July, 1922] BAILEY — POLLINATION OF MARCGRAVIA 
379 
ously glandular, broadly elliptical or obovate, 5.5-8.5 cm. long, 3.5-5.5 
cm. broad, or small oblong or ovate, 2.5-4.0 cm. long, 1.5-2.5 cm. 
broad. Umbelliform racemes dark greenish purple, pendulous. Pedi- 
cels numerous, slender, 3.5-4.0 cm. long, regularly and evenly distributed, 
straight and nearly horizontal or curved downwards at time of flowering; 
corky excrescences few, inconspicuous; bracteoles widely separated from 
the calyx. Calyptriform corolla ovoid or ovoid-conical, thin, coriaceous, 
8-11 mm. long, 7-9 mm. diameter. Stamens non-coherent; filaments 
stout, short, zh 5 mm., quadrangular, abruptly tapering above; anthers 
broadly lanceolate, reddish, d= 5 mm. long. Ovary turbinate-obovoid, 
± 3.5 mm. diameter, abruptly tapering into a short, blunt style, 2.5 mm. 
long. Fruit small, globose, =b 10 mm. diameter, reddish, Scabrous; style 
and stigma conspicuously protuberant. Nectaries elongated, 5.5-6.5 cm. 
long, bilaterally compressed, swollen at the apex, tapering towards the 
base, orifice surrounded by a slightly protruding rim or lip; petiole ± 7 
mm. long; .fused pedicel showing as a slightly embossed, verrucose rib; 
flower bud rudimentary, depressed or protuberant. 
Banks of the Mazaruni and Cuyuni Rivers near Kartabo, British Guiana. 
Bailey nos. 136, 179, 192, 195, deposited in the Gray Herbarium of Harvard 
University. 
Pollination of M. cuyuniensis and M. purpurea 
(I) . , 
I first was led to question the accuracy of Belt's generalization con- 
cerning the pollination of Marcgravia by a detailed study of the arrange- 
ment of the nectaries and flower buds in M. cuyuniensis and M. purpurea. 
In these species, as in M. umhellata L. and most representatives of the 
Brachyracheae, the pitchers are placed close to the pedicels of the flowers 
(text figs. 2, 4). Therefore, according to Belt's hypothesis, birds would 
have to visit these inflorescences from above, but the flower buds and flowers 
are not turned upwards. 
During my stay at Kartabo, I did not succeed in finding birds in the 
vicinity of the inflorescences of M. cuyu?iiensis, although many of the 
nectaries contained a sweetish liquid. Dr. and Mrs. Alfred Emerson and 
Miss Anna H. Taylor were more fortunate in the case of M. purpurea, 
which flowered after my return to the United States. A humming bird 
was seen to hover above an inflorescence and to sip nectar from one of 
the pitchers. The following observations of Miss Bryant (1905) are also 
significant in this connection. She states concerning M. umbellata L. (?): 
The plant is common here climbing to the summit of the forest trees, and is frequently 
visited by humming birds. The bird settles on the top of the flowers (inflorescences) 
and inserts its long curved beak into the pitchers below. 
Such observations as these suggested, of course, that the highly special- 
ized inflorescences of these Marcgravias are not efficient mechanisms for 
insuring cross-pollination by birds. 
