422 PROF. P. MARSHALL ON THE [Aug. I906, 



The Papanui Dolerite has a wide occurrence in Otago, and is 

 extremely uniform in structure and composition. The trachyte and 

 alkaline trachytoid phonolites are also very closely related, very 

 uniform, and are of widespread occurrence in the Dunedin area. 



The obvious criticism that must be first offered concerning this 

 explanation, is the complete mixing that the magmas have under- 

 gone. Both the basanites and the trachydolerites appear to be 

 uniform and homogeneous throughout. The passage of steam- 

 bubbles through the magma, and the flow-movements consequent on 

 eruption, seem to me inadequate to account for so complete a 

 mixing. Yet, admitting this apparent inadequacy, the explanation 

 is perhaps more satisfactory than any that might be based upon 

 magmatic differentiation. 



Y. Summary. 



The Dunedin district is a distinct petrographical unit. The 

 igneous rocks are of post-Oligocene age. 



The igneous rocks are very variable in nature, embracing plutonic, 

 hypabyssal, and volcanic members. They range in their petro- 

 graphical character from very basic dolerites to highly-acid and 

 alkaline trachytes. 



Complete descriptions are given of many typical members of the 

 several groups. 



The geological occurrence of the rocks is described. 



Chemical analyses show that the petrographical differences are 

 often coincident with chemical differences, and also that they are 

 generally similar to rocks of the same classes in other parts of 

 the world, more particularly to those of East Africa. 



A classification is attempted, according to the methods proposed 

 by distinguished American petrologists. 



Sections are described that indicate the order of succession of 

 the lavas. This order does not appear to be explicable on any 

 theory of magmatic differentiation, but it is suggested that it may be 

 due to the mixing of magmas before and during eruption. 



Postscript. 



[Prof. Eosenbusch, with characteristic enthusiasm, has kindly 

 examined slides of many of the rocks described in the foregoing 

 pages, and his remarks on them have just reached me. In nearly 

 all respects he agrees with the classification that I have adopted, but 

 there are some important points as to which his wide experience 

 and knowledge have prompted him to give me important infor- 

 mation : — 



The diorite of Bell Hill he classes as an exceptional form of essexite. 



The andesite of the Paper-Mills and the trachytoid phonolite of 

 St. Leonard's are, he says, rather more closely allied to the trachy- 

 dolerites than to the classes in which I have placed them. The 

 former he regards as closely similar to the ' andesite ' of the Sieben- 

 gebirge. 



The camptonite also he refers to the trachydolerite-group. 



