REDONDA AND ITS PHOSPHATES. 87 



tion of which they were at liberty to go home if they chose, or 

 were discharged if no longer needed. 



On that Sunday there were one hundred and one people on the 

 island, of whom five only were white. However, this proportion 

 of black to white was no greater than in the neighboring islands. 



In the afternoon, while taking a nap, I was awakened by what 

 I vaguely thought to be the thunder of a coming storm ; but it 

 proved to be Chalmers trying to drive a goat off the iron roof, to 

 which it had sprung from the steep incline behind the house. 

 Though it was now the beginning of the hurricane season, the 

 weather was calm and fine during our whole stay in the tropics. 



On Monday the brigantine Foley arrived for a cargo of phos- 

 phate, and we went to the lookout west of the house to see her 

 drop anchor. We were at least six hundred feet above the sea, 

 and as the vessel lay in the shelter of the cliff she looked like a 

 boy's ship floating on a pond. The wind was blowing briskly at 

 the time, but the island afforded a perfect shelter against it, and 

 the calm area could be seen extending like a shadow over the sea 

 for half a mile. This protection from the wind also caused it to 

 be almost unbearably hot down on the beach in the afternoon 

 sun, which was reflected from water and cliffs. 



On Tuesday, July 8th, we bade good-by to Mrs. H and 



Miss Dorothea, and descended the wires for the last time. Cap- 

 tain H went with us aboard the sloop which was to take us to 



Montserrat. We were soon on our way, and the ensign of Great 

 Britain, flying in front of the house, was dipped three times. We 

 waved a final adieu, and the lofty walls of Redonda were there- 

 after seen by us only from a distance. 



Mr. Jacques W. Eedwat expresses the opinion in the Geographical Journal 

 that the reason wliy the prairies and plains of the United States are treeless, is 

 because they have never been seeded with trees, and this because thej have never 

 been exposed to inundations from tree-bearing districts. '' Water," he says, "has 

 been the chief agent in the distribution of trees, and the treeless regions are the 

 greater part in regions that have not been disturbed by physiographic agencies. 

 From the southern limit of glaciation to the made lands along the coast of the 

 Gulf of Mexico, the central plain of the United States is the level bed of a Pahneo- 

 zoic sea. Excepting such places where the streams of Champlain times have cut 

 channels through the upper strata, the surface of this vast plain is undisturbed; 

 it is at once a sedentary soil of Silurian disintegrations and a Quaternary epoch. 

 Throughout much of its extent it is treeless, not because of prairie fires, nor yet 

 of unwholesome conditions of the soil, but from the simple fact that the seeds of 

 forest trees have never been distributed over its surface at fortuitous times. 

 Prairie fires have doubtless had more or le^s to do with retarding the distribution 

 of forestry; so undoubtedly have unwholesome conditions of the soil. Neither 

 condition, however, is suflSciently potent to prevent the emlolsment (tree-clothing) 

 o a treeless area; it is still less able to deforest a timbered area." 



