72 H. F. Blanford — ^ome further results of tJie sun-tliermomeler. [No. 4, 



8. Euplecta ornatissima, Benson. 



9. Do. unknown. 

 10. Do. crossei, Pfr. 



^11. — Some further results of sun-thermometer observations with reference 



to atmospheric absorption and the supposed variation of the solar heat. 



— By Henry F. Blanford, F. R. S., Meteorological Reporter to the 



Government of India. 



[Received 28th December, 1882.] 



In 1875, I read a paper before tlie Society, in which I discussed the 

 temperatures observed with the sun-thermometer, at eleven Indian stations, 

 during the j^ears 1868-1874, and arrived at the conclusion that the solar 

 heat had undergone a rapid increase from 1868 to 1871, and a less rapid 

 decline afterwards, up to 1874. 



The data were discussed according to various methods, but that on 

 which I chiefly relied, as taking count of tlie largest amount of data, and 

 being the best calculated to exclude the disturbing influence of atmospheric 

 variation, was to select days on which there was either no cloud, or on 

 which the cloud canopy, on the average of the 10 a. m. and 4 p. M. observa- 

 tions, did not exceed one-fifth of the sky expanse ; and having taken the 

 monthly averages of all the sun-thermometer readings on such days, to 

 compare these averages for homonymous months at each station, in each 

 pair of consecutive years. The months June, July, August and September 

 were excluded from this comparison, inasmuch as, at nearly all the stations, 

 the registers of which were discussed, these months are too cloudy to fur- 

 nish a sufficient number of available readings. For the remaining months, 

 the mean progressive variation of all the stations was taken, for each pair 

 of years ; and finally, the mean of the eight months gave the adopted varia- 

 tion for the consecutive years. 



In order to ensure that such comparison should be valid, the investi- 

 gation was restricted to stations, at which the same instrument had been 

 in use in each pair of years compared, exposed in the same way, and on 

 the same site. The curve of annual variation, resulting from these data, 

 coincided, in a marked manner, with the sun-spot curve ; but, in reality, 

 striking as it was, this result was vitiated by errors from two sources, both 

 of which tended to disturb and diminish the coincidence. One of these 

 was the inclusion of the Silchar registers, which, as I afterwards discovered, 

 had not been kept under similar conditions throughout ; so that those of 

 all the earlier years gave too lovv a temperature ; and hence a marked increase 

 of insolation temperature, shewn by this station in the later years, (when, 

 according to the general result, that temperature was falling,) was not real. 

 The other, the effect of which was however small, was an error of method : 



