262 H. S. Summers: 



The early geological survey of Victoria subdivided these basalt 

 rocks into older and newer Volcanic. In some cases where the lavas 

 come into relation with the Kainozoic marine series the age can be 

 fairly definitely fiyed, but in a large number of cases field evidence 

 as to exact age is wanting. Professor Skeats (15) has discussed the 

 relative ages of these basalts and has pointed out that " the reference 

 of a basalt to the older or newer series is frequently based only on 

 conjecture or analogy." 



The basalts of the plains to the north of Melbourne, however, 

 clearly belong to the younger series, as they overlie the Kainozoic 

 marine series at Essendon, Keilor and elsewhere. 



The plains are continuous up to the slopes of Mount Macedon, 

 and there is not the slightest doubt that the basalts of the plains 

 near Macedon belong to the newer series. 



Somewhat older than these basalts are the interesting series of 

 alkaline rocks which flank the slopes of Mount Macedon. 



The alkali rocks which have been recognised in the Macedon 

 District consist of solvsbergite, anorthoclase trachyte, anorthoclase 

 olivine trachyte, olivine anorthoclase trachyte, limburgite, mace- 

 donite, woodendite, and anorthoclase basalt. For reason given in 

 the paper on the Macedon District, Professor Skeats and the author 

 consider the probable order of extrusion was : — 



1. Anorthoclase trachyte. 



2. Solvsbergite. 



•?. Anorthoclase basalt, Macedonite, Woodendite. 



4. Anorthoclase olivine trachyte, Olivine anorthoclase trachyte. 



5. Limburgite. 



The solvsbergites are in the form of plugs at the Camel's Hump, 

 Hanging Rock and Brock's Monument, while the remaining types 

 occur as more or less widely spread lava flows. The age of these 

 rocks is considered to be probably mid-Kainozoic. 



III. Differentiation of the Rocks. 



One of the most interesting problems to the petrologist is that of 

 the causes which give rise to the great diversity of types of igneous 

 rocks. As has been stated earlier, the aim in the present paper is 

 1 athc'i- to shoAv the relationships of the end products than to attempt 

 an c'Xj)lanation of the causes which have led to the production of 

 tlie various types belonging to any one scries. 



Harker (4) has shoAvn that the serial j-elationship of the rocks of 

 any petrographic province may be expressed by moans of variaticm 

 diagrams consti-uctcd with the silica percentages of the several 



