44 THE VOYAGE OF H.M.S. CHALLENGER. 



of which no notification is given, and apparently no allowance is made for the change. 

 Thus at Hobart Town for some years, the hours of observation appear to have been 



9 a.m. and 1 and 5 p.m., and the mean of the observations at these hours was adopted 

 as the mean temperature, with the result of winters apparently 2° and summers 6° 

 warmer than before. The figures for Hobart Town in the table have been brought 

 to mean temperatures by correcting each year's observations by the table of corrections 

 for hourly range at this place. It may be mentioned that these faulty mean tempera- 

 tures at Portland and Hobart Town for long thrust the isothermals of this part of the 

 globe seriously out of their proper positions. 



In a large number of instances the monthly means in the table are the means of 

 particular hours of observation uncorrected in any way, such as 6 A.M., 2 p.m., and 



10 p.m. ; 7 a.m., 1 p.m., and 9 p.m.; 4 a.m., 10 a.m., 4 p.m., and 10 p.m. ; 8 a.m. and 

 8 p.m. ; 9 a.m. and 9 p.m. The means were corrected for daily range where such 

 corrections were required, and after being corrected for height, the resulting means 

 were entered in their places on the map. 



In correcting for height, the correction adopted is at the rate of 1° Fahr. for 

 every 270 feet in height above mean sea level ; and this correction has been uniformly 

 applied to the temperature observations for all seasons and countries. The rate 

 unquestionably varies with season and climate ; but as regards the manner and degree 

 of this variation, our information is so scanty, and the worked-out results in many 

 cases are so doubtful, and sometimes even so inconsistent with each other, that it is 

 more in accordance with the present state of our knowledge to adopt provisionally a 

 uniform rate of correction throughout, than a rate varying with season and climate. 



Of the causes producing variability in the rate of diminution of temperature with 

 height, the more prominent are season, hygrometric state of the atmosphere, and 

 situation. During the transition from winter to summer, when the great annual rise 

 of temperature is in progress, the rate of diminution of temperature with height is 

 greatest, for the simple reason that at this season the lower layers of the atmosphere 

 are more quickly heated by simple proximity to the earth's surface, thus increasing 

 the difference between the temperatures at low and high levels. On the other hand, 

 in autumn, when the great annual fall of temperature occurs, the lower strata of the 

 atmosphere are more cooled by the now rapidly cooling surface of the earth, and 

 accordingly the difference between the temperatures of the low and high levels is 

 proportionally lessened. 



Observations prove that the more aqueous vapour there is in the atmosphere in 

 the form of cloud, and to a large degree even in a purely gaseous form, the more is the 

 earth's surface protected from the effects of solar and terrestrial radiation. It follows 

 therefore that in rainy climates, and during the rainy season in the tropics, the rate 

 of diminution of temperature with height is comparatively a stable quantity hour by 



