Mancheste}' Memoirs, Vol. lix. (191 5), No. \%. 45 



compared with those of the sheet, and were found to 

 point to its date as lying between October, 1806, and 

 July, 1807. Whilst the figures for oxygen, azote, sulphur 

 and phosphorus show that the sheet was not produced 

 before 1805 or 1806, the most significant figure is that for 

 iron, which at the first appearance of this sheet was 40. 

 Now, in September, 1806, Dalton put the atomic weight 

 of iron as 29 or 58 ; on October 22nd of the same year it 

 became 40, but in July, 1807, was again altered to 50, 

 and did not again become 40 so far as can be discovered. 

 The other atomic weights are all in accord with the date 

 ascribed, with the exception of two : the figure for carbon 

 is 5'4 15 on the sheet, a value not found in the published 

 works or laboratory books until 18 10 (" New System," 

 I., ii.) ; for mercury the atomic weight on the sheet is 

 167, a figure which appears first in 1808 (" New System," 

 I., i.) where it replaces the 166 of July, 1807, and October, 

 1806. 



The latter difficulty is probably of no significance, since 

 the difference is only of one unit in the large value 166 to 

 167. The former difficulty — the atomic weight of carbon 

 being 5*4 when in contemporary lists it appears as 5 — was 

 removed as the result of an enquiry as to the reason for the 

 compilation of this sheet of atomic weights. It seemed 

 reasonable to expect that this was drawn up for use 

 during some important lecture, and it was therefore no 

 surprise to find that Dalton gave courses of lectures in 

 Edinburgh and Glasgow in April, 1807. The notes for 

 these lectures, which are also among the relics in the 

 Society's library, contain worked out several atomic 

 weights, together with reference to a ' l scheme " (appar- 

 ently the present sheet). The atomic weight of carbon 



16 Without doubt the figure in the first decimal place 1 elonged to the 

 original compilation of the sh< 



