SEASON FOR PIRIUa TRACES. 433 



sufficiently effective for the time being before there is any danger 

 at all of fire entering from outside, -while the latter need not be 

 ready until just before the grass inside the forest has become dan- 

 gerous. 



Taking up first the case of boundary traces, we note that ia 

 preparing them one of the three following distinct sets of condi- 

 tions has to be dealt with : — 



(1) The trace may he flanked hy a road or a watercourse, which 

 thus itself constitutes a part of the width of the trace, Now we 

 know that during the early part of the fire-conservancy season, 

 conflagrations can neither spread rapidly nor are so violent as not 

 to be stopped by a comparatively slight obstacle. Hence there is 

 no need for any hurry, and, uidess strong constant winds blow 

 towards the forest, the trace need not be fired until most or all of 

 the grass on it has become inflammable. 



(2 The grass on the entire or partial width of the trace may dry 

 up appreciably earlier than in the forest. This may be due to the 

 trace being on a well drained ridge, or to the fact that the bound- 

 ary of the forest is demarcated by a cleared line exposed to every 

 weather extreme while the area to be preserved is covered with 

 more or less canopied forest. Here also there is no need for hurry 

 and the firing may be delayed until all or most of the grass on the 

 "trace, or at least on the portion of it devoid of tree-growth, is fit 

 to bum. We have a special and the simplest instance of the 

 present case when the area outside the forest is not to be protected 

 against fire and the grass on it dries up early. 



(3) The grass on tJie trace may not become inflammoMe hefore 

 that immediately inside the forest. This is a real difficulty to over- 

 come, especially if the grass on the trace, from being in moisture 

 soil, remains green longer than in the adjoining protected belt. 

 If there is no cleared Hne to mark the boundary of the forest, one 

 should be made without delay so as to bring about, if possible, the 

 earlier drying up of the giass on the trace. But if that measure 

 should prove ineffectual or insufficient, one or both of two remain- 

 ing expedients may be adopted. The green standing grass on the 

 cleared portion of the trace may be beaten and crushed down with 

 poles or hght wooden rollers (according to the broken or even 

 nature of the ground), so that it may dry early and bum more 

 easily and completely in its now recumbent position. This opera- 

 tion may be found too costly and not indispensable, but the follow- 

 ing simple device, carried out with care and intelligence, will 

 answer in every case. Starting originally with somewhat wider 



