207 



AN ACCOUNT OF THE GENUS ARGEMONE. 

 By D. Prain. 



(Continued from p. 178.) 



The difficulty occasioned by the union of these dissimilar forms 

 has been got over by the statement that A. mexicana is highly 

 variable. This statement is, however, purely hypothetical, and, as 

 applied to the original A. mexicana of Tournefort and Linnajus, 

 quite incorrect. This is shown by an examination of the numerous 

 specimens of the plant reported from various parts of the Old World 

 where this form exists as an introduced plant, and where no other 

 form of Arrfemone occurs. In India, for example, where the species 

 extends from the Punjab and Kamaon to Ceylon, and from Malabar 

 to Bengal, growing everywhere from sea-level to an elevation of 

 5000 ft. in the Himalayas and the Nilghiris, and flourishing equally 

 well in the humid atmosphere of the Gangetic Delta and on the 

 dry table-land of the Deccan, there is probably no species indigenous 

 or introduced that accommodates itself so readily to altered con- 

 ditions and yet remains so absolutely true to its essential characters 

 as does A. mexkaim. An examination of the specimens from Africa, 

 where it occurs from the Cape to Algeria, from Socotra to Senegal, 

 shows that in this continent also the same is absolutely true. 

 Seeing that this is so, we are compelled, unless we are prepared to 

 forego any attempt at classification whatsoever, to separate, once for 

 all, the white-flowered Argemones from the yellow-flowered ones. 

 But within the yellow-flowered group itself we find quite sufficient 

 difficulty, for though the statement that A. mexicana is variable 

 outside America is incorrect, we do find in the New World specimens 

 with yellow flowers that seem at first to bear out the general con- 

 tention as to the variability of A. mexicana. Here again, however, 

 a more careful examination shows that the assumption is hypo- 

 thetical, and is due not to any variability in the flowers or fruit of 

 A. mexicana, but to a want of care in separating from it a quite 

 distinct form as constant apparently in essential characters as 

 itself. There is no doubt that what has been the cause of this 

 misapprehension is the fact that this second plant is really a 

 Mexican one, while that which bears the name A. ynexicana is in 

 Mexico only an introduced one, occurring occasionally on the 

 eastern coast in the vicinity of seaport towns. Though the 

 differentiation of these two is always easy, and though no inter- 

 mediate forms are as yet reported, I have, following Lindley, 

 treated them here as varieties of one species, which thus includes 

 all the yellow-flowered forms except A. fruticosa ; even if Sweet's 

 differentiation of A. ochroleuca be ultimately sustained, there is no 

 doubt that the two constitute a natural group of forms. 



The question which next arises is whether the whole of the 

 white-flowered forms can be treated as one species. If this were 

 possible, we should then have but three species in the genus, viz., 

 A . fruticosa, A. mexicana, and A. alha. But a closer examination of 

 these white-flowered Argemones shows that this treatmeut is 



