20i TETRADYNAMIA— SILIQUOSA. Matthiola. 



pods, being perfectly distinct, as it seems to me, from the others, 

 nor is there any forked or starry pubescence to be found. Leaves 

 crowded, stalked, lanceolate, acute, almost invariably entire j 

 the lowermost, if any, slightly serrated ; all deep green, with 

 more or less of a minute silvery hoariness, especially at the back. 

 Fl. corymbose, sweet-scented ; their petals always of a uniform 

 bright golden yellow, not stained with brown or blood-red as in 

 the garden Ch. Cheiri, though the calyx is purplish. Pods race- 

 mose, erect, l^or 2 inches long, covered with close hairs chiefly, 

 if not altogether, pointing upwards ; each valve marked with an 

 elevated central line, often vanishing about half way up, and 

 hardly discernible at all in Mr. Davall's Swiss specimens ; though 

 very strong in some French ones, with shorter broader pods, 

 which most accord with Dr. Hooker's, the style excepted. Stijle, 

 in all the specimens I have seen, about a line in length in the 

 flower, rather more on the ripe pod, stout, angular, a little 

 bristly, crowned with the cloven stigma, whose lobes are finally 

 brought close together. The seeds are flat, with a narrow, mem- 

 branous, deciduous border at one side, as well as at the summit, 

 of each. 

 The late Mr. Crowe, whose remarks were always worthy of atten- 

 tion, and to whom we owe so much for his unrivalled discrimi- 

 nation of Willows, observed that the petals of our wild Cheiran- 

 thus merely become recurved as they advance towards decay, 

 and do not hang loosely flaccid, like those of the true Ch. Cheiri, 

 or Blood Wall-flower of the gardens. There is indeed a culti- 

 vated double variety of Ch.fruticulosus, always with plain yellow 

 Jiowers, and though more luxuriant than the wild plant, still 

 unlike the Ch. Cheiri. Dr. Hooker appears to me quite cor- 

 rect in his Fl. Scot., except a slip of the pen, leaves for petals ; 

 but I quote his Fl. Lond. and its luminous dissections, with he- 

 sitation, on account of the strongly-ribbed valves of the short 

 pods, and the almost total want of a style, such as I have never 

 seen in any Wall-flower. Ch. Cheiri and its supposed varieties 

 enumerated by DeCandolle, require more correct examination 

 than they have, as yet, received. I do not presume to give a 

 decisive opinion concerning them, but merely describe what I 

 have seen, depending with implicit confidence on my friend 

 Hooker for the accuracy of his representations. Viola lutea, 

 Fuchs. Hist. 458, f, comes nearest to his plate and description. 



338. MATTHIOLA. Stock. 



Br. in Ait. H. Kew. t\ 4. 1 1 9. DeCand. Syst. v. 2. 1 62. Comp. ed 4 

 108. ' ' 



Cal. converging, a little compressed ; leaves linear-oblong, 

 concave, erect, deciduous, 2 opposite ones protuberant at 

 the base. Pet, obovate, spreading, entire, or with a broad 



