102 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



system, as was originally suggested for the northern ridge. This 

 consists, as already partly described, in making use of all natural 

 depressions wherever earth is being deposited and not subject to 

 erosion. Natural depressions or pockets are excavated three to four 

 feet in diameter if the surrounding rocks permit, and to corresponding 

 depth if possible, wherever porous disintegrated stone is found, the 

 pit (see diagram A) is dug out and refilled with clay from below. 



The pits or holes are excavated on an average 25 feet apart, which 

 gives an average of seventy pits to the acre. Each of the pits is planted 

 with a hardy pot- or pipe-grown tree, supported by three hardy 

 similarly grown undergrowths, and beneath them again seeds of 

 hardy species are sown at the beginning of the rains. It will be seen 

 that each pit really accommodates four nursery- grown saplings. 

 The process was arrived at after repeated failures to get seeds to 

 germinate and thrive when afforestation was to be conducted by 

 sowings. In many instances the main tree or one of the undergrowths 

 thrives indifferently, but what actually remains meets the objects 

 aimed at. There is no doubt that a certain amount of thinning out 

 will have eventually to be done, but this will not be necessary for 

 years, as the rate of growth under such adverse conditions is relatively 

 slow. It may be added that most of these pits are surrounded by 

 Opuntia Dillenii (fig. 10) as a protection against porcupines and 

 hares ; rats are easily disposed of with arsenic, and as to white ants — 

 there is very little to be done as a protective measure — the best is to 

 eliminate the species that are usually attacked by them. 



Ravines (see diagram p. 105) will have to be gradually trained 

 to stop as much as it is practically possible the " run off " by a 

 series of bunds that will eventually prevent the disintegration of the 

 sides and collect the erosion from the upland, thus eventually forming 

 so many small tablelands. The sides of these ravines being in some 

 cases almost perpendicular, it is proposed to clothe them with suc- 

 culents such as Aloes, Agaves, Fourcroya, Euphorbia, and Opuntia, 

 which are gradually being collected and grown for the purpose. These 

 ravines, when established, will create a most pleasant contrast. It 

 may be interesting to record that, without artificial watering, it would 

 be an impossibility to carry out the work in the time it ought to be 

 completed, and it would be best to leave Nature to do its work, 

 though it may take half a century or more. 



The method aimed at, is to tend these plantations for a period of 

 four or five years (according to their position), by which time the vegeta- 

 tion should be sufficiently established to thrive by itself. These 

 artificial waterings will gradually diminish, so that by the end of the 

 fourth or the fifth year, as the case may be, when the monsoon sets 

 in, the vegetation will be left to thrive by itself. 



The distribution of water is, of course, one of the most difficult 

 problems. Take, for instance, the watering of tree-pits at the rate of 

 seventy an acre, amongst rocks and thorns over an area that will eventu- 

 ally cover approximately 2,500 acres. This will convey an idea ot the 



