SECTIONAL TRANSACTIONS.— C. 315 



whose exact source of origin is unknown, but which may have been derived from the 

 floor of the Baltic Sea. 



In North America several elements of this fauna are common in the highest beds 

 of the Stonehouse formation at Arisaig, Nova Scotia. 



Afternoon. 

 Excursion to Portishead and Clevedon. 



Friday, September 5, 



Discussion on The Validity of the Permian as a System. (Ordered by the 

 General Committee to be printed in extended form.) 



Dr. Bernard Smith. — The validity of the Permian as a system, or as to what its 

 limits should be, even if accepted as a system, has already given rise to much dis- 

 cussion and bids fair to cause more controversy and expenditure of energy than that 

 attaching to the limits of the Cambrian and Silurian Systems in the days of Murchison 

 and Sedgwick. 



The question appears to depend largely upon two points : — 



1. The definition of a system. 



2. The determination of the vertical range of the (so-called Permian) rocks that 



naturally fall into such a system. 



Geikie has laid it down that a system or period should contain a number of Series, 

 Formations or Epochs ; a Series a number of Groups or Stages ; and a Group, two or 

 more Beds which in turn may be split into Zones. Again, a System should show, in 

 general, a fauna and a flora that is distinct from that above or below and characteristic 

 in certain striking cases. 



If the stratigraphical record, or even large parts of it, were perfect, and represented 

 by a series of rocks deposited (in the sea or even in land-locked continental areas) in 

 uninterrupted sequence, geological history would show a blending and transition from 

 period to period, and no sharp division could be made. But the record has been 

 constantly interrupted, and these interruptions have served two purposes : — 



(1) Firstly, the geographical and topographical changes introduced by widespread 

 earth movements have either given a spurt to or put a check upon evolutionary 

 processes — affecting in the course of time even those areas where sedimentation was 

 continuous. 



(2) Secondly, have caused in certain areas a suppression — by denudation or non- 

 deposition — of parts of the record. 



It is interruptions of the second type that have been seized upon most eagerly by 

 stratigraphers in the past for drav/ing their boundaries between one formation, or 

 one system, and another, and for arranging their history into chapters. 



This works very well locally, but since Systems have been named, and limits have 

 often been given from localities in which a series of rocks has been first studied in 

 detail and presumed to be typical but which may be curtailed by a considerable 

 hiatus at the base, and even at the top also, the geologist begins to find difficulties 

 when he tries to push his systems farther afield, and has to take, in part or whole, 

 of what may be termed the transition or ' liaison ' deposits between his system and 

 the one above or below in cases where the succession has been virtuallj- unbroken. 

 He then finds that he must limit his system by faunas and floras and not by 

 unconformities. We have well-known instances of this in our attempts to separate 

 Cambrian from Ordovician, Silurian from Devonian, Cretaceous from Eocene, and so 

 on in different parts of this country and of the world. 



The question before us to-day is due to similar causes. In some parts of the 

 world diastrophic movements have caused the suppression of basal beds which if 

 deposited would have been claimed by certain geologists as Permian ; in other districts 

 either there is or it is claimed there is a stratigraphical break between Permian and 

 Trias. On the other hand, there is in some districts an apparent perfect passage up 

 from the Carboniferous, or up into the Trias. 



A much thicker belt of rocks is now claimed by certain geologists as Permian than 

 Murchison and his colleagues ascribed to that formation when it was first established. 



