242 



DISTRIBUTION OF PLANTS. 



" Convolvulus." (Compare No. 1 Otafuan coral-islands, and I. turpethum of Gardner's 



coral-island, Metia, and the rocky Hawaiian Group). Observed by Mr. Rich. 

 " Cordia." (Compare No. 1 Otafuan and Gardner's coral-islands, and the rocky Taheitian 



and Feejeean Groups). Observed by Mr. llich. 

 Tournefortia (bis No. 1 Paumotuan to Gardner's coral-island and the rocky Feejeean 



Group). Perhaps the most abundant plant. Observed by Mr. Rich, in some 



instances " twenty feet" high. 

 "Pisonia; thirty feet." (Compare Calpidia No. 1 Paumotuan to Rose, the Otafuan and 



Gardner's coral-islands). Observed by Mr. Rich. Doubtless the small trees I saw in 



the distance, near the cluster of cocoa-palms. 

 Boerhaavia; bis Gardner's coral-island, (and compare No. 1 Paumotuan coral-islands). 



Frec[uent. Mentioned also in Mr. Rich's catalogue. 

 " ; a second species." Observed by Mr. Rich. 



Urtica ; bis Gardner's coral-island, (and compare Paumotuan, Bellinghausen, and Otafuan 

 coral-islands). Smooth. Mentioned also in Mr. Rich's catalogue. 



Granien. incert. An undetermined grass, growing in tufts. Mentioned also in Mr. Rich's 

 catalogue. 



24. Enderby's coral-island. Continuing Eastward, the Vincennes 

 next arrived in sight of Enderby's coral-ishmd ; situated in " S. Lat. 

 3° 8' and W. Long. 171° 08'," find found to be "three miles long, by 

 two and a half wide ;" from the mast-head, having the .aspect of a 

 grassy plain, with a few bushes in two separate localities. I could 

 distinguish no variation in level ; but those who landed, spoke of a 

 central " depression," and other indications of a shallow dried-up 

 " lagoon, as soft mud, and an incrustation of salt." 



The land-surface being "eighteen feet" in extreme height, the "dead 

 tree" lying on the top of the island, could hardly have drifted there 

 without some extraordinary and general overflow. For it may be 

 remarked, that the shelf around coral-islands is an impediment to 

 the dispersion of plants; as it prevents the surf from throwing drifted 

 seeds far enough for the wind to convey them inland. The Pandanus 

 was again absent. 



Besides the nautical means of identifying a small coral-island visited 

 by different navigators, this may sometimes be done by the flora, or 

 the plants growing thereon. The specimens collected by Mr. Bracken- 

 ridge on Enderby's coral-island were examined by myself, and a list 

 made out: subsequently, Mr. Rich communicated a list of the plants 

 growing on one of the Phoenix coral-islands visited by the Peacock. 

 On comparing the two lists, the result is as follows : 



