104 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
fish-like, neatly ciliate, 18 to 15 mm. long, 8 to 4 mim, broad, including ciliate 
margin. 
Hacienda Belnira, near Santa Maria de Dota, Costa Rica, altitude 1,450 
mneters, Tonduz, January, 1898, flowers and fruit (Instituto fisico-geografico de 
Costa Rica, ne. 11619) ; Angostura, near Turrialba, alti- 
tude about TOO meters, Cook & Doyle, no, 35, April 11, 
19038 (U.S. National Herbarium, no. 577471, type). 
Tonduzia stenophylla (Donnell Smith) Pittier. 
PLATE TX, 
Reaarolfia stenophylla Donnell Smith, Bot. Gaz. 44: 
115. 1907, 
Glabrous: upper internodes 8 to 20 mm, long; leaves 
elliptic-lanceolate, attenuate, chartaceous, varying in 
size in the same whorl, ) to 17 cm. long, 1 to o cm. 
broad; secondary nerves 30 to 85 pairs, arcuate near 
the margin; veins indistinct; petioles canaliculate, T to 
18) mm. long: cymes pseudoterminal, coryinbiform, 
reaching about one-third of the length of the nearest 
leaves, dichotomous and many-flowered; pedicels 2 to 
G6 mm. long: bracteoles 0.5 mm. Jong; calyx with 
Fig. 6.—Fruit of Tonduzia — younded-triangular tips, 1.5 mm. long; tube of corolla 
parvifolia, One-half nat- 
ural size 
puberulent outside, hirsute inside, about 2 mm, long, 
lobes of equal length or slightly shorter; stamens in- 
serted on lower half of tube, glabrous: carpels separate, style cleft for about 
the first third of its lower length; follicles cylindrical, apiculate, smooth or very 
finely striate longitudinally, 7 to 13 cm, long; seeds flattened, obovate-elliptic, 
delicately ciliate. . 
Around San Salvador, Salvador, altitude SOO to 1.000 meters, Carlos Tkenson, 
no, 289, 1905, flowers; no, 335, 1906, fruit (both U.S. National Herbarium). 
EXPLANATION OF PLATE IX.—Leafy branch and fruit, one-half natural size; pistil, seed, 
bud, and stamen variously magnified. 
Specimens with flowers were first received and, not being able to identify 
them myself satisfactorily, I had them forwarded to Captain Donnell Smith, 
who, misled by the very similir floral structure, described them as a new 
species of the genus Rauwolfia. Meanwhile I had been studying no. 11619 of 
the herbarium of the Instituto fisieco-geogrifico de Costa Rica, arriving at the 
conclusion that it belonged to an undescribed genus, About the time Mr. Don- 
nell Smith published his new species I was greatly surprised to receive from 
Doctor Renson a fruiting specimen of the Salvadorean plant, which showed 
at a glance Captain Donnell Smith's mistake, and also the close affinity of his 
Rauwolfia stenophylla to my Tondusia parvifolia. 
\ a 
