449 



Plot F. Sprayed u> solution arsenite of soda; 

 „ G. Treated i\j solution arsenite of soda applied 



through carpet machine; 

 „ H. Changkolled and roots picked. 



Notes on Above. — Plots B and F were treated alike, 

 they were sprayed on ten occasions and burnt over twice. 



The effect of two sprayings was spoiled by rain com- 

 ing on soon after the spraying was finished. The lalang 

 appears to have weakened in growth considerably, but looks 

 as if it would still take some time to kill out, and if left alone 

 for a short time would again grow vigorously. 



. Plot C shows very little difference from B and F, but 

 is if anything less affected by the treatment, it has been 

 noted that if a shower of rain happens to come after spray- 

 ing, the mixture has much less effect than in the case of 

 those sprayed with the stronger solution. 



Plot D. — In this plot the ground was changkolled over 

 on three occasions, in March, April and May, at a cost of 

 $')-'}. 90 per acre, coolies being paid 35 cents per day. After 

 three turnings the ground was practically clear of lalang 

 and with ordinary weeding could have been kept clean. 



The ground was wet from July to December without 

 any attention whatever, it was then turned over once more 

 at a cost of $9.80 per acre, the lalang even then was well in 

 hand. 



This apears to me to be the best and most practical 

 means of eradicating this pestilent weed, not only are the 

 weeds disturbed, but, in addition, the ground receives the 

 benefit of cultivation. The cost would probably be very 

 considerably cheapened if large areas were treated and 

 plough and harrow used in place of the ordinary changkol. 



Plot E. — The ground was changkolled over once, at a 

 cost of $16.75 per acre, and then planted with the wild 

 passion flower (Passifiora fcetida), the ground being high 

 and somewhat dry at times ; the growth of the passion 

 flower was extremely slow and cannot be considered at all 

 a success. 



Plot F, see Plot B. 



Plot G. — This plot was treated with one-tenth solution 

 of arsenite of soda, applied as follows: 



A box containing the solution, mounted on wheels, and 

 a blanket saturated with the solution, with one end in the 

 box, was dragged slowly over the grass in both directions. 

 The device when tried on level ground acts very well and 

 results in a considerable saving of solution. But for prac- 



