166 MEMOIRS FROM THE DEPARTMENT OF BOTANY OF COLUMBIA UNIVERSITY 



The species of the genus Fragaria are very closely related to each other, and it is very 

 often difficult to distinguish them from each other. The number that has generally been 

 recognized for the United States is four, and for Mexico one, but the number is, how- 

 ever, Avithout doubt much larger. Until lately only two species were supposed to 

 grow in the Atlantic States, and four in the Pacific States. The so-called vesca and Vir- 

 giniana of the two regions are so unlike that it is impossible that they can be the same 

 species, and the true F. vesca L. is not a native of America. The Fragarias of the Rocky 

 Mountain region have been in such confusion that the names they bear on herbarium 

 labels are hardly ever correct. The trouble has been that altogether too small a num- 

 ber of species has been recognized. The only reasonable treatment of the genus is to 

 acknowledge about 20 species in North America, for should a smaller number be rec- 

 ognized the question would arise as to how to dispose of the numerous intermediate forms. 



I have endeavored to arrange the species in natural groups in somewhat the same way 

 as the arrangement adopted in PotentiUa and Horkelia, but all these attempts were in vain 

 on account of the close relationship of the members of the proposed groups. The general 

 arrangement that is found in manuals and monographs consists in dividing the genus 

 into two series, one with superficial achenes, the other with the achenes set into pits. 

 This will not do for the North American plants, for in some species, as for instance in 

 F. Chiloensis and F. Californica, the pits are very shallow. The latter, although with evi- 

 dent pits, is a rather close relative of the European F. vesca and F. collina, both of which 

 have superficial achenes. 



The pubescence of the scape and petioles has been used to distinguish related species, 

 even between groups. In some species the pubescence on these parts is appressed or 

 shghtly spreading, in others it is spreading nearly at right angles to the stalk or even 

 somewhat refiexed. Although this is not a reliable character in all cases, it has been 

 employed by me in this work in the key to the species, as it is in most cases a very 

 convenient one ; it must, however, be remembered that the key is artificial, and too much 

 stress should not be laid upon it. The general description will give a better and more 

 reliable definition than the key affords. 



Fragaria Chiloensis, F. grandiflom, and F. Carolinensis^ constitute a very natural group 

 with very large flowers and thick and rather coriaceous leaves which are strongly veined 

 and are more or less puberulent under the long silky hairs on the lower surface. F. cuneifo- 

 Ha Nutt., however, agrees perfectly with this group, except in being of a much smaller size 

 and having very small flowers. F. crinita {F. Chiloensis var. Scoideri Wats., in part) has 



' This is not, as may be supposed, a native of Carolina. Its origin is entirely obscure. It may be a hybrid between 

 either of the other two species and F. Virginiana, or simply a garden form perhaps derived from F. Chiloensis. 



