56 NEW YORK STATE MUSEUM 



„ I 2 3 " 4 S 



YJ" 4-45 723 5.00 3.34 s-oo 



Ab 23.06 \ ^-. „^ 26.20', ^. ,_ 31.-^7 ■ ^, fi_ 25.15 ,a .« 20.96'' , 



An 33.64; 56-0 18. 07J '*■'•'' 13.90; ^5.87 13. 3^; 38.49 16. 4of 37.36 



Kaolin.... 3.10 3.43 5.42 4.65 3.60 



Calcite . . . . .20 .80 .10 .50 .90 



CaO.SiOo.. 2.67I 6.96] iLas'i 3.48] 12.531 



.Mg°dlfoV.: |:to:;-3-4' i::^ ■'•8^ •■;„>■■•'« ;;JS^« ,?;l^•3..,x 



MnO.SiOa.. .26, .40. .53 j .39 J > .26 I 



2FeO.Si02. 8.461 ,^ „, 8.36' , 5, io.2o\ ,„ . 13.67 1.73' 



sMgO.SioV 8. 47 j '^-93 5.46/^3.82 8.96/19.16 1^.80/30-47 ,llf 2.93 



Magnetite.. 2.09 2.32 3.25 3.25 6.73 



Ilmenite... 2.13 6.23 4.62 2.16 9-73 



Apatite 34 1.34 .70 .35 .67 



Pyrrhotite. .18 .35 .26 .18 .65 



Garnet 7.20 



Spinel 3.13 2.13 



123 45 



Plagioclase AbiAn.3 AbiAnj.3 AbiAnos AbjAn, AbjAni-^ 



Light colored minerals 64.45 55.73 56.39 46.98 46.86 



Dark colored minerals 35 -08 43-95 43.10 52.40 53-42 



1 "VTall rock of titaniferous magnetite. Split Rock mine, "Westpcrt. Analysis by W. F. 

 HUlebrand. 



2 Woolen mill i nule west of Elizabethtown. Analysis by W. F. HiUebrand, See 

 above p. 40. 



3 Gneissoid gabbro, 2 miles south of Elizabsthtown. Analysis by W. F. Hiilebrand. 



4 Massive gabbro. same exposure as no. 3. Analysis by \\ . F. HUlebrand. 



5 Wall rock of titaniferous magnetite. Lincoln pond. Analysis by George Steiger. 



In the quantitative system, no. i is class II Dosalane; order 5 

 Germanare ; rang 4 Docalcic, Hessase ; subrang 3 Persodic, Hessose. 



Nos. 2, 3, 4 and 5 are all class III Salfemane; order 5 Gallare; 

 rang 4 Docalcic, Auvergnase ; subrang 3 Auvergnose. 



In recasting the above analyses, nos. i, 2 and 5 could be done in 

 the normal way. That is, aside from the accessories such as mag- 

 netite, apatite, pyrrhotite, ilmenite and calcite about which there can 

 be little doubt, the soda and potash were assigned to albite and 

 orthoclase ; the combined water to kaolin, and the remaining alumina 

 used for anorthite. There then remained sufficient silica to care for 

 the excess of lime as the bisilicate and for the ferrous iron and 

 magnesia partly as unisilicates, partly as bisilicates. In nos. 3 and 4 

 this proved impossible, because if this course is followed for 

 anorthite there is not enough silica to satisfy the remaining bases 

 even as unisilicates which we know are not the sole dark silicates 

 present. To obtain sufficient silica, the only feasible course was to 

 reduce the anorthite, and for no. 3 a plagioclase molecule 

 AbiAn^., was assumed. It became possible then to reach a solution. 

 In no. 4 similarly Ab^Anj^ was assumed, and both the garnet and 

 spinel molecules were called in. These assumptions have no par- 

 ticular advantage over the ordinary^ calculations of the highly 

 ingenious quantitative system, except that we confine ourselves to 

 making assumptions of minerals known to exist in the rock. 

 Probably ever\- one of these rocks had some garnet, to whose 

 substance both anorthite and bisilicates contributed. Possibly 



