Figure 5. — Sporadic flaming typified the 15-ton-per-aare plots; this was probably caused 

 by the relatively small amount of fuel available and the olumpiness of particles. 



Discontinuity of fuel and heat flux. — Deviations were highest for the lowest values 

 of packing ratio and fuel loading (table 5) . This was especially true for the ponderosa 

 pine plots. A discontinuity in the arrangement of fuel particles and a corresponding 

 decline in the supply of heat for preignition could have existed because of the spac- 

 ing between fuel particles. The spacing was greater in the least compact fuels: the 

 branching habit and manner of needle growth created various sized gaps between particles, 

 even though the slash was distributed uniformly during construction of the plots (fig. 

 2) . Some gaps were probably so large that a substantial propagating heat flux was re- 

 quired to ignite unburned particles. In contrast, the mathematical spread model was 

 developed primarily using fuel beds having evenly spaced particles without any large 

 gaps between them. 



Deviations for the plots having the highest loadings were less, possibly because 

 the higher fire intensities furnished ample heat flux for a nearly uniform rate of 

 particle ignitions. Fires in the low loading plots generally burned more sporadically 

 because particle ignitions were not occurring at a uniform rate (fig. 5). 



On one of the low loading Douglas-fir plots (no needles present) fire would not 

 sustain itself; on another similar plot, fire spread was barely sustained. The amount 

 of fuel, particle size, and particle spacing were the dominant limiting factors. The 

 spread model contains mathematically continuous functions that are defined by scant 

 data at the limiting end of the fire scale. 



15 



