BOTANY OF FEENANDO NOEONHA. 13 



small sj)ecies of moths haunted at night the bushes of Scoparia 

 dulcis. Cassias, &c. on the open spaces. A single species of 

 butterfly was very abundant on Eat and the main island, but 

 we never saw it visiting flowers. 



The most important fertilizer was a small endemic hornet 

 belonging to the genus Polistes, which gathered honey from the 

 Leguminosa3 and Cucurbitacese ; and three small black species of 

 Salictus were caught in the flowers of the melons, Momordica 

 Charantia^ Oxalis Noronhce, and the mustard. The latter plant 

 was also haunted by Teinnoceras vesiculosus, a pollen-eating 

 Syrphid, The only other insects which could also be considered 

 as ])ossible fertilizers were TacTiytes incoiispiciius, n. sp., and 

 Monedula signata, two sand-wasps, Pompilus nesopliila, n. sp. 

 (Hymenoptera), and Psilopus metallifer (a Dipteron), but none 

 of these were seen at or near flowers. A small black beetle also 

 was found in the flowers of an Acacia in the Grovernor's garden. 



Though the number of species of insects was not large, yet 

 the individuals, especially of the Polistes and Halicti, were very 

 numerous, but at the same time they seemed out of all propor- 

 tion to the immense number of flowers to be fertilized. It is 

 very probable, however, that the majority of the Leguminosae 

 and some of the other plants were self-fertilized. 



Groups of Plants rare or unrepresented. 

 The, absence of plants or groups of plants from a given locality 

 often throws as much light on the origin of the flora as the 

 presence of others does. In the present case the absence of 

 marsh-plants is among the most striking ; for in the first place 

 they are exceedingly abundant on the adjacent mainland, and, 

 again, there is every reason for their being introduced by the 

 wading-birds which fly across from Brazil. The genera Eleocharis, 

 Utricularia, Poepalantlnis, Saleria, and many others remarkably 

 abundant on the adjacent mainland, are here quite absent. 

 Indeed, the only really marsh-loving plants met with were 

 Jussieua linifolia. Ammonia latifolia, and Panicum hrizoides. 

 The dryness of the island during the dry season accounts for this 

 in the main ; but there are spots which are permanently damp, 

 and here one might reasonably expect to find marsh-plants. 

 The chief plant which grows along the streams and on these 

 damp spots is, however, Philoxeriis vermicularis ; and as the 

 water has a brackish taste, it is probable that the salt or other 



