158 EOTAKY OF THE VOYAGE OE H.M.S. IIEKALD. 
longi ; tubulo brcvissimo vix semilinearij puberulo in faucem ampliatam sesquilineam longam, basi 
puberulam CBeterum glabram, nends 5 inter sinus dentiuni excurrentibus pcrductam, apicc brevitcr 
5-dcntatam transeuntCj dcntibus obtusiusculis extus hispidis. Filamcnta ima faucc inscrta, latiuscula, 
vix torta j aniherae sub antlicsi pauilisper exsertse, brunnesej basi ecaudataij supra conncctivnm in 
appendicem liberam, ovatamj basi attenuatam, acutlusculanij liyalino-fcrmgineam, uninervem pro- 
ductse. PolUnis granula spbserica, undique echinulata. Ovaria oblonga^ comprcssa, glabra, calva. 
Stylus basi in bulbum majusculum incrassatus, glaber; stigmatis rami exserti, utrinque recurvi, dense 
papillosi, appendicula longiuscula attenuate- acuta superati. Achania raatura oblonga, crassa, sub- 
tetragonaj compressiusculaj glabra, mai'ginc angustissimo, bine illinc paullulum clevato coronata, 
CBeterum calva. 
This species seems to be related nearest to Gijmnopsis? Costaricensis,'BQTiii\iQ.m (in * Compositaj Centro- 
Americance/ gathered by 6rsted, p. 2G, no. 114) which I have not seen, but it is said to have trinerved and 
broader leaves, with larger petioles. Moreover our plant seems to differ by its achaBuia being crowned by 
a very narrow but manifest margin. On the other hand it has the habit of a Wulffia, but is well distin- 
guished from this genus by the ach^enia not being fleshy, and by the triplinerved leaves. The genus 
Gymnopsis, Be Gaud., however, includes indeed incongruous plants, as remarks Asa Gray in a Flora of 
!N'oi*th America, torn. 2. p. 317 ; and surely I do not kuow why De CandoUe joined with it Aldmia of De la 
Llave and Lexarza, which is so strikingly characterized hy its achsenia entirely enclosed by, and cohering 
with the at length increased chaff, and which truly merits to be restored. 
591. AYuLFFiA platyglossa, De Cand.? Prodr. torn. 5. p. 563. no. 3.— Frutex 10-pedalis. Pa- 
nama, in savanas. 
I am not quite certain whether Dr. Seemann*s specimens, being in fruit only, without any trace of 
flower, belong to this species or to Wuljia stenoghssa, De Cand., which has much longer ligules. But 
ha^'iug compared many specimens of both species, I found the leaves of W. stenoghssa being in general more 
acutely serrate and of a firmer texture, than in W. platyglossa and Dr. Seemann's plant. Both however 
seem to be distinguished by very slight characters. 
592. Oyeo^a verbeshioides, De Cand. Prodr. torn. 5. p. 577. no. 1. — Delessert, Icones Select, 
torn. 4. tab. 34. Helianthus arboreus, WiUd. ! Herbar. no. 10,413. Volcano of Chiriqui, Veraguas. 
The specimen in WiUdenow's Herbarium is collected by Humboldt at Caracas, but seems not to bo 
described in the ]S"ov. Gen. et Sp. — It occurs abundantly in Venezuela, whence I have seen specimens 
gathered by IMoritz and others. 
593. BiDENs leucantJia, WiUd. ? De Cand. Prodi', torn. 5. p, 598. no. 26. Common about Panama. 
Dr. Seemann's specimens are all in fruit. The achfcnia are imequal in length, from 4 to 9 lines, some- 
what quadr-angular linear, some of them being densely hirsute, but the most part quite glabrous ; the lower 
leaves are long-petiole d, with a petiole of 1-2 inches, the terminal leaflet somewhat rhomboid-shaped is 
cuneate towards the base.^ — Abstracting from the flowering capitula, which are wanting in our specimens, 
they seem indeed to belong to the species which is considered by De Candolle as Bidcns leiicantJia, AViUd., 
whose short diagnosis in the Prodromus well agrees. The genuine species of Willdenow will remain for 
ever a problem, since in his own herbarium, no. 16,022, under this name are associated three incomplete 
specimens, belonging to three different species. 
■ 
594. BiDENs tereticaulis, De Cand. Prodr. torn. 5. p. 598. no. 31. Savanas about Panama. 
The flower-heads of Dr. Seemann's specimens, being not developed, do not admit of a 8iu*e definition. 
