74 SCIENCE PROGRESS 



of the criticisms being valid, as Mr. Lambert asserts in the 

 passage quoted above, the one of major consequence zs justified 

 by his own admission. Moreover, the criticism is still applicable 

 to his later work, as he cannot possibly have heated the entire 

 surface of his apparatus, and the degree to which he heated 

 parts of it must have been such that it is not likely that he 

 did more than drive off the major part, let us say, of the carbon 

 dioxide condensed on and within the glass, thereby reaching an 

 equilibrium, perhaps, but never removing the whole of the gas. 



In order that there may be no misunderstanding, let me say 

 that I hold it to be impossible to obtain valid results with an 

 apparatus of so complicated a character, in which so large a 

 surface of glass is exposed, as that used by Mr. Lambert ; 

 infinite opportunity is given in such an apparatus for the 

 retention of carbon dioxide at the glass surfaces. 



Mr. Lambert's views are summarised in the statement : 



"The glass walls of all the vessels with which the water 

 and oxygen came in contact had been subjected to the exhaustive 

 treatment described above and so it may be said to be proved 

 beyond any reasonable doubt that oxygen and water — both of 

 the highest obtainable purity — have the power, of themselves, 

 of causing commercial iron to rust. 



" Further, the rusting seemed to take place as quickly as it 

 does in ordinary air or oxygen and so it cannot any longer be 

 maintained that carbon dioxide or any other acid is the ' dominant 

 factor in the atmospheric corrosion of iron,' where commercial 

 forms of the metal are meant." 



Reading back we learn what is here meant, by "commercial 

 forms of the metal " : 



" Three different kinds of commercial iron were used, one 

 in each vessel — namely, (i) a pure commercial electrolytic sheet 

 iron ; (2) Kahlbaum's pure iron foil ; and (3) a cylinder of iron 

 turned from a large nail taken from the roof of Merton College 

 library while repairs were being carried out. This nail was 

 made of very soft iron and was more than two hundred years 

 old. 



" The iron in each case was carefully polished with fine 

 carborundum and then with Swedish filter paper. The results 

 in all three cases were the same and did not differ in any respect 

 from the results obtained with other good specimens of com- 

 mercial iron used in earlier experiments. Corrosion was visible 

 in a few hours and a considerable quantity of rust had formed 

 within a few days." 



Mr. Lambert took none of the precautions to cleanse the 



