106 it. B. Riggs — Separation of Magnesium Chloride 



Weight of 

 KC1 taken. 



Weight of 

 NaCl taken. 



Weight of 



chlorides 



found. 



Corr. weight 



of chlorides 



found. 



Error in corr. 



weight of 



chlorides 



found. 



Weight of 



MgO found 



in chlorides. 



grm. 



grm. 



grm. 



grm. 



grm. 



grm. 



(15) 0-0544 



0-0529 



0-1066 



0-1076 



0-0003 + 



trace 



(16) 0-0545 



0-0531 



0-1068 



0-1077 



o-oooi + 



0-0004 



(17) 0-1087 



0-1063 



0-2131 



0-2142 



0-0008 — 



0-0002 



(18) 0-1085 



0-1063 



0-2i22 



0-2134 



0-0014 — 



trace 



(19) 0-1083 



0-1064 



0-2118 



0-2137 



o-ooio 



not det. 



Weight of 



MgO takeD. 



Weight of Corr. weight of Corr. weight of 

 MgS0 4 found. MgS0 4 found. MgO found. 



Error in corr. 

 weight of MgO 

 found. 



grm. 



grm. 



O' 



6 



rm. 



grm. 



grm. 



(15) 0-0827 



0-2487 0-2475 



0-0825 



0-0002 — 



(16) 0-08C0 



0-249 



5 0-2484 



0-0828 



0-0002 — 



(17) 0-1664 



0-5000 0-4987 



0-1662 



0-0002 — 



(18) 0-1663 



0-500 



7 0-4995 



0-1665 



0-0002 + 



(19) 0-1660 



0-5010 0-4987 



0-1662 



0-0002 + 



These results speak for themselves. It is evident that we 

 have here a fairly good quantitative method. The errors are 

 generally negative. This is to be expected. The large nega- 

 tive errors in the weights of the insoluble chlorides, in ex- 

 periments (26)-(28), are probably due to loss from manipulation. 

 "2 grams of either the insoluble chlorides or magnesia are 

 perhaps too large quantities to deal with. 



10 cc of amyl alcohol hold 1'14 grm. of magnesium chloride 

 in solution at 14° C. and 1-23 grm. at 50° C. In the majority 

 of the experiments the residual volumes of alcohol were less 

 than 10 cc . They might with safety have been further reduced, 

 thus reducing the solubility corrections. 



In experiments (7), (8),' (13), (14), (18) and (19) two precipi- 

 tations of the insoluble chlorides were made, decanting the 

 liquid, washing the residue, redissolving it and repeating the 

 precipitation. The advisability of this is questionable. A 

 more complete separation may be thus effected but the possi- 

 bilities of loss from manipulation seem to counterbalance this 

 advantage. Almost the same degree of completeness of sepa- 

 ration may be attained by redissolving the precipitate with as 

 little water and acid as possible and repeating the precipitation 

 in the original solution. 



In dehydrating the mixed chlorides it was noticed that the 

 nature of the precipitate depended much on the relative 

 amounts of the magnesium and the insoluble chlorides. 

 Roughly speaking, when the weight of magnesia is equal to or 

 exceeds that of the insoluble chlorides, the precipitate is 

 thrown out as a fine powder. When however the insoluble 



