THE AMAZON VALLEY. 307 



any striking flowering tree or shrub. This is partly owing toi 

 the flowers of most tropical trees being quickly deciduous : they \ 

 no sooner open, than they begin to fall ; the Melastomas in ' 

 particular, generally burst into flower in the morning, and the 

 next day are withered, and for twelve months the tree bears 

 no more flowers. This will serve to explain why the tropical 

 flowering trees and shrubs do not make so much show as 

 might be expected. 



From the accounts of eye-witnesses, I believe that the forests 

 of the southern United States present a more gay and brilliant 

 appearance than those of tropical America. 



Humboldt, in his " Aspects of Nature," repeatedly remarks 

 on the contrast between the steppes of Tartary and the llanos 

 of the Orinooko. The former, in the temperate zone, are gay 

 with the most brilliant flowers ; while the latter, in the tropics, 

 produce little but grasses and sedges, and only few and incon- 

 spicuous flowering plants. Mr. Darwin mentions the brilliancy 

 of the flowers adorning the plains of Monte Video, which, 

 with the luxuriant thistles of the Pampas, seems hardly 

 equalled in the campos of tropical Brazil, where, with some 

 exceptions, the earth is brown and sterile. The countless 

 beautiful geraniums and heaths of the Cape cease on entering 

 the tropics, and we have no account of any plants equally 

 striking and brilliant supplying their place. 



What we may fairly allow of tropical vegetation is, that there 

 is a much greater number of species, and a greater variety of 

 forms, than in the temperate zones. Among this great variety 

 occur, as we might reasonably expect, the most striking and 

 brilliant flowers, and the most remarkable forms of stem and 

 foliage. But there is no evidence to show that the proportion 

 of species bearing brightly coloured, compared to those bearing 

 inconspicuous flowers, is any greater in the tropics than in the 

 temperate regions ; and with regard to individuals which is, 

 after all, what produces the effects of vegetation it seems 

 probable that there is a greater mass of brilliant colouring and 

 picturesque beauty, produced by plants in the temperate, than 

 in the tropical regions. 



There are several reasons which lead us to this conclusion. 

 In the tropics, a greater proportion of the surface is covered 

 either with dense forests or with barren deserts, neither of 

 which can exhibit many flowers. Social plants are less common 



21 



