170 COBB— RELATIONSHIPS OF WHITE OAKS [April 23, 



tiary, was, at the boreal end, continuous from Asia across America 

 to Europe, gives the possibility of the center of distribution being 

 either in Europe or in America. My data on European oaks are 

 insufficient to decide this point; it seems, however, highly prob- 

 able that the white oaks with thin, deciduous, lobed leaves originated 

 in or near northern Mexico. 



The early members of the group Leucohalanus, then, marked by 

 entire, evergreen leaves, gave rise, probably in North America, to 

 a form with thin, deciduous, lobed leaves. This type is now domi- 

 nant over the greater part of the United States, while the older form 

 holds its own in the southwest and in Mexico, where the climate has 

 probably known no great fluctuations since the Cretaceous, and 

 where it still finds suitable dry and arid habitats. This evergreen 

 type occupies the Mexican highlands, Arizona and New Mexico, 

 extending east into Arkansas, and west into California. Quercus 

 virginiana seems also to have been a very early offshoot ; with its 

 varieties it forms a well-marked coastal group, ranging from North 

 Carolina south along the shores of the Gulf into Mexico (where 

 it stretches inland up the mountain sides), and appearing also on 

 the California coast. 



III. Deciduous White Oaks of North America. 



The oaks with which we are familiar in this part of the country 

 are of the lobed-leaf type. Geographically, at least, there are three 

 parts to this group, — the eastern, the Rocky Mountain, and the Cali- 

 fornian lobed-leaf oaks. It is not clear, however, whether or not 

 these geographical groups can be separated taxonomically. They 

 may be parallel groups, cut off. from one another comparatively 

 recently; or, possibly, the Cahfornian group may be more closely re- 

 lated to the deciduous oaks of Europe (type Q. robur) than it is to 

 the oaks of the Rocky Mountains and the east. The habit, leaf 

 form and texture, and bud form of the Californian oaks have sug- 

 gestive resemblances to those of the English oak ; and it is perhaps 

 not venturing too much to speculate as to whether these oaks, like 

 certain other forms on our Pacific slope, may not have their closest 

 relatives, not in America at all, but in Europe. There is besides at 

 least one oak in CaHfornia, Q. sadleriana, which appears to find its 



