CHEMICAL EFFECTS OF FUSION 437 



any doubt that the samples were representative to an unexpected degree. 

 Our results, therefore, furnish a very trustworthy basis for a discussion 

 of the chemical effects of fusion. 



In order to compare the analyses, Ave have reduced them to a common 

 silica basis by recalculating the analysis of the fused portion. Columns 

 4 and 5, Table I, therefore state the amounts of each oxide associated 

 with 61.76 grams SiOg in both fused and unfused. Columns 6 and 7 

 show dilferences and columns 8 and 9 express these as percentages of the 

 amounts present in the original unfused sediment. In these last two 

 columns figures in brackets represent original differences between the 

 fused and unfused material. Eor constituents present in accurately de- 

 terminable amounts the differences range from 2 per cent to 8 per cent. 

 Figures not in brackets represent the effects of fusion, and range from 

 57 per cent to 137 per cent. The differences in KgO, TiOg, P2O5, MnO, 

 .and possibly ALOg are within the limits of analytical error. Those in 

 MgO, CaO, and perhaps AUOg represent actual differences between the 

 two samples. A calculation from these two columns shows that the iron 

 pipe has contributed 1.78 grams of iron (or 2.29 grams stated as FeO) 

 to 100 grams of the original material, on fusion. In the oxidation of 

 this additional iron to FeO and in the concomitant reduction of the 

 original amount of FegOg to FeO, there has been a change in oxygen 

 ■content of -|-0.35 gram. This oxygen may be assumed to have come 

 from the decomposition of water by metallic iron at the elevated tem- 

 perature. 



All the CO2, practically all the water, and about one-half of the sul- 

 phur have been driven off during the fusion. Only 0.26 per cent of 

 water has actually remained in the fused product — a fact to which refer- 

 ence will be made later. The remaining percentage differences are to be 

 attributed to actual variation, in reality very slight, in the nature of the 

 material in the sedimentary column penetrated. 



Temperature of Fusion" 



We have made tests on the material of the core to determine the tem- 

 perature required for fusion. When heated in air the rock powder is 

 ■changed to a l^rick-red frit in which the iron has evidently been largely 

 oxidized to the ferric state, and tests carried out under these conditions 

 would give little indication of the temperature of fusion under the con- 

 ditions of drilling. The tests were therefore made in a current of CO^ 

 .and steam after a method devised by E. S. Shepherd, of this Laboratory, 

 who has found that in such an atmosphere the state of oxidation of the 



