566 



PAPAVERACEAE 



the summer period may not flower at all the next spring, or the next year, just as vernal 

 flowering plants of a given year may not flower the spring of the next year. Plants grown 

 on the coast at Berkeley from seed of the E. crocea form do not, in such a situation, go through 

 the successive life-history phases wliich characterize that fonn in the interior valleys, but tend 

 to simulate closely the coastal form. Transplantation of perennial Eschscholtzias is difficult, 

 but we have made in early winter successful transplants of the E. croc«a form from Vaca 

 Valley in the interior to Berkeley on the coast. These individual plants on flowering the next 

 spring resembled, not their Vaca Valley associates, nor indeed their own previous flowering, 

 but the smaller flowered and often somewhat scapose plants of tlie Berkeley Hills! 



The Eschscholtzia crocea fonn furnishes a characteristic illustration of variation in the 

 perennial group. In this form extreme types of habit and of flowers are produced on one 

 root-crown during one season 's reproductive period. Specimens from one locality representing 

 these dift'erent phases, have been described by two authors as species, and specimens of the 

 different phases from various localities have also been described by the same authors as 

 species. Such proposals obviously cannot stand as species. Our work upon the E. crocea form 

 alone is sufficient to destroy the value of the characters used as differentiae in the diagnoses 

 of the numerous segregates of E. californica made by Greene and by Fedde, for the reason 

 that the normal range of variation in the successive flowering stages of this one form covers 

 all the characters used by these authors in describing their segregates. 



Fig. 122. Eschscholtzia c.vlifornica Qiam. var. crocea Jepson (E. crocea Bentli.) ; habit; 

 summer-flowering (July) stage. These stems arise from the same root-crown that produced 

 the erect large-flowered shoots shown in Fig. 121. A basal portion of one of the heavy erect 

 flowering stems of spring, now dead and dry, is shown at h. X Vi- 



Before going further it should be emphasized that the amount of variation in Eschscholtzia 

 californica is not by any means as great as would naturally be supposed from the large number 

 of specific segregates which have been published. Single sheets of EscJischoltzia specimens 

 have been taken in the main by the segregators as sui speciei. If this method were generally 

 employed it might be applied with similar results to hundreds of California species which have 

 never been subject to segregation and which do not as yet possess a single synonym. We now 

 proceed to a detailed analysis of the synonymy, although we regard the extent of our treat- 

 ment as out of proportion to its intrinsic importance but not to its mass. 



In Greene's Revision of Eschscholtzia (Pitt. ,'5:205-293) appear descriptions of 39 perennial 

 species and 5 varieties, of which 32 species and all the varieties are new, -i of the 7 remaining 

 species having been fully described elsewhere by the same author. The annual species number 

 73 (or 1 a possible perennial) with 3 varieties, .30 of the species and the 3 varieties being new, 

 11 of the remaining 17 species having been previously described by Greene. Fedde in his 

 account of Eschscholtzia (Engler, Pflzr. 4"':144r-202) follows Greene and accepts his species, 

 with the addition, however, of several new species and varieties, reducing only one of Greene's 

 species to varietal rank. 



