18 INTRODUCTION. 



the Uinbelliferce, the corymb elongating into the 

 raceme of the Crucifene, the spikelets of the 

 Grasses are all high developments compared to the 

 variable inflorescence in Rubiaceae, Saxifragacea3, 

 Rosacea, and Liliaceae, where in most instances 

 can be traced their gradual development out of 

 solitary axillary flowers whose supporting leaves 

 represent bracts, frequently still showing the tran- 

 sitional stages. 



I must add here an observation in regard to the 

 geographical distribution of the higher evolutions. 

 Their proportion increases in the Northern Hemis- 

 phere, while the greater number of ancestral types 

 have been preserved in the Tropics, and in the 

 Southern Hemisphere, where arborescent ferns, 

 and conifers of otherwise fossil t} r pe, are still found. 

 The preservation of such forms can easily be ac- 

 counted for by the equable warmth and moisture 

 of a tropical climate, but it is more difficult to ex- 

 plain the existence of these types in the Temperate 

 Zone of the Southern Hemisphere. It may be that 

 they owe their preservation under less propitious 

 skies, to the isolation of the continents, and per- 

 haps also to the circumstance that agriculture and 

 civilization entered that part of the globe at a com- 

 paratively late period, so that the decimation of 

 ancestral forms by the struggle for existence did 

 not begin so soon as in the Northern Hemisphere 

 with its many centres of civilization. 



The arrangement of the genera in this volume is 

 not, I am aware, in strict accordance with the 



