oj the Island of Barbadoes. 



m 



peraturcs ; that notwithstanding these precautions, the temperatures 

 near the hour of minimum were somewhat affected by the slight excess 

 of heat in the room and surrounding materials, above that of the air, I 

 have no doubt j but am inclined to think that it must be very slight, 

 and will not materially affect the deductions to be drawn from the whole 

 body of observations. 



The foUoAving observations were made at mean time, as shewn hy a 

 good watch, the errors and rate of which were found by altitudes of the 

 sun once a week. Every observation was made by myself, and I think 

 that nowhere in the hourly observations can the difference, on the mean 

 of a month, between the real time of observation and corresponding 

 hour, exceed a minute either way. 



Table I., shewing the means of Temperature for each hour in the months 

 of June, July, and August 1841, at Barbadoes. Also the whole num- 

 ber of Observations at each hour, and the corresponding mean Tem- 

 peratures for the three months, together with a series of numbers 

 calculated on the supposition that the diurnal changes of Tempera- 

 ture may be represented by j.arabolic curves. 



