INTRODUCTION. XXXV 



the trees that were prostrated at the Bay of Islands were found 

 lying with their tops to the northward. The gale came on with 

 the wind from the southward and eastward at all the stations, this 

 was nearly at right angles to its actual line of movement : the veer- 

 ing of the gale was from left to right. It was one of the most de- 

 structive that had occurred at New Zealand, the water rising to a 

 great height, and overflowing a part of the island which had before 

 been thought beyond the reach of the sea. The lowest standing of 

 the barometer on board H. B. M. Ship Herald was 28 - 75 inches, when 

 the gale was most violent. The change of wind took place at 2 A. M. 

 of the 1st of March, and on board the Flying- Fish it was at its height 

 at 3 A. M. Throughout the gale the thermometer did not descend 

 below 66 Fahrenheit, the temperature of the surrounding sea. 



On our passage to Tonga, one of the Friendly Islands, we did not 

 enter the trades until we reached the latitude of 24 26' south, when 

 they blew from the east. The barometric column again decreased 

 139 in., while the temperature of the air increased 13-3, and the 

 water 9-7. 



Tongataboo comes under the class of low coral islands, but its cli- 

 mate is dissimilar ; it is more humid, and the heat more oppressive. 

 The daily transitions from heat to cold arc much complained of, 

 though we found but 8 difference in the temperature. The dews 

 are remarkably heavy, which may account for the feeling of cold at 

 night; they are said to render the climate unhealthful. The mean 

 temperature of the air, during our stay of ten days, in the month of 

 April, was 79-05, that of the water 79-28; the barometer stood at 

 29-808 in.; the degree of moisture '893. 



Hurricanes are frequent in this group. The months from Novem- 

 ber to March are those in which they occur. According to the 

 accounts given me, these begin with the wind at northwest, go round 

 to the eastward, and end at the southeast. Sudden changes during 

 the height of the gale are experienced, and trees are prostrated with 

 their tops in all directions. It appears that the progress of these 

 storms is from the northeast to the southwest, corresponding to what 

 we found to be the case at New Zealand, and nearly at right angles 

 to that before described at the Samoan Islands. The courses of these 

 gales having been well ascertained, we must conclude that the paths 

 of these storms are deflected, somewhat like those of our own hemi- 

 sphere, only in a contrary direction, moving from the northwest to 

 the southeast in the lower latitudes, and thence passing to the higher 

 latitudes by a course towards the southwest. The same gale at the 



