234 Correspondence — F. Dixey. 



Further, I have referred to other cases in which the name of a 

 characteristic constituent has been applied to a whole group of 

 rocks ; in these cases the name has been allowed to stand even when 

 the special constituent is inappreciably developed in the extreme 

 acid or basic members, or in unusual types. For instance, the 

 alkaline and calc-alkaline suites are well-known to coincide at their 

 acid and basic ends, and, moreover, intermediate members of the 

 one suite can sometimes be distinguished from those of the other 

 only by their close association with a characteristic type. 



As regards the suggested list of magnesian provinces, I have 

 admitted the possibility that further work may demand some 

 modification, and it is certainly not to be expected that the areas 

 named would be analogous in every respect. Nevertheless, at the 

 time I drew up the list there appeared ample justification for the 

 inclusion of all the areas named, but I thought it unnecessary, in 

 a short general paper, to do more than refer to original papers. 

 In my present circumstances, having no opportunity of consulting 

 the necessary works again, I am unable, even if it were desirable, 

 to put forward a detailed comparison of the various areas. 



It is well known that the presence of hypersthene does not 

 necessarily indicate a high proportion of magnesium in a rock ; 

 this also has been mentioned ia my paper. 



In conclusion, I may perhaps be permitted to suggest that it is 

 still rather early to base arguments concerning the crystallization of 

 deep-seated magmas upon chemical equations. Moreover, as regards 

 the contact metamorphism of norites, the chemical reactions 

 suggested by Mr. Tilley do not appear, in the case of various 

 occurrences that have been examined in detail, to be in accord with 

 the results of petrological examination. I believe I am correct in 

 stating that recent additional work upon the Huntly intrusions,^ 

 referred to by Mr. Tilley, has indicated that the cordierite of these 

 rocks is due entirely to absorption of sedimentary material, and not, 

 even in part, to invasion of one intrusion of norite by another. 

 Again, in the case of the norite complex of the Sierra Leone 

 Peninsula, where several phases of metamorphism have been 

 developed as a result of the invasion of earlier members of the 

 complex by later, the actual changes observed are also very different 

 from those suggested, especially in that no trace of cordierite has 

 been found. 



F. Dixey. 



Nyasaland. 



January, 1922. 



[The foregoing letter was unavoidably crowded out of the April 

 number ; however, a proof was submitted to Mr. Tilley, whose reply 

 is printed below. — Editor.] 



^ Cf. H. H. Read on " Some Contaminated Eock Magmas", paper read 

 before Section C (Geology) at the Edinburgh meeting of the British Association 

 for the Advancement of Science, September, 1921. 



