82 THE CLIMATE. 



winter temperature for those hours is 60. 4; and as it has been found that the results of 

 records at 8 A. M., noon, and 4 p. M. exceed the true mean temperature 6. 5, one may well 

 appreciate what its inhabitants enjoy. Commander Wilkes found the mean temperature of 

 May, 1839, 55. 2, and the range of the thermometer from 46 to 65. It has been supposed that 

 the cold current of water from the vicinity of the Antarctic circle, that has been traced as far 

 as the coast of Peru, and which, from its .discoverer, has been named Humboldt current, has 

 some influence in modifying the climate of maritime Chile ; but there are too few observations 

 to prove that its eastern limit is much within the island of Juan Fernandez on this parallel. 

 The temperature of the water found by the United States Exploring Expedition was 56, from 

 which degree of heat Commander Wilkes had reason to believe that it varied very little through- 

 out the year. 



As we proceed northward from Santiago, the humidity of the climate varies more rapidly than 

 a mere difference of latitude seems able to explain. Beyond the spur uniting the two chains of 

 mountains, there are no extensive plains or valleys except the transversal depressions through 

 which flow the mountain streams ; and within one geographical degree of the capital the number 

 and duration of the rain-storms is diminished quite one half. In the parallel of 30, even on 

 the coast, there are rarely more than five or six moderate showers, and sometimes not more than 

 two, though night-dews at this season are heavy and constant much further north. Deposits 

 of snow on the Andes are frequent, and from the melting of these the inhabitants anticipate 

 benefits scarcely inferior to those which winter rains afford. This snow-water brings a current 

 of cold air through the ravines with it, and fogs are frequent over the mouths of the streams 

 where the contact of the cold with the warmer and moister atmosphere of the ocean takes place. 



No register of meteorological phenomena has been regularly kept except at La Serena, in lati- 

 tude 29 54'; and even this embraces only pressure and temperature, except a hygrometric 

 remark on the occasion of an earthquake. Situated at the margin of the sea, the climate of 

 La Serena is sensibly modified by it ; its mean winter temperature (54. 8) being more than 

 5 greater than that of Santiago, whilst the mean of the extremes between 8 and 9 A. M., 

 and 9 and 10 p. M., is only IV . 4. Taking the decrease of temperature with elevation found 

 by Humboldt and Boussingault at 1 for each 330 feet, the difference of level of the two cities 

 will account for 5. 5, leaving 0.3 only to be explained by the 3 32' of geographical latitude 

 and the difference of their topographical locations. Though less extensive than at Santiago, the 

 barometric oscillations are greater than at any other season. Its mean diurnal tide is only 

 0.010 of an inch. No violent winds occur. The southerly monsoon prevails, though it is 

 not uniformly attended by clear skies. When it changes to north or N.W., clouds and precipi- 

 tation of moisture usually follow. 



In latitude 27 we are almost at the northern limit of the rain zone, and there the risk of 

 injury from rain is so small that the larger number of houses are roofed with canes externally, 

 plastered only with mud. If three slight sprinkles occur during the year, the country is 

 regarded very fortunate, and the sandy surfaces of terraces and ravines are soon clad with 

 verdure and flowers. But it is rare to have two entire cloudy days following each other, and 

 at mid-winter the prevailing westerly wind most annoyingly drives the dust of this region 

 before it. There are no instrumental data to refer to, except during a few days of July in 1851. 

 On the coast, I found a low barometric pressure, small diurnal tide, and extreme oscillations, 

 with predominant though light winds from the northward. On one occasion, when the baro- 

 meter had been 0.250 of an inch below its normal elevation for three days with a partially 

 overcast sky, a sudden change of the wind from W.N.W to S.W. was almost instantly fol- 

 lowed by rain, though half an hour previously the temperature of the air and of evaporation 

 differed by 4. 8. The mean of these differences during five days was 4. 38 ; mean'temperature 

 during the same period, 56. 6 ; highest at any hour, 66 ; and lowest at midnight, 50. 5. 



Fifty miles inland the hygrometric condition of the atmosphere is wonderfully different. 

 There, whilst the mean temperature of evaporation was 51. 3, that of the air was 61. 2, and 



