290 EDMUND W. SINNOTT AND IRVING W. BAILEY 



In fact, the builders of all the various hypothetical land bridges 

 have used an abundance of phytogeographical data. 



Botanical evidence which has heretofore been of use to the 

 geologist, however, has been mainly derived from a study of flora 

 rather than of vegetation; a study of the species, genera, and 

 families which inhabited a region in past or present time and not 

 of the various plant types which they represent. It is now pos- 

 sible, however, to supplement this evidence by an investigation of 

 the habit of growth of the various elements in a flora, with especial 

 reference to the relative proportions of herbs and woody plants, 

 particularly among dicotyledonous angiosperms.^ Data thus 

 obtained derive their importance from the following facts which 

 recent investigation seems pretty thoroughly to have established: 

 (i) The earliest angiosperms were woody plants. (2) Herbs 

 attained but very little prominence until the beginning of the 

 Tertiary, since which time they have increased very greatly in 

 number. (3) The gradual refrigeration of the climate of the 

 temperate zones during the Tertiary, with the appearance of a 

 well-marked winter season, seems to have been the factor respon- 

 sible for the development of most herbs, which are plants well able 

 to withstand cold winters, in the form of seeds or underground 

 roots and stems. (4) This herbaceous vegetation, which reached 

 its greatest development in the land mass of the north temperate 

 zone, was thus composed of plants which were very hardy, agres- 

 sive, and rapidly dispersed, and it consequently spread far and 

 wide into other and warmer portions of the globe. 



The evidence on which these conclusions are based may be sum- 

 marized briefly as follows: 



From paleobotany. — The vast majority of angiospermous re- 

 mains, especially from the Cretaceous, are of plants whose nearest 

 modern relatives are always trees or shrubs. In the Tertiary, espe- 

 cially the middle and latter parts of the period, remains of her- 

 baceous plants are much more frequent. This geological evidence, 



'The monocotyledons were apparently derived in very ancient times from the 

 primitive dicotyledonous stock in adaptation to an aquatic habitat. They are almost 

 entirely herbaceous, and their few woody forms are evidently recent rather than 

 primitive. 



