ATOMIC WEIGHTS. 537 



would unite, with not the slightest excess of either. The 

 value for the volumes of hydrogen which unite with one 

 volume of oxygen is, according- to Morley, 2-00023, Leduc 

 2*0037, Scott 2 '00245. 



Undoubtedly Leduc's result has not the same value as 

 a direct measurement, for a very minute error in the den- 

 sity determination of the gaseous mixture would affect the 

 ratio of the volumes very considerably, and he himself 

 points out that variations in the pressure at which the gas is 

 being evolved very markedly affect the density of the mix- 

 ture and therefore still more so its composition. His ratio 

 of the volumes is, however, of the same order of magnitude 

 as that of the author, whilst that of Morley shows a very 

 much smaller (only about one-tenth) variation from the 

 exact value of 2 : 1. 



The apparatus in which the gas is measured by Morley 

 is practically that of the author as described in his prelimi- 

 nary note (17) to the Royal Society in 1887, but with this 

 important difference that it is also used as the exploding 

 vessel, and hence the products of combustion in the form 

 of water are left and must interfere with the accuracy of 

 the readings, since Morley claims that he can read his 

 mercury levels to ^ju of a millimetre and that he does so to 

 ^iy. in a wide tube. His apparatus for the preparation and 

 purification of his gas is also very complicated and volu- 

 minous. His method of testing the residual gas, after the 

 explosion is over, is not at all satisfactory, as it would give 

 it a minimum value, especially if the residue contained hy- 

 drogen or oxygen which had escaped combination in the 

 final explosion, and according to Dixon's experiments there 

 seems to be a small quantity which does miss fire in every 

 such explosion, and this agrees quite with the author's ex- 

 perience, although when the gases are very nearly in their 

 exact combining proportions it becomes inappreciable, yet 

 when there is, for example, a fair excess of oxygen then 

 always a small quantity of hydrogen also remains, and the 

 greater the excess of oxygen the larger, as a rule, also is the 

 quantity of hydrogen in the residue. Morley states that 

 his " explosions were always made in the presence of a 



