306 PROCEEDINGS OP THE AMERICAN ACADEMY 



whether end-measures behave in the same way as line-measures under 

 a varying temperature. 



(i.) The exact relation between the length of the Imperial Yard and 

 the Meter of the Archives. 



The Imperial Yard has never been directly compared with the Meter 

 of the Archives. Our knowledge of the value of the equation between 

 these standard units depends solely upon, — 



First. The indirect comparison of the platinum meters of the Royal 

 Society with the Meter of the Archives by Kater, Baily, Clarke, and 

 Chisholm. All these determinations rest upon the relation of the Royal 

 Society platinum meters to the Meter of the Archives given by Arago 

 in 1818. 



Second. The comparison of the iron meter of the United States 

 Coast Survey with Troughton's scale, by Hassler (27 m# — 63 m# ) in 

 1832. 



The following are the relations given by the authorities named above. 

 I have subtracted .00087 inch from the value given by Hassler, this 

 being the amount by which the yard on Troughton's scale exceeds the 

 length of "Bronze 11 " when reduced to 62°. 



Kater, 1818, " Metre a Traits " = 39.37076 inches. 



" Metre a Bouts " 39.37081 



Kater, 1820, Dolland's Scale 39.37045 



Baily, 1836, " Metre a Traits " 39.36968 



" Metre a Bouts " 39.36937 



Clarke, 1866, "Metre a Traits" 39.370-18 

 Chisholm, 1870, Standard Meter on Baily's metal 39.37112 



Hassler, 1832, Original Iron Meter 39.38005 



Mr. Hassler compared several other standards with Troughton's 

 yard, but they are not included in the above table ; first, because, with 

 the exception of the brass meter by Lenoir, there is no evidence that 

 the standards compared are authentic copies of the original; and, sec- 

 ond, because the certificate of the meter by Lenoir refers to the Meter 

 of the Observatory, and not to the Meter of the Archives. 



It is obvious from this table, that our present knowledge of the rela- 

 tion between the length of the yard and the meter is suhject to great 

 uncertainty. In deciding upon the weight which should be assigned 

 to any particular value, it is to be remembered, — 



(a.) That in the comparisons by Kater, Baily, Clarke, and Chis- 

 holm, the constant relation given by Arago is subject to great doubt. 



