PROTECTION OF NASCENT ORGANS. 53 



correlated, like the resinous exudate of glanduliferous plants, with the 

 relative importance of the tran spiratory functions of these organs. 



I have not yet accounted for a certain scant hairiness that occurs in 

 some of the desert shrubs. The leaves of Aster mohavensis and Grayia 

 spinosa are described above as bearing a few scattered hairs. Although 

 very different in form, they possess the characteristics that have been 

 mentioned as common to the individual hairsof the densely coated spe- 

 cies enumerated in the preceding paragraph. To repeat, they consist of 

 dry cells containing only air, and they are adapted, therefore, to be non- 

 conductors of heat. Scattered thinly over the surface of the leaf, how- 

 ever, they provide a very unimportant protection to it. If now a leaf 

 in its young stage be examined, it is found, conspicuously in the case of 

 Grayia spinosa, that the hairs attain their full size early, while the leaf 

 is still very little expanded. In this stage of the leaf's development 

 the small number of hairs is ampJy sufficient to cover it and to prevent the 

 moisture of its tissues from evaporating. Later in the development of 

 the leaf, during the period of rapid spring transpiration, when the 

 epidermal tissues have become strong, the covering is not needed, and 

 still later, in the arid months, the leaves dry and drop off. In several 

 other desert shrubs of the deciduous-leafed type, the same character of 

 hairiness is exhibited, and the same function is evident. 



In connection with this subject of the protection of nascent organs I 

 wish to point out the fact that scaly buds are almost unknown in these 

 desert shrubs. The form of protection that has been described above, 

 or some equivalent in epidermal modification, is almost universal in 

 these plants. In a humid climate the familiar type of winter bud, both 

 in trees and in shrubs, is a bud covered and protected by scales. Only 

 in rare instances, for example in Gormts fiorida and the species of Vibur- 

 num, do the organs that constitute the outer covering of winter buds 

 develop later into organs with other functions. In a large majority of 

 cases the bud-covering is made up of scales which, after the bursting 

 of the bud in spring, are incapable of further development and drop 

 away. The typical leaf in a humid climate is large, thin, bright green, 

 and provided with a film-like epidermis. The development of an in- 

 durated bud-scale, which has been exposed for months to wide ex- 

 tremes of temperature and sunlight, into such a delicate structure as 

 that just described, is hardly compatible with the principles of leaf 

 growth and is rarely accomplished. We have therefore in humid 

 climates a sharp differentiation between bud scales and foliage leaves. 

 In the case of desert shrubs the same influencing conditions do not 

 exist. The environment of the latter consists essentially of extremes in 

 temperature combined with aridity of the atmosphere. The close, hairy 

 coating of nascent leaves, which has been described above, and which 

 exists all the more effectively in leaves that are permanently covered 

 with hairs, is entirely adequate to resist these conditions; and since in 

 spring and summer a similar covering, or at least an indurated epidermis, 



