Mr. J. Miers on the Bignoniacege. 257 



liis T. albiflorum (Fl. Ind. Occid. tab. 20) and of T. crucigerum 

 (Plum. Am. tab. 254), on the one hand, with T. parasiticum 

 (Sw. icon, cit.) and with T, [Schlegelia) lilacinum, Miq. (Aubl. 

 Guian. tab. 254), on the other, no one can doubt that the two 

 former species are generically distinct from the two latter. In 

 the former group the plants are scandent, their leaves conjugate, 

 with a long cirrhus, as in Bignonia ; the calyx is green, long, 

 and tubular ; the corolla is white, pubescent within and without, 

 with a very narrow hypocrateriform tube, of unusual length (6 or 

 7 inches), with an undulately crispate 5-lobed border ; the sta- 

 mens and style (of great length) are exserted ; the anther-lobes 

 are linear, widely divaricated, with a terminal excurrent connec- 

 tive ; the fruit is very large, oblong, often a foot in length; and 

 the seeds are large, broad, compressed, and not imbedded in pulp. 

 In the latter group the stem is radicant ; the leaves are quite 

 simple, as in many of the Catalpece ; the calyx is coloured, short, 

 and globosely campanulate ; the corolla is deep violet or purple, 

 quite glabrous, scarcely more than | inch long, much swollen 

 and ventricose above a short basal constriction, with an oblique 

 bilabiate border, the upper lip of which is erect, bifid, scarcely 

 cleft to the base, and the lower lip is tritid, reflected, with the 

 middle lobe considerably tlie largest, and enveloping all the 

 others in aestivation; stamens and style only half the length of 

 the short corolla, and of course included ; anthers very small, 

 ovate, white, with nearly parallel lobes ; fruit globose, only | inch 

 diameter in one species, and not more than ^ inch in the other, 

 Avith projecting seminiferous placenta, rendering it falsely 2- 

 locular, as in Kigelia, and containing numerous minute, angular, 

 oblong seeds enveloped in pulp. These characters are severally 

 as opposite as possible, rendering it evident that Schlegelia is not 

 only generically distinct from Tanaecium, but appertains to a 

 different family. The former genus manifestly belongs to Crcs- 

 centiacece, while Tanaecium will probably find its place near Ade~ 

 7iocah/iimia in Bignoniea, because it possesses a similar habit, has 

 the same kind of cylindrical elongated fruit, as we have seen 

 (p. 167), and its seeds are, in like manner, large, apterous, and 

 closely packed together, without intervening pulp. 



The remaining genus, Parmentiera, placed by DeCandolle near 

 Tanaecium on account of its bilocular ovary and indehiscent 

 fruit, is referred to his tribe Crescentiete by Dr. Seemann, who, 

 in detailing its generic character, affirms that the ovary is at 

 first unilocular, but that by the enlargement of the placentae it 

 becomes 2-4-locular in the fruit * ; but he nowhere states that 

 he had examined the ovary or had witnessed the organization 

 just mentioned, and we may infer that he copied this character 

 * Hook. Kcw Journ. Bot. ix. 82. 



