670 SUPPLEMENT.— POLYGALACEiE. 



Order POLYGALACE^. 



1. POLYGALA, 2^- 126. 



1. P. sanguinea. — This, the P. sanguinea of Nuttall, is not the plant of 

 Linnaeus. It is also the P. fastigiata, Nutt. ! according to specimens in 

 Hooker's herbarium, and therefore oi' Hook. S^'Arn.! in jour. hot. 1. p. 195; 

 but not the P. fastigiata which Mr. Nuttall communicated to us, which is 

 only a form of P. cruciata. It is not improbably the plant figured by Plu- 

 kenet {Mant. t. 438, /. 5.) As a new name seems to be necessary, it may 

 be called P. NuttaUii. 



2. P. purpurea. — This is the original P. sanguinea, Linn..' (fide sp. 

 Kalm in herh. Linn.) as well as of the earlier American authors. It is also 

 P. viridescens, Linn.! spec. 2. p. 705 (a young state), excl. syn. Gronov.! 

 which relates to P. In tea. 



3. P. cruciata, Linn. {pi. Gronov. !) — Linnaeus had never seen this plant, 

 but founded the species on P. foliis quafernis, Gronov.! Ji. Virg. (which 

 is just P. cuspidata. Hook. S^-Arn.!), and P. (juadrifolia seu cruciata, <5cc., 

 Pluk. aim. p. 301. Of this last we could find no specimen in Plukenet's 

 herbarium; but, as he states it to come from Virginia (Banister), it cannot be 

 the exclusively northern form (Massachusetts) which Hooker (in^. Bar.- Am. 

 1. p). 85.) takes for the true P. cruciata. The plant of the Southern States 

 is the original species ; but Hooker's plant cannot be separated, as we have 

 numerous intermediate forms in which the point of the viings is nearly 

 wanting. 



5. P. corymhosa. — Having ascertained that Walter's P. cymosa is identical 

 with this, his name must be adopted. 



6. P. acutifolia. — We are now convinced that this plant is only a larger 

 form of the preceding, with narrower leaves. It is also P. graminifolia, 

 Poir. ! (fide sp. in herb. Mus. Par.) 



7. P. cymosa. — As Walter's name is transferred to P. corymbosa, that of 

 Nuttall must l)e adopted for this species. — The synonymy of these species 

 will therefore stand as follows : 



P. cT/?ftosa, Walt. ! — P. corymbosa, Michx.! (in part,) Nutt.! DC. I. c. 

 P. ramosa. Ell. ! 

 0. graminifolia. — P. graminifolia, Poir.! DC! P. acutifolia, Torr. ^' Gr. 



P. attcnuata, Nutt. — P. corymbosa, Ell.! P. cymosa, Torr. Sf Gr. excl. 

 syn. Walt. S^- Poir. 



13. P. bicolor. — The plant described by Hooker & Arnott under this name 

 is P. Boykinii, Nutt., but not of Kunth, as we have ascertained by a careful 

 comparison of authentic specimens. The exterior sepals in P. Boykinii are 

 nearly equal, and destitute of the "2 thick parallel nerves" which are so con- 

 spicuous in P. bicolor; the wings also are broader, and the flowers do not 

 turn orange wlien decaying, as in the latter species. It is jnuch more nearly 

 allied to P. asperuloides, Kunth! from which it chiefly differs in the denser 

 spike, broader wings and white (not rose-colored) flowers. 



21. P. grandiflora. — This is either the same with P. violacea of Aublet, 

 or exceedingly near that species, as was obligingly pointed out to us by Mr. 

 Bennett, who showed us a Brazilian specimen in the Banksian herbarium. 



