TKOPICAL FLORAS T5 



quickly onwards and do not appear to kill or injure the trees 

 or even the small herbaceous plants. In fact, numbers of 

 these plants as soon as the rains come produce foliage earlier 

 than Avhere there has been no fire, and often produce flowers 

 when unburnt trees or shrubs of the same species remain 

 flowerless. Mr. Warming and otlier botanists believe that the 

 practice of firing the campos was a native one long before the 

 European occupation, and that many of the plants have become 

 adapted to this annual burning so as to benefit by it. 



It is interesting to note here the opinions of two eminent 

 botanists, only thirty years ago, as to the comparative riches 

 of certain tropical and temperate countries. In his great work 

 on The Vegetation of the Globe, Griesbach thus refers to the 

 Brazilian flora : ^^ The results of the explorations of Martins, 

 Burchell, and Gardiner, cannot be compared with those fur- 

 nished by the Cape. The number of endemic species may per- 

 haps reach 10,000, but the area is twenty times greater than 

 that of Cape Colony, and we may conclude that, as regards its 

 botanical riches, the Brazilian flora is very far from rivalling 

 that of the extremity of South Africa." Gardiner, however, 

 after spending three years in collecting over a large portion 

 of the interior of Brazil, though chiefly in the campos and 

 mountain ranges, concludes his account of his travels with these 

 w^ords : " The countrv is beautiful, and richer than anv other 

 in the world in plants." This general statement may not be 

 strictly true, but it seems clear that the facts already adduced 

 are sufficient to show that, as regards the comparison of tem- 

 perate with tropical floras, there can be no doubt as to the 

 superiority of the latter. This point will, I think, l)e made 

 still clearer in the following discussion of some almost unno- 

 ticed facts. In the case of Brazil and Cape Colony, however, 

 it is clear that Griesbach was creatlv in error. Tlie wliole 

 area of extra-tropical South Africa has probably been as well 

 explored botanically as Brazil, the richest portions of which 

 have been only as it were sampled. Yet wo find less than 

 14,000 species in the former against 22,S00 in the latter. It 



