196 SCIENCE IN AFRICA 



five years. It is clear that prompt action is required in reducing 

 the rate of cutting, but quite apart from this, it is probable that a 

 much larger area of perinanent forest than the 8,000 square miles 

 anticipated would prove a source of wealth to the country in the 

 future. 



The peculiar physiographical and meteorological conditions of 

 the Gold Coast lend themselves to a special selection of areas. The 

 object of the department is to develop reserves in such a way that 

 (i) all the principal water-sheds will lie in forest reserves; (2) there 

 will be a belt of permanent forest on the hilly escarpment which 

 forms the north-eastern limit of closed forest and high rainfall, and 

 is the water-shed between the Pra to the south and tributaries of 

 the Volta to the north; (3) there will be a series of shelter belts in 

 the closed forest region, which is also the region of cacao cultiva- 

 tion; these will run parallel with each other at right angles to the 

 direction of the south-west monsoon; (4) small reserves will be 

 made near the townships to supply local timber for building and 

 for firewood. To provide a basis for the selection of areas the Direc- 

 tor of Forestry, Mr. Marshall, recommends^ the establishment of 

 additional meteorological stations in the closed forest region, and 

 also in the area of savannah forest north of the escarpment where 

 rainfall stations are at present very few. 



The provision of shelter-belts is especially important in the Gold 

 Coast in the interest of the cacao industry on which so much of that 

 country's prosperity depends. Many cacao plantations have 

 ceased production as a result of forest destruction in the neigh- 

 bourhood and the ensuing reduction of humidity around the cacao 

 trees. The exact effect of this exposure appears to be a problem of 

 ecology and microclimatology comparable with that of coffee 

 plantations referred to in Chapter IV. For cacao it is coming to 

 be recognized that small 'reticulate' shelter-belts among the plan- 

 tations and the raising of shade-trees are likely to be of more imme- 

 diate benefit than larger shelter-belts at wide intervals. A recent 

 article by H. W. Moor (1937) is of interest in relation to this and 

 other problems of forestry in the Gold Coast. 



The principle of the shelter-belt in forestry has been developed 

 especially in America, and some of the conclusions reached there, 



' Private communication. 



