382 PROF. P. MARSHALL OX THE SEQUENCE OF [Dec. 1914, 



18. The Sequence of Latas at the North Head, Otago 

 Harbour, Dukedix (New Zealand). By Patrick Mae- 

 shall, M.A., D.Sc, F.G-.S., Professor of Geology in the 

 University of Otago. (Bead May 27th, 1914.) 



[Plates LIII & LIV.] 



Contents. Page 



I. General Geology 382 



II. Petrography 388 



III. Chemical Composition 393 



IV. Classification of the Rocks 396 



V. Origin of the Different Lavas 398 



VI. Summary and Conclusions 405 



I. GrEXEEAL GrEOLOGT. 



The North Head of Otago Harbour is situated in lat. 45° 47' 30" S. 

 and in 170° 45' long. E. It is 13 miles distant from the city of 

 Dunedin, in a north-easterly direction. The head is a precipitous 

 cliff facing nearly due east in its southern portion ; but, bending 

 slightly westwards in its northern portion, it runs from south 

 15° east to north 15° west, and therefore in this portion faces 15° 

 to the north of east. 



The cliff has an altitude of 530 feet at its highest point almost 

 in the middle of the section, and throughout its length it is more 

 than 300 feet high, except at the southern end, where it slopes 

 somewhat gradually to the sand-hills which skirt its base. The 

 lower part of the southern end of the cliff is partly obscured by 

 scree-slopes and blown sand, and for 200 yards from its northern 

 extrenrhy" the lower part of the cliff is hidden beneath an immense 

 slip. . Elsewhere, for a length of a mile and a half, the cliff presents 

 a surface that is almost perpendicular, and it thus affords a remark- 

 ably clear section nearly bare of vegetation and quite free from 

 all those difficulties of interpretation that are always associated 

 with folding and faulting. This cliff was briefly mentioned in a 

 previous paper of mine, published in the Quarterly Journal of this 

 Society in 1906 (vol. lxii, p. 417). In this description the general 

 nature of the succession was indicated. The lava-flows exposed in 

 the clear section include a trachyte, three trachytoid phonolites, 

 one kaiwekite, two trachydolerites, and numerous basalts. It is, 

 therefore, evident that a complete examination of the different 

 lavas might be expected to throw a considerable and perhaps 

 important light upon the causes which promote those differences 

 that are sometimes found in the succession of lavas that issue from 

 a single volcano, or at least in the same volcanic district. 



In the first place, evidence was sought with the object of ascer- 

 taining whether the various lavas which are exposed in the cliff 



