354 Morris Morris: 



Taking KicliarcLs' analysis (16) of tlie Hypersthene Dacite at 

 Upwey, in addition to the above four, we are thus provided with 

 a complete set, representing each of the five volcanic series. The 

 variation diagram has been prepared from these analyses, and I 

 have also introduced analyses of the following closely-related dacitic 

 rocks of the Macedon district, described by Skeats and Summers 

 (15), viz : — 



M^, Dacite, Braemai' House. 



M2, Granodiorite, Braemar House. 



M3, Dacite, Hesket. 



M"*, Grandodiorite, Baringo. 



The petrological affinities of the Macedon and the Dandenong- 

 districts are indicated by the conformity of the analyses of the 

 former with the curves of those of the latter, as far as the three 

 Dacite Series (c. d and t) are concerned. These all form a serial 

 group. The two toscanites {a and 6), however, fall together on 

 curves of their own. It is. therefore, obvious that either they have 

 pursued an independent line of differentiation in the same cycle, or 

 else, that they belong to some period older than the dacites. 



In the absence of other evidence, I have preferred the former view 

 for the following reasons: — (1) The distribution of a and h is such 

 as to keep them everywhere in the closest association with the three- 

 other series. (2) The chemical analyses show family characteristics 

 between the two groups, e.g.. they both have a very high percentage 

 of BaO, indicating probably a barium felspar throughout the four 

 series; further, the series all fall in or near the sub-alkali division. 



Order of extrusion. — The variation diagram shows that a and h 

 form an inseparable pair. For reasons set forth earlier, a is 

 regarded as being older than h; c has been found superposed on b. 

 This gives us the order, «, 6, c. For the reasons given below, e 

 must be regarded as the last, and d, the penultimate, series of the- 

 cycle. Therefore, the order must be a, h. r, d. e. and this is the- 

 order which is everywhere indicated by the evidence of super- 

 position. 



Magmatic different iatiou . — There aj)pears to lie little d()ul)t that 

 these five series, excepting possil)ly the toscanites, are consan- 

 guineous. They are, therefore, regarded as the differentiation pro- 

 ducts of a primordial magma, which there are the following 

 reasons for lielieving to have had a chemical composition, similar 

 to that of the Upper or Eypersthene Dacite Series (p), and the 

 Granodiorite, and the Wandin Dyke : — 



1. The above three rocks represent respectively tlie volcanic,. 



