Nevertheless, such exceptional cases do not detract from the wisdom of clean weeding in general. Every planter of experience realises how easily fields become infested with weeds if the regular work is suspended or delayed. It is probably quite true that the harm due to the presence of some weeds on an occasion is negligible; but apart from this debatable point, there is the solid fact that if once an area is allowed to become weedy it may soon demand a much greater expenditure to bring it back to normal condition than if it had been regularly weeded. This is common experience, and for that reason alone a general policy of clean weeding is thoroughly sound; especially if combined with some system of silt-retention. GRASS SQUARES.--On some estates the practice of clean weeding is undertaken in combination with a system of silt-retention, which depends upon the development and maintenance of ridges. These are built up from the débris of weeding in the form of hollow squares. Grass is allowed to sprout and grow in these ridges, and when it attains a certain height it is trimmed down so as to keep it within bounds. The soil within the hollow square is clean weeded; and it is maintained that loss of soil by wash is avoided. Under certain conditions there is a great deal to be said in favour of the method, but in the opinion of the writers it should be regarded only as a method of expediency. It is not to be preferred to the more thorough practice of soil-retention by means of silt-trenches, although the latter method may be slightly more expensive in the end. [Illustration: TYPICAL YOUNG CLEARING, WITH TIMBER. Planted "rubber-stump" in foreground.] "LALLANG" ERADICATION.--The greatest bugbear of the planter in connection with weeding is the incidence of lallang. Many proposals have been put forward at various times for the complete eradication of this pest; but at present, under ordinary circumstances, there would seem to be no better method than by heavy and deep digging, followed by regular attention. The method is acknowledged to be expensive, but any half-hearted measure otherwise taken will eventually prove to be even more costly. One has to differentiate, of course, between the incidence of lallang attributable to negligence on the estate itself, and the occasional outbreaks near boundaries, due to seeds having been wind-borne from patches of lallang outside the boundaries But, in general, it would be safe to remark that the appearance of lallang could be taken as evidence of a failure to cover the area at sufficiently short intervals. As already intimated, the usual method of eradication of areas of lallang is by thorough digging, and the exposure of the strong root system to the sun. As a matter of interest it may be noted that recently some success has been obtained by another method[1] on areas which one may have in view for planting at some future date.