CYCLONES OF THE WESTERN PACIFIC. 353 



with the wind at south ; and at no time was in advance of the center or axis of the cyclone. 

 The harometer stood lowest, 29.20, about 8 a. m. of the 14th, at the time when the topmasts 

 were blown away. 



KINGSMILL ISLANDS, GILBERT ARCHIPELAGO. 



These islands, situated on and near the equator, longitude about 175° E., were visited by the 

 United States ship Peacock, of the exploring expedition under Captain Wilkes. Variable winds 

 from the northward and westward prevail from October to April ; and they have violent gales 

 from the southwest. According to Kirby, who was taken off the islands, these storms are 

 typhoon-like, and last three or four days. The westerly sides of the islands receive most damage, 

 and both land and trees are swept away. 



THE RADACK ISLANDS. 



These islands are scattered between 6° and 11° N., and longitude 168° to 173° B. Captain 

 Kotzbue ascertained that hurricanes of great violence sometimes occur in September and October, 

 and the natives always anticipate with dread the recurrence of those months. 



THE JAPAN'S TYPHOON. 



In December, 1832, the Japan, a new ship, encountered a severe hurricane in latitude 13° 

 N., longitude about 160° west. 



This position is about on the meridian of the most western of the Sandwich Islands. 



The tracks or paths of those cyclones, of which the dates and positions have been given, are 

 indicated on the annexed chart. 



SANDWICH ISLANDS. 



At the Sandwich Islands, latitude 19° to 22° N., longitude 155° to 160° W., the cyclones 

 which occur are, commonly, not of great severity ; although the native huts are sometimes 

 unroofed or destroyed. The kona, or southerly wind, by which the trade wind during part of the 

 year is much interrupted, may be referred, at least in part, to those cyclones which have their 

 center-path northward of the islands, or which complete their recurvation in that region. The 

 easterly gales which accord nearly with the trade winds in their direction, indicate an axis-path 

 which lies southward of the islands. The actual presence or influence of a cyclone may, com- 

 monly, be determined by the oscillation of the barometer. 



The absence of intense violence in any of the cyclones which visit this group of islands, may 

 possibly be due to their geographical position. But it is equally probable that this qualified 

 exemption may result from a diversion of the cou*se of the central vortex of the cyclone, occa- 

 sioned by the great height and compact form of Hawaii, the most southeastern of these islands. 

 For the group lies in almost a direct line, which is parallel to the ordinary courses of the 

 cyclones in the lower latitudes ; being, from the summit of Manua Kea to the centre of Kauai, 

 north sixty-one degrees west. A cyclonic vortex, if moving previously on this line of direction, 

 would be displaced by the eastern angle of Hawaii, which island has an area of near four thou- 

 sand square miles, a portion of which rises far above the upper horizon of the cyclones, and at 

 two points reaches an elevation of nearly fourteen thousand feet. The protection, or partial 

 diversion of course thus occasioned, must extend in'good measure to the high but smaller islands 

 which lay in the same track. 

 45.? 



