Cochlcaria. CULX'IFKK.E. 140 



♦ * * * Flowers large, both the petals and sepals ileeji joirplo: poii slender but eonic- 

 what compressed ; cotyledons accumbent. 



B. pygmseum, Gav. Dwarf biennial with simple stem and very numerous crowded linear 

 or narrowly lauce-liuear entire or few-toolliod leaves, jiulx'scent with a]>prcHsed 2-i>ointed 

 white hairs: inHorcsceuce very ileuse : sojials oblong stronjjly sjiccjite at the baso ; margins 

 thin, white ; petals 5 to 9 lints long : pr(ii<i'ls ;i.Hionding, 2 to 4 lines in lengtii : pods pube>«- 

 cent, scarcely tapering at all at the ap(;x ; stigma slightly 2-ltil>cil. — (jay, 1. c. 4. Cheiran- 

 thus pi/qnuKus, Adams, Mem. Soc. Nat. Mosc. v. 114. V. I'ulluini, I'ursh, Fl. ii. 43C. JIta/teria 

 pi/ymwa, Hook. Fl. Bor.-Am. i. 60, t. 19. //. miuiTiui, Torr. & Uray, Fl. i. 90. II. Pallasii, 

 ToTT. & Gray, 1. c. 667. II. Ilookcri, Ledeb. Fl. Ross. i. 174. Sisi/mlirium j>i/f/iiiin(m, Trautv. 

 Act. llort. Petrop. i. 60. — Arctic America from Greenland to Alaska. (Siberia.) 



33. COCHLEARIA, Touru. (Latin cochlear, .si)oon, from the form of the 

 leaves.) — Glabrous succulent herbs of Northern Hemisphere, chieHy boreal and 

 arctic, often maritime. — Inst. 215, t. 101; L. Gen. no. .028: DC Syst. ii. 3.08, 

 & Prodr. i. 172; Reichenb. Ic. Fl. Germ. ii. t. 17. — Considering its moderate 

 size, one of the most confused and difficult genera of the order. Until all mem- 

 bers of the group can be subjected to a thorough revision, the certain identifica- 

 tion of our American forms with the closely related European species is impossible. 

 However, as the occurrence of the European O. Anglica, C. Danica, and C. offi- 

 cinalis in Arctic America rests upon the high authority of Sir William Hooker, 

 Torrey & Gray, and others, it seems best to continue to (.'numerate them, although 

 no specimens from this continent in American herbaria can be cited as exactly 

 representing the typical gerontogeous forms. The remaining species are here 

 interpreted nearly as in Lange's careful treatment of the genus in his Conspect. 

 Fl. Groen. [By B. L. Robinson.] 



C. Anglica, L. Radical leaves long-petioled, ovate or sub-orbicular, rounded at the base or 

 slightly and broadly cordate, subentire ; lower cauline leaves similar, short-petioled; the mid- 

 dle and upper ovate-oblong, sparingly and bluntly toothed, sessile by a more or less auricu- 

 late biise : capsule subglobose, reticulated with prominent veins. — Syst. Nat. ed. 10, 1128, 

 & Spec. ed. 2, ii. 903 ; DC. Syst. ii. 364 ; Hook. Fl. Bor.-Am. i. 57 ; Torr. & Gray, Fl. i. 109. 

 — Antico.sti, J/rtcoim ; coast of Labrador to the Arctic Ocean and Alaska. A species dubi- 

 ously di.stinguished from the following by its reticulated capsule. 



C. oflBcinalis, L. Very similar to the preceding in habit and technical characters: leaves 

 somewhat more inclined to be lobed, and the radical more deeply cordate : pods globose, 

 smooth or obsoletely reticulate-veine<i. — Spec. ii. 647; DC. 1. c. ; Hook. 1. c— Shores of 

 the Arctic Ocean, ace. to Hooker and others. Specimens with the large flowers of the 

 European form have not been seen from America by the writer. Onr commonest Cocfitearin, 

 however, extending from Vancouver Island, Macoun, to Abiska, corresponds in all essential 

 points with this species save in its smaller flowers and often more stunted growth, differences 

 a.'scribable perhaps to climatic influences. 



C. tridactylites, Banks. Cauline leaves coarsely sub-trilobed with a single obtuse tooth 

 on e.ach side : silicels ovate-globose, as large as in the la.st ; style short, capitate ; seeds 2 to 

 4 in each cell. — Banks in DC. 1. c. 367 ; Hook. 1. c. — Labrador, lianlcs. No specimens 

 acce.^sible to the writer exactly represent this imperfectly described species (here charac- 

 terized from the original description). Enough are at hanfl, however, to show much varia- 

 tion in the toothing of tlie leaves without change of more essential characters, thus cjisting 

 much doubt upon the distinctness of a species separated npon this feature alone. Lange 

 may be right in referring the plant doubtfully to C. (innilandim, or it may be a form of 

 C. Anglica. 



C. Ddnica, L. Leaves smaller than in the first two species, only 1 J to 2 or 3 lines in di- 

 anuter, " all pctiolate," deltoid, and hastately toothed at the kise : ca])8ule ov.ate to ellip 

 soidal, nearly or rpiitc a.'' lonij a? the pedicel. — Sj" • " 'I D.m. t. 100; Eng. Px'i 



10 



