90 INHERITANCE IN ANIMALS 



prepared and weighed. It is impossible in the short time 

 available to give you any adequate idea of the work 

 involved in such a weighing. Most of you have probably 

 not tried to weigh anything accurately. If you will try 

 to prepare oxygen so that it is sensibly pure, and then to 

 weigh a bottle of it at known temperature and pressure 

 until your records agree to three figures, you will under- 

 stand some of the reasons why there is a margin of un- 

 certainty to our knowledge about the weight of a known 

 volume of oxygen, and you will understand what it 

 means to have reduced this margin of uncertainty 

 within the narrow limits indicated by Lord Rayleigh's 

 record. 



When he had made this series of weighings, Lord 

 Rayleigh tried to determine the weight of pure nitrogen 

 contained in the same bottle at the same temperature and 

 pressure : and some of the results he obtained are given 

 here : 



Weight of the Nitrogen contained in a certain Flask at 15 C. and 

 760 mm. Barometric pressure. (Rqyleigh.) 



2-30143 grammes 2-31017 grammes 



2-29890 2-31001 



2-29816 2-31024 



2-30182 2-31010 



2-29869 ,, 2-31028 



2-29940 2-31035 



2-29849 2-31026 



2-29889 2-31034 



I have arranged these results in two columns. You 

 notice that the numbers in each column agree with each 

 other very well, though not so remarkably well as those 

 obtained for oxygen : but any number in one column 

 differs from those in the other by about one-hundredth of 

 a gramme; and to a skilful physicist such a difference 



