SCHISTOCERCA, SPHINGONOTUS AND HALMENUS 417 



almost no soil covering the surface lava. Hood, James, Barrington 

 and Tower probably never had a single central crater, but all of them, 

 except James, shovv^ little evidence of age. 



The winds of the archipelago are almost invariably from the south- 

 east. The humidity of the region is so small that islands less than 

 1 ,000 feet in height have but little precipitation. Consequently high 

 islands lying to the south and southeast are the most favorably situated 

 for receiving rain. Chatham and Charles possess both of these quali- 

 ties. Hence, on them the condition of the surface cannot be relied on 

 as a definite index of their age. They may appear older than another 

 island of the same age that is lower and less favorably situated. Hood 

 is just as well located as either Chatham or Charles, but it is too low 

 to condense much moisture. 



The contrast is so great, however, between Chatham and Charles 

 and the other islands, that we cannot but suppose, even after making 

 allowance for all other influences, that they were at least the first 

 inhabitable islands of the archipelago. Albemarle rises far above all 

 the islands east of it. To be sure, air currents must reach it somewhat 

 impoverished in humidity by Chatham, Indefatigable, Charles and 

 James, but its perfect craters and great fields of barren lava attest its 

 newness. James and Indefatigable appear to be older than Albemarle 

 but younger than Chatham and Charles. What the relative ages of 

 such low islands as Hood, Barrington and Tower may be, however, it 

 is difficult to judge. The rainfall on them is so slight that they may 

 have remained unchanged for a long time. 



A more serious difticulty in judging the age of an island arises from 

 the consideration that only its present surface may be recent. A much 

 older and fertile surface may be buried beneath it. This is true of 

 Narboro. Most of the surface of this island consists of unweathered 

 and utterly barren lava. There occur, however, scattered over its sides 

 numerous isolated patches of vegetation growing on a rather rich soil. 

 These areas vary from a few rods to half a mile in diameter, and are 

 walled in on all sides by the recent flows of lava that have covered all 

 the remaining surface of the island. These flows have occurred at 

 different times and each has been of small extent. On such an island 

 a fauna might easily migrate from one place to another as successively 

 occupied areas became covered up. 



In conclusion, then, with regard to the relative ages of the islands, 

 all we can affirm is that Chatham and Charles have the appearance of 

 being the oldest and first habitable islands of the archipelago. As we 

 have seen, Schistocerca literosa is represented on Chatham and S. 



