ON THE KEW OBSERVATORY. 367 



The details will be stated wlicn these, together with additional comparisons, 

 shall have received more complete discussion than they yet have. 



The following experiments and remarks were made in the course of these 

 inquiries, &c. 



Instead of taking the mean between the reading of the immersed thermo- 

 meter at the first appearance of dew on the bright bulb of the Daniel, and the 

 reading at the time of its disappearance for the determined dew-point, in order 

 to eliminate the error which might arise in the estimation of the dew-point 

 from the continued fall of the thermometer after the first appearance of dew, 

 the observer checked the internal evaporation from the aether in the bright 

 bulb, " whenever the dew has shown itself by touching with the hand the 

 covered bulb (upon which the aether is poured), and thus preventing an undue 

 accumulation of water" on the briglit bulb, and giving time for the mass of 

 aether, the immersed thermometer and the bulb itself to acquire a uniform 

 temperature at the dew-point moment. 



In the cisterns of the Regnault hygrometers, the continual agitation of the 

 aether (by the passage of air bubbles through it) served to produce the re- 

 quisite uniformity of temperature throughout the mass of aether and the bulb 

 of the immersed thermometer. 



In experimenting with the improved Regnault, the difficulty alluded to at 

 p. 345, of "catching the exact instant of the formation of dew," was not found 

 to have been completely obviated by the method there described*. But " the 

 agreement between the two dew-point instruments is generally good, the 

 discordance rarely amounting to one degree ; the dew-point by Regnault 

 being generally somewhat lower than by the Daniel." 



" The expenditure of aether is somewhat greater in the Regnault than in 

 the Daniel, which however may perhaps be compensated by the inferior qua- 

 lity consumed." 



Experiments similar to those made on May 16 (vide p. 366, ante), on the 

 effect of a current of air on the wet bulb, " have been repeated several times 

 since, and always with a like result. It has also been found that a moderate 

 agitation of the air produces the maximum amount of depression. Similar 

 results have been found in the ordinary comparisons in the open air ; the 

 agreement between the computed and observed dew-points being greater 

 when a moderate wind was blowing than when the air was still." 



Barometers. 



Between September 23 and November 16, 1850, a series of 167 com- 

 parisons was made between the mountain barometer and the photo-barome- 

 trograph ; " but as there can be little doubt that considerable discrepancies 

 which the comparison showed might be partly owing to the imperfect stand- 

 ard of comparison employed, no confidence is placed in the results from this 

 series." 



In October Colonel Sykes recommended the acquisition of a good standard 

 barometer. 



In January the standard barometer of Newman (vide p. Sie, ante) arrived 

 from Colonel Sabine, and soon afterwards, by his desire, its new attached 

 thermometer was added. 



Between January 23 and March 30, about 200 comparisons, &c. were 

 made between this barometer and the photo-barometrograph ; the principal 

 results of which were, that the range upon the Daguerreotype plate (at E, 



* Further trials on this point are in progress. 



