examined to mathematically represent the range of particle sizes in the distributions. Total specific energy (MJ/Mg) was defined as size reduction energy to operate the hammer mill plus that imparted to biomass. Effective specific energy was defined as energy imparted to biomass. Total specific energy for switchgrass, wheat straw, and corn stover grinding increased by 37, 30, and 45% from 114.4, 125.1, and 103.7 MJ/Mg, respectively, with an increase in hammer mill speed from 2000 to 3600 rpm for 90 degrees-hammers. Corresponding total specific energy per unit size reduction was 14.9, 19.7, and 13.5 MJ/Mg mm, respectively. Effective specific energy of 90 degrees-hammers decreased marginally for switchgrass and considerably for wheat straw and it increased for corn stover with an increase in speed from 2000 to 3600 rpm. However, effective specific energy increased with speed to a certain extent and then decreased for 30 degrees-hammers. Rosin-Rammler equation fitted the size distribution data with R-2>0.995. Mass relative span was greater than 1, which indicated a wide distribution of particle sizes. Hammer milling of switchgrass, wheat straw, and corn stover with 3.2 mm screen resulted in 'well-graded fine-skewed mesokurtic' particles. Uniformity coefficient was <4.0 for wheat straw, which indicated uniform mix of particles, and it was about 4.0 for switchgrass and corn stover, which indicated a moderate assortment of particles. Size-related parameters, namely, geometric mean diameter, Rosin-Rammler size parameter, median diameter, and effective size had strong correlation among themselves and good negative correlation with speed. Distribution-related parameters, namely, Rosin-Rammler distribution parameter, mass relative span, inclusive graphic skewness, graphic kurtosis, uniformity index. uniformity coefficient, coefficient of gradation and distribution geometric standard deviation had strong correlation among themselves and a weak correlation with mill speed. Results of this extensive analysis of specific energy and particle sizes can be applied to selection of hammer mill operating factors to produce a particular size of switchgrass, wheat straw, and corn stover grind, and will serve as a guide for relations among the energy and various analytic descriptors of biomass particle distributions. (C) 2009 Elsevier B.V. All rights reserved.