TRANSACTIONS OF SECTION C. 



661 



superior quality of sucli iron has been thought by some metallurgists to be due to 

 the presence of titanium in it. 



Even a very small percentage of nickel in an iron ore would be of value if the 

 nickel could be extracted along with the iron in smelting, as the resulting alloy 

 might be used directly in the production of nickel-steel. 



There is reason to believe that magnetites will be found containing a higher 

 percentage of nickel than those already examined, just as some of the Canadian 

 pyrrhotites, which are also considered to be of igneous origin, contain amounts of 

 nickel which make them valuable as ores, while others contain the metal in lesser 

 amounts. 



5. Differentiation in Igneous Magmas as a result of Progressive 

 Crystallisation. By J. J. H. Teall, M.A^ F.R.S. 



Crystal building in an originally homogeneous igneous magma necessarily 

 produces differentiation into portions of different chemical composition, a fact the 

 importance of which was first impressed upon the author sixteen years ago in study- 

 ing the andesitic lavas and their associated quartz porphyry dykes in the Cheviot 

 district. 



As is well known. Professor Eosenbusch has classified the common constituents 

 of igneous rocks into (1) the ores and accessory constituents (including magnetite, 

 &c.), (2) the ferro-magnesian constituents, (3) the felspathic constituents, (4) free 

 silica, and has maintained that members of group (1) are the first to form in the 

 process of crystallisation, and that while there are irregularities of order between 

 members of group (2) as compared with those of group (3), yet the members of 

 these groups separate out inter se in the order of increasing acidity. This order of 

 crystallisations has been emphasised by many writers, though it has also been 

 clearly recognised that the law is not constant in different magmas and under 

 diflerent conditions. The object of the present communication is to call attention 

 to what is at least an important exception to this law. 



Among an extensive series of rocks and fossils collected by the Jackson- 

 Harmsworth expedition in Franz Josef Land, recently examined by the author and 

 Mr. E. T. Newton, are many basalts essentially composed of labradorite, augite, and 

 interstitial matter, in which labradorite formed first, then augite, and last of all 

 the interstitial matter either with or without further differentiation. The main 

 interest of these rocks lies in the composition and relations of the interstitial 

 matter. This is occasionally present as a deep brown glass, but more often is 

 represented either by palagonite or by a turbid and more or less doubly refracting 

 substance crowded with skeleton-crystals of magnetite. In many specimens it is 

 only in this form that magnetite occurs, the labradorite and augite being free from 

 inclusions of this mineral. These facts prove that magnetite may belong to a very 

 late stage of consolidation, and that progressive crystallisation may lead to a con- 

 centration of iron oxides in the mother liquor. 



The palagonite has undoubtedly been formed by the hydration of a deep brown 

 glass. An analysis was made of it with the following results : — 



Silica . 



Titanic acid . 



Alumina 



Ferric oxide . 



Ferrous oxide 



Lime 



Magnesia 



Soda 



Potash . 



Loss on ignition 



98-54 



100-00 



