WESTERN TROPICAL AMERICA. 127 
ASCLEPIADACER. 
508. SarcostemMA cumanense, Humb. et Kunth.—Desne, in DC. Prod. v. 8. p- 539.— 
Columbia. 
509. Sarcostemma bilobum, Hook. et Arn. Bot. Beech. p- 438.—This is not taken up by 
Decaisne, but comes very near to his S. cynanchoides (in DC. Prod. vy. 8. p. 540), if it be not the 
same.— Acapulco. 
510. Ascuxpras curassavica, Linn,—Desne. in DC. Prod. v. 8. p- 566.—A. incarnata, Hook. et 
Arn. Bot. Beech. p. 438. non Linn.—Mexico to Guayaquil. 
511. Ascuepras longicornu, Benth.—Desne. in DC. Prod. v. 8. p- 570.—Realejo. 
512. BuePHaRopon mucronatum, Desne. in DC. Prod. v. 8. p. 603.—Guayaquil.—I can find 
no difference between these specimens and those from Vera Cruz gathered by Galeotti and Linden 
and described by Decaisne, 
GENTIANACER. 
513, GYRANDRA speciosa, sp. n., (Plate XLYV.), caule diffuso vel adscendente ramosissimo 
acute tetragono, calycis profunde 5-fidi laciniis tubum corolla aquantibus superantibusve, corolle 
lobis acutiusculis, antheris filamento gracili sublongioribus.—Zrythrea macrantha, (3 major, Hook, 
et Arn. Bot. Beech. p. 438.—Tepic. 
Planta glaberrima, ut videtur perennis, semipedalis usque ad sesquipedalis, habitu Chironse. Rami 
alterni, graciles, angulis fere alatis, apice laxe et dichotome 1-5-flori. Folia 4-8 lin. longa, raro ensssan _ 
acutissima, margine scabriuscula, basi non angustata, superiora parvasetacea. Internodia — — 
superiora fere pollicaria. Pedicelli ultimi 6-10 longi. Calyx 3 lin. longus, lacintis — angustis — 
vix ad quartam partem connatis. Corolle (roses ?) primum florentis tubus laciniis = multo ee in 
corolla emarcida persistente elongatur et lacinias calycinas tandem equat vel superat, sigitacs rotatus, ee aaa 
lobis ovato-lanceolatis acutiusculis fere 7 lin, longis, 24-3 lin. latis. Anthere Intew, loculis absque connectivo 
connatis, demum gyris 4-5 torte. Stylus stamina superans ; stigma late bilamellatum. 
I have had some doubts whether this beautiful plant might not be either 
identical with, or a slight variety of, the G. chironioides described by Grisebach, 
(DC. Prod. v. 9, p. 44), but that accurate botanist states expressly that the stem 
is nearly terete, and the anthers shorter than the filaments. The calyx of rs 
plant is usually not divided to the base, and the lobes of the corolla more eae 
than described by Grisebach. The variety a latifolia of Hooker ane =a 
Erythrea macrantha is a different plant. In the flower I examined the a a 
were much injured, but they were evidently much shorter and Jess twisted, os 
appeared to have the cells less closely combined. At the same time the very sho 
