-182 AN ACCOUNT OF THE GENUS ARGEMONE. 



Argemone armeniaca Lmn. — is at present treated as the type of the 

 very distinct § Miltantha of Pajmver, quite as distinct from any of 

 the other sections as half the proposed genera of PapaveracecB are 

 from each other. When its subordination from utihtarian motives 

 to one of the larger allied genera is decided upon, it becomes a very 

 open question whether it ought to find a place in Tournefort's 

 Papaver or in Viguier's Meconopsis : its stigmas are arranged and 

 its capsule dehisces exactly as in the latter genus ; the only 

 character that separates it from Mcconopds and justifies its asso- 

 ciation with Papaver is the absence of a distinct style. But though 

 its characters go far to justify Linnaeus in removing this plant 

 from Papaver, they do not in any way support his location of it in 

 Argemone. The result, however, of the treatment in the Speries 

 Plantarnm is that the Linnean genus Argemone can only by courtesy 

 be quoted as synonymous with that of Tournefort. Whereas the 

 Tournefortian genus, by completely excluding the Argemone of 

 Bauhin, and thus of necessity also the classical Argemone, remained 

 as a result of Tournefort's definition and limitation the apparently 

 natural genus that we still accept, the Linnean Argemone is a mere 

 arbitrary conglomeration of membra diajecta without even possessing 

 the excuse of attempting to conserve the classical incidence of the 

 name. The only possible explanation of this treatment is that 

 Linnfeus had not seen, when the Species Plantarum was issued, 

 either his A. pgrenaica or his A. armeniaca, and it is interesting to 

 find that he probably never saw them, for neither is represented in 

 his own herbarium. In 1753 Haller published an Argemone which 

 is evidently a Papaver, and is probably a form of P. nudicaule 

 (alpinum)* But no other author has added a species to the genus 

 in the Linnean sense, and the confusion introduced by Linnaeus 

 did not long persist, for in 1784 Lamarck again restricted it witliin 

 the Tournefortian limits ; t with perhaps the single exception of 

 Lestiboudois,! he has in this been followed by all subsequent 

 authors of any importance. 



The place usually assigned to the genus Argemone is alongside 

 of Meconopsis, Papaver, and the allied genera that constitute the 

 EiipapaveracecB. This is not an altogether convenient arrangement, 

 because it places among genera in which the flowers are usually of 

 2-merous type one in which the floral arrangement is normally 

 3-merous. It is quite impossible, however, to find any character 

 in this troublesome order that does not at times break down, and in 

 the present instance the two most distinctive characters, a constant 

 or almost constant 3-mery and the presence of horns under the 

 apex of the sepals, both fail us within the limits of the group of 

 species that we are accustomed to treat as forming the genus 

 Papaver. Not only is the presence of 3-merous flowers an occasional 

 feature in wild examples of § Scapijiora (P. nudicaule), and in a 

 species [P. lateritium) that forms a connecting link between 

 § Scapijiora and § Calomecon, but it is a normal character in both 



* Haller, Plant. Goetting. 89. t Lamarck, Encyc. Meth. i. 247. 



I Lestiboudois, Bot. Belg. iii. pt. 2, 132 (1799). 



