of the Rutherfurd Photographs. 221 



the coordinates would require a correction to reduce them to 

 what they would have been had the temperature remained con- 

 stant. Investigation shows, however, that such a correction is 

 unnecessary by reason of its minuteness, at least within the limits 

 of the range of temperature at which the present plates were 

 measured. To ascertain the amount of the correction, two well 

 defined specks, such as may be found in the film of any plate, 

 were selected, one near either edge of the plate, and the distance 

 between them was measured at various temperatures. On the 

 morning of April 30, 1897, this distance was measured six times 

 each by Mr. Kretz and myself with the following results : 



Kretz : 

 104. 1505 mm. 

 514 

 505 

 497 

 54o 

 517 





Schlesinger : 





104.1548 mm 





67 





36 





30 





50 





25 



Mean, 



I04-I543 



Probable Error, 



±0.00042 



104.1513 



±0.00041 



The temperature of the measuring room had been kept at 6g.°$ 

 during the measurement by means of artificial heat, the heating 

 apparatus being at the other end of the room from that occupied 

 by the measuring machine. The heat was now turned off and the 

 plate allowed to assume the natural temperature of the at- 

 mosphere ; on the afternoon of the same day, three hours having 

 elapsed since the first series was completed, the distance between 

 the specks was again measured by the same observers as follows : 



Kretz : 

 104.1557 mm. 



44 

 48 

 60 

 53 

 75 





Schlesinger : 





104.1550 mm. 





70 





63 





87 





90 





70 



< 



Mean, 



104.1572 



Probable Error, 



±0.00041 



104.1556 



±.00031 



The temperature for this series was 52. °2. Denoting by v the 

 increase in the measured distance due to an increase of i° in the 

 temperature we have 



Annals N. Y. Acad. Scr., X, May, 1898—15. 



