50 CENTRAL AMERICAN RUBBER TREE. 



than where the surface of exposed soil is loosened by stirring and thus 

 forms a layer which hinders the access of dry air and is a noncon- 

 ductor of heat. In previous discussions of shade in the culture of 

 Castilla this distinction between open culture and clean culture seems 

 to have been overlooked, and the question of shade has continued to 

 be confused with that of water supply. The statements of various 

 writers that the leaves are unable to withstand exposure to the full 

 sun because of their delicate texture are quite erroneous. The tree 

 needs sunlight, and is benefited by it as long as the water supply is 

 sufficient, but when this becomes deficient the leaves shrivel. The 

 light is no brighter and the temperature no higher in the dry season, 

 which in Mexico occurs in the winter months; but the dry atmos- 

 phere demands more water, while the soil supplies less. 



The rapidity with which diy atmosphere takes water from a plant 

 may be judged by the promptness with which the leaves of a broken 

 branch wilt and shrivel, and this happens very promptly with Castilla. 

 Many plants have developed no expedients for resisting evaporation 

 and are accordingly confined to continuously humid regions, but Cas- 

 tilla, as has already been seen, is adapted in several ways for resisting 

 drought. The leaves themselves are, it is true, of rather loose texture 

 and have only the slight assistance of the hairs of the lower surface as 

 a protection against excessive transpiration. The leaves suffer when 

 they are obliged to part with more water than they can obtain, and 

 their falling off is then an advantage because it decreases the demand 

 for water. Thus, although Castilla is not a desert plant, the falling of 

 its leaves in the dry season is the same physiological phenomenon 

 which appears so conspicuously in deserts, viz, the loss of the leaves 

 as a protection against drought. Many desert plants such as Parkin- 

 sonia, Fouquieria, Peireskia, and species of Euphorbia put out leaves 

 for the wet season only, while most of the Cactaceje and many 

 Euphorbias have discarded leaves entirely and expose as little surface 

 as possible to the air. 



This digression may help to make it apparent that the planter who 

 desires to give intelligent consideration to the agricultural question of 

 shade should dismiss the notion that the rubber tree derives a direct 

 advantage from standing in the shadow of another tree; on the con- 

 trary, it is probable that interference with the sunlight is always a 

 direct disadvantage. Shade, if used at all, is to be applied and justi- 

 fied on the ground that it will preserve the moisture of the soil or of 

 the atmosphere or serve some other cultural purpose. B}^ conserving 

 the soil moisture, clean culture may produce some of the desirable 

 effects commonly ascribed to shade. Open culture may be, and prob- 

 ably is, less advisable than either clean culture or a moderate shade 

 culture. 



Open culture with relatively little cleaning at first would be more 



