216 Speyers — Molec u lav Weigh ts of some 



The four were placed in a water bath whose temperature 

 was regulated to 0-4°. 



A current of air direct from the laboratory passed through 

 alcohol tube 1, from whicli it took up alcohol to saturation ; 

 then, to remove all the alcohol, through three spiral tubes 

 containing about 90 cm of sulphuric acid in all. The dried air 

 then passed through the solution in tube 1 and from that into 

 the suction apparatus. 



A second independent current of air, also direct from the 

 laboratory, passed through alcohol tube 2, then through another 

 set of sulphuric acid tubes just like the first set, then through 

 the pure solvent in tube 2 and on into the suction apparatus. 



The loss of the alcohol tubes is proportional to the volume 

 of air drawn through them. Let V be the loss of one alcohol 

 tube, let I" be that of the other alcohol tube, 1/ the corre- 

 sponding loss of tube 1, L" the corresponding loss of tube 2. 

 Then 



gives what would be the loss of tube 1 if as much air passed 

 through it as passed through the tube 2. The loss for that 

 tube whose alcohol tube showed the smallest loss was always 

 corrected. That is, the smallest loss of alcohol was put as the 

 denominator in the correction, the other loss as the numerator, 

 Let Lj and L 2 be the corrected values for tubes 1 and 2. Then 

 we may put L, and L 2 proportional to p' and p respectively in 

 equations 1 and 2. That is, 





«L 2 — aLj L 2 — L x n p—p 

 «L X L } N p 



and 







dL, L 2 _ p __ n 



•«l, l, -y~w 



Since n - 



= w/m and N = W/M we have 





wML ; 





m " wcl-lj 



and 







wM 0-4343 



m = . ^ =^r- 



(3) 



- W(IL-IL,) (4) 



The error of the method could be partly determined by 

 using pure solvent in tubes 1 and 2, correcting the smaller loss 

 as given by the alcohol tubes, and comparing with the 

 observed loss of the other tube. For instance, when pure 

 water was used in both tubes 1 and 2 : 



