1891.] D. Prain — On the present condition of Barren Island. 



87 



(Stray Feathers ii, 104) " tlie thermometer showed a temperature of 

 " nearly 140° F." This remark occurs in the narrative portion of a 

 paper on the ornithology of the islands of the Bay of Bengal and is 

 clearly written from memory only ; attention is drawn to it now merely 

 in order to prevent any one who may subsequently read the statement 

 from relying upon it as exact. Mr. Mallet's observations are very com- 

 prehensive and precise. Only the bare result of Mr. Daley's observa- 

 tion is given and only one temperature is mentioned : it is not stated 

 that it was the highest, and it is possible that the temperature was taken 

 at one spot only. 



The rate of cooling of the spring as indicated by the various highest 

 readings of the preceding table is shown in the table subjoined : — 



Period. 



No. of years. 



Amt. of cooling. 



Rate per annum. 



1866-73 



7 



33° F. or 18-30^0. 



4-7° F. or 2-6°C. 



1873-84 



11 



14° F. or 7-70C. 



1-2° F. or 0-6°G. 



1884-86 



2 



6° F. or 3-3°C. 



3-0° F. or l-7°0. 



1886-91 



5 



4° F. or 2-2°C. 



0-8° F. or 0-4°C. 



The rates per annum shew that the existence of an error in the 

 1886 observation is not only possible but highly probable ; the spot 

 selected for examination can hardly have been that in which the tem- 

 perature was highest at the time. If it be neglected the rate per annum 

 for the whole 1884-91 period is ^ or 1-4° F. ( = 0-7°C.) a rate that, 

 allowing for errors unavoidable in such observations, is practically 

 uniform with that during the 1873-84 period. Between 1866 and 1873 

 it cooled nearly four times as fast as it has cooled since ; if the 1886 

 observation is correct, six times as fast as it is cooling now. If the 

 rate continues uniform henceforth the spring will, if the 1886 observa- 

 tion be correct, probably cease to deserve the appellation " hot " by 

 1925, if the 1886 observation was not taken where the water was hottest 

 at the time, the spring will probably be cool by 1910. 



