AN ACCOUNT OF THE GENUS ARGEMONE. 369 



throughout Central and Northern Mexico, and in Lower California, 

 extending also throughout Arizona and Southern California. This 

 is tlie A. munita of Greene, but not of Diirand and Hilgard. 



y. That with purple flowers, from Southern Texas, which other- 

 wise does not seem to differ from form /3. I have not conserved 

 Mr. Coulter's name of A. platycenis var. rosea for this form, partly 

 because the name is made to include a form of what must be treated 

 as another species, partly because the mere difference of colour does 

 not satisfactorily separate the plant. The name should moreover, 

 if possible, be dropped, owing to the possibility of bibliographic 

 confusion between this plant and A. rosea Hook. 



^. Tliat with extremely prickly capsules, the spines of the middle 

 line of the valves being sometimes upwards of 25 ram. long, and 

 much branched. This form occurs in Texas, where it was collected 

 by Trecul near New Braunfels, and by Drummond ; it forms part 

 of the third communication by this unfortunate collector to Sir 

 William Hooker, and therefore, though the specimens have no 

 field-ticket, we know from the historical account in the Bot. Misc. 

 that the specimens were collected somewhere to the north-west of 

 Austin, between the Brazos and Colorado rivers. In Herb. Kew 

 Mr. Planchon has proposed to consider this a distinct species, and, 

 had the plant been European, I should have felt little hesitation in 

 suggesting its distinction as A. Planchonii ; as it is, the matter must 

 be left to American botanists to decide, and I leave the form for the 

 present in A. j)l(ityceras, of which it has the foliage and the flowers. 

 It has escaped notice that this particular plant has been mixed with 

 A. Iiispida from the beginning, although it is not hispid, for Gray's 

 original description covers the characters of fruit met with in this 

 plant. In Herb. Kew, Herb. Paris, Herb. Drake, Herb. Cosson, 

 Herb. Durand, Herb. DeCandolle, there are only flowering examples 

 of Fendler n. 16, on which A. hispida is based. All these are hispid 

 specimens, and all belong to the plant described here as var. hispida. 

 The only fruiting example of Fendler n. 16 that I have seen is in 

 Herb. Brit. Mus., and it is not the same as the flowering plant; it 

 is this densely aculeate (but not hispid) plant with branching spines 

 on the capsules. 



In Herb. Durand there are good New Mexican examples of the 

 true A. hispida in fruit. These show that that plant, at least 

 normally, has fruits indistinguishable from those of the North 

 Mexican and Californian form of A. j^lattjceras, which are well 

 represented, in fig. 86 b of Prantl and Kundig's account of the order 

 Papaveracece, under the erroneous name of A. mexicana. The con- 

 fusion that has existed among these forms has been extreme, and 

 the only American authors who have appreciated the differences 

 between these forms have been Mr. Colville, whose synonym A. 

 plati/ceras is the only one by any North American botanist that 

 applies without qualification of any kind to Link & Otto's plant ; 

 and Mr. Greene, who has definitely separated A. platyceras from 

 A. hispida in his Flora Fravciscaiia. But Mr. Greene has given tlie 

 name A. munita of Durand to the plant common in Southern 

 California ; this is not admissible, because I believe that the 



JouKNAi, OF Botany. — Vol. 33. [Dec. 1895.] 2 b 



