36 



kets, crawl over the pistillate flowers to sip at the drop of nectar 

 which is secreted in the angle between their stigmas. The se- 

 cretion tastes slightly sweet, and glitters like a dew-drop in the 

 morning sunlight. So by the combined pollen and nectar-seek- 

 ing propensities of these bees the flowers are cross-pollinated. 

 Whether wind pollination would be effected in the absence of 

 bees, and how much transfer of pollen by the wind actually 

 does take place, was not determined, but judging from the ex-' 

 posed position of both anthers and stigmas, this form of pollina- 

 tion seems not unlikely. 



The fruit is set freely at Almirante, and ripens in about 

 three months from the time of pollination. The mature drupe 

 is ellipsoidal, about 15 mm. long, shining, bright red, and sub- 

 tended by the withered perianth segments. The thin, fleshy ex- 

 ternal pericarp is much sought by birds, especially by the Blue 

 Tanager {Thranpis carta diaconus Ridgw. and Nutt.) and the 

 Black-winged Palm Tanager (T. palmar um atripennis Todd). 

 These peck away the edible portion of the fruit while it is still 

 attached to the tree, finally letting the hard seeds, surrounded 

 by the fibrous internal portion of the pericarp, fall to the ground, 

 where they germinate in large numbers. A larger bird, such as 

 a toucan or large dove, might swallow the fruits whole, later 

 rejecting the indigestible seed, and so disseminate the species. 



Observations on the pollination of palms are apparently not 

 numerous. According to Knuth (1), who devotes a scant page 

 to the family, the nectar-secreting, proterogynous flowers of 

 Sabal adansoni are pollinated by hymenoptera. Species of 

 Chamaedorea are also entomophilous. Cocos, Syagrns, Phoenix 

 dactylifera and Chamaerops humilis, according to the same au- 

 thority, are wind-pollinated. Spruce (2) records the following 

 interesting observations regarding the fragrance of palm flow- 

 ers: "The flowers of palms are, it is true, comparatively small, 

 and being usually of a pale yellow color, are conspicuous only 

 when massed on the large spadices of the taller-growing species; 

 but in their exquisite odor they often yield to no flowers what- 

 ever. In many cases the odor is that of mignonette, but I think 

 a whole acre of that darling weed would not exhale as much 

 perfume as a single male spadix of the Carana palm (Mauritia 

 carana) of the Rio Negro. The flowers of the slender Sangapilla 

 palm of the Peruvian Andes preserve their fine scent for months, 



