moisture content. This behavior substantiates the theory that the effective holding time of 

 short-term retardants is limited only by the total amount of moisture on the fuel. 



No such Limitation exists for the long-term retardant. When the total moisture content 

 was reduced to within 1 or 2 percent of the untreated fuel moisture level (dashed line, fig. 11), 

 the rate of spread was still well below the rate of spread for untreated fuel at the same mois - 

 ture level. Thus the limitations of solar heating and fire heating mentioned earlier will not 

 shorten the effective holding time of long-term retardants as they would short-term retardants. 

 These facts should be considered seriously in the purchase of retardants, together with relative 

 costs of short-term and long-term retardants. 



The effect of initial application amount of long-term retardants is shown in figure 12, 

 where the actual rate of spread is plotted against an approximation of the equivalent National 

 Spread Index. The data now separate and align and best illustrate the effect of initial applica- 

 tion amount and retardant drying. Note the marked difference in slope of the Lines for initial 

 application amounts of 1 gal. /1 00 sq.ft. and 2 gal. /1 00 sq.ft. For the fuel loading used, an 

 initial concentration of 1 gal. /1 00 sq.ft. was not sufficient to effectively retard the fire. The 

 rate of spread increased sharply with both Spread Index and retardant dryness. However, when 

 the concentration is doubled, it can be seen to be effective even though the Spread Index in- 

 creases and the retardant loses 95 percent of its moisture. Whatever fire propagation occurred 

 when 2 gal. /1 00 sq.ft. was applied took place near the bottom of the fuel bed, where the re- 

 tardant did not penetrate . 



The "t" tests confirm these observations. Drying time was significantly different when 

 only 1 gal./lOO sq.ft. was applied, but was not significant when 2 gal./lOO sq.ft. were applied. 

 Where 2 gal./lOO sq.ft. were initially applied and the long-term retardant was fully dry (95 

 percent), the difference due to environment was not significant. In every case, a high 

 significance level was found for differences in initial application amount. 



This result should not be entirely unexpected since the long-term retardant contains a 

 fire -inhibiting salt which must contact the fuel if it is to alter the combustion characteristics of 

 the fuel. A certain minimum ratio of retardant to fuel may be expected to exist, and appli- 

 cations below this amount should not be expected to provide sufficient treatment to enough fuel 

 to be effective in suppressing the fire. Our work indicates the ratio for the initial amount of 

 long-term retardant is near 0.4 pound of retardant per pound of dead fuel. This ratio should, 

 of course, be put in terms of the necessary ratio of retardant salt to dry fuel required to hold 

 a fire. Such a ratio would provide a basis of comparison for long-term retardants. 



CONCLUSIONS 



1 . All five of the retardants tested had similar drying rates within each of nine drying 

 combinations . 



2. Rates of spread in fuel beds treated with long-term retardants were well below the 

 value which might be expected if moisture alone were causing the effectiveness. In contrast, 

 rate of spread in fuels treated with short-term retardants appears to depend entirely upon total 

 moisture retained. 



3. Long-term retardants are effective even after their moisture has evaporated when the 

 initial amount of retardant is sufficient. Short-term retardants remain effective only when the 

 moisture retained around or in the fuel is at least 22 percent of the dry weight of the fuel. 



25 



