ACTUAL EIRINQ OP THE TEACEfl. 439 



and not far from, the first be now lit, each line of flames, instead 

 of rising np straight, -will bend towards the other nnder the 

 influence of the markedly preponderating draught from the off 

 side, the space between the twO' lines being obvionsly too narrow 

 for the production of any appreciable draught towards either fire. 

 Next, let ns suppose that a wind is blowing while a single line is 

 burning ; the flames must in that case be bent forward in the 

 direction of the wind the deflection fron* the vertical being propor- 

 tionate to the velocity of the latter. If the second line of fire is 

 now lit to windward of the first, the wind will, on reaching this 

 new line,^ be more or less forced upwards, partly owing to a rise of 

 temperature, partly and more especially inconsequence of encounr 

 tering the broad ascending column of air over the two fires. So 

 much of it as is able to traverse the new line of fire and reach the 

 first across the narrow interval separating the two lines will have 

 too high a temperature to receive much acceleration from the 

 combustion of this farther line, and even this acceleration must, 

 as the two fires approached each- othec, diminish to such an extent 

 as to be unable to oppose the draught of air coming in from 

 the other side, which would then be ultimately strong enough to 

 bend the flames in its own direction. We thus see that by a very 

 simple and inexpensive device the- long flames, which rise whea 

 the cushion of dry grass along, the edge of a trace is set on fire^ 

 can be turned away from the broonL-men even, if a contrary wind 

 is blowing. 



From the above considerations we derive the following practical 

 rules : — K the air is still, an inner parallel line of fire lit a 

 few yards away from each edge of the fire-trace will cause the 

 fire along the edge to bend away from the broom-men. A single 

 line along the middle of the trace will, however, suffice if the 

 width of grass to be burnt does not exceed, say, 40 feet, and none 

 at all will be necessary if it should be less than half that figure. 

 If a wind is blowing across the trace, an inner line of fire should 

 be run dose to the lee edge of the trace ;, no. suck line will be re- 

 quired for the opposite side of the trace^ Lastly, it must not be 

 forgotten that i^ during more or less still weather,, a line of fire is 

 drawn round any area of small extent, converg^ing currents of air 

 will be produced which will bend all the flames inwards and hasten 

 the complete burning of the enclosed area. 



During the firing of a trace one or two men should always fol- 

 low the firing party at a short distance behind, in order to put out 

 or brash or throw away still smouldering fire that is in dangerous 



