296 EDMUND W. SINNOTT AND IRVING W. BAILEY 



even to species, which are identical with those in northern latitudes, 

 and are therefore in all probability recent immigrants. 



The slight degree of endemism among these southern herbs is the 

 more significant when we remember that such plants, from the brev- 

 ity of their life-cycle, are likely to change much more rapidly than 

 woody forms and hence would develop into endemic types in a 

 shorter time. The notable uniformity of flora, too, over most of the 

 globe during the Cretaceous, as shown by fossil evidence, renders it 

 very improbable that, if herbs had been a dominant feature of the 

 vegetation of that period, they would now be so uniformly absent 

 from the ancient portion of the southern floras. The union of 

 southern Africa with Eurasia in the Miocene and the union of South 

 America with North America in the Pliocene evidently mark the 

 periods of the herbaceous invasion of these two continents. Aus- 

 tralia has apparently been sufficiently isolated from Asia since 

 the Cretaceous so that land animals have been unable to enter it. 

 The northern members of its flora, however, have doubtless effected 

 their entrance by a migration, in comparatively recent times, over 

 the Himalayas and along the East Indies. Had herbs been numer- 

 ous in the latter part of the Mesozoic, when the connection of 

 Australia with Eurasia was evidently much more intimate, the 

 present vegetation of the island continent would certainly contain 

 a much higher percentage of such plants. That the very early 

 Tertiary flora of Australia included few if any herbs is made even 

 more certain by the fact that although in all probability northern 

 Australia and New Zealand were connected at that period by 

 a land bridge, none of the northern herbs which entered 

 Australia from the East Indies are now found in New Zealand; 

 and we are obliged to infer that they had not then arrived in 

 Australia. 



The period at which herbs became an important feature of the 

 north temperate vegetation and began to spread thence southward 

 is therefore pretty clearly set at somewhere in the early Tertiary.^ 

 That herbaceous dicotyledons existed in the latter part of the 

 Mesozoic, however, especially in mountainous regions, is most 



^ Recently discovered evidences of glaciation just at the close of the Cretaceous 

 are of importance in this connection. 



