40 Journal of the Mitchell Society l^^ciy 



"Thej belong to eight genera which are systematically scat- 

 tered, for they represent six families, Ranunclaceae , Fumar- 

 iacea, Umbelliferae, Primulacae, Lentihularieae [sic]^ Nyctag- 

 ineae. Clearly these species cannot have inherited the peculiar 

 form of their seedling from a common ancestor. It must be due 

 to similar external conditions affecting certain species of very 

 different descent in the same way. 



"One feature is common to all of the pseudo-monocotyledons 

 in my list — they all possess some underground member which 

 is thickened into a tuber. In Ranunculus Ficaria one of the 

 earlier cauline roots became tuberous ; in the other species the 

 hyi^ocotyl is more or less thickened. 



"Moreover, the most complete list I can make of dicotyledons 

 with their cotyledons partially united for some distance from 

 the base upwards included twenty genera. It contains but one 

 genus — Rhizophora — in which the hypocotyl is not very much 

 shortened, if not actually thickened. In the great majority the 

 hypocotyl becomes a conspicuous tuber. The seeds of the 

 single exception germinate under peculiar conditions, which 

 would account for most any amount of modification in the 

 structure of the seedling " 



And a little further on, she says : 



"The formation of assimilating organs in the seedling of a 

 geophilous plant is, however, very greatly limited by the short- 

 ness of the growing season and the necessary formation of sub- 

 terranean organs. Here lies the explanation we are seeking: 

 the reduction of the cotyledons and the formation of a tuber 

 are both adaptations to the geophilous habit " 



Is it not remarkable that these observations should apply so 

 exactly to Quercus virginiana, the oak which shows the strong- 

 est tendency^ to the geophilous habit ? 



In the Botanical Gazette, Vol. XXXVII, p. 62, I published 

 a drawing of an acorn Quercus Prinus L. from which three 

 healthy young plants were sprouting. Most of the acorns from 

 the same tree were also multiseeded. There is in Chapel Hill, 



^AU oaks show a slight tendency towards the geophilous habit in the con- 

 centration of the early growth in the root, and Englemann mentions the oc- 

 casional fusion of the cotyledons in Quercus pungens, a shrub of dry regions 

 in the West. 



