Both surface area and volume were measured on the same fuel specimen to aid in accu- 

 rately determining surface-to-volume ratios. Sample material was at a moisture content 

 of 5 to 8 percent. For long, narrow particles (pine needles, cheatgrass stalks, 

 peduncles, awns, and spikelets) , volume and surface area were determined using measure- 

 ments of particle cross-sectional area, perimeter, and length. Cross-sectional area 

 and perimeter of pine needles were measured using photomicrographs of right-angled 

 cross sections. Diameters measured microscopically permitted calculation of cross- 

 sectional area and perimeter of the cheatgrass particles. 



For flat particles (cheatgrass leaves and glumes and miscellaneous leaves in the 

 forest floor), volume and surface area were determined using measurements of leaf 

 thickness and blueprint outlines of leaf areas. 



An electrolytic technique similar to that developed by Tibbals et al . (1964) em- 

 ploying silver castings was used to estimate surface area for cone scales. Volume was 

 determined by displacement in water. Volume and surface area for the staminate flowers, 

 bark flakes, and grass were determined from counts of number of particles in each plot 

 and estimates of their average dimensions. For complete details on measurement of 

 volume and surface area see Brown (1970). 



Particle density was determined using ovendry weights and airdry (5- to 8-percent 

 moisture content) volumes. 



CALCULATIONS OF (r\, PARTICLE SPACING, AND ORIENTATION 



Most of the properties are calculated from measurements in a manner obvious from 

 their definitions. Calculations of oA , particle spacing, and orientation, however, are 

 not so obvious. It is necessary to determine an average a in calculating aX for fuel 

 complexes, where more than one particle size is present. In this study, where six to 

 eight different particles existed, o was calculated as an average, weighted by volume 

 of each type of particle present. 



Spacing of particles, as viewed in the sampling plane, is computed from 



Sp = V~Vd"- VT (1) 



where: 



Sp = average particle spacing, 

 Vd = void space per particle, and 

 A = average cross-sectional area of particles. 



It is assumed that particles are square in cross section and that void space is 

 distributed evenly about each particle. Solution of equation (1) requires knowledge 

 of average right cross-sectional area of particles and sample information on the number 

 of particle intersections with a sampling plane. Void space per particle equals the 

 area of sampling plane divided by number of particle intersections. 



The following two hypotheses were subjected to the chi-square test for goodness-of- 

 fit for evaluating randomness of particle orientation: (1) Ponderosa pine needles are 

 randomly oriented in a horizontal plane; (2) cheatgrass particles are randomly 

 oriented in a vertical plane. 



The number of particle intersections in each of the six 30°-wide sectors arcing 

 about one side of the sampling plane was compared with the expected number of particle 

 intersections in these sectors. 



7 



