IXXX PROCEEDINGS OF THE GEULOGTCAL SOCIETY, [vol. Ixxvii, 



SO for the v.orkl at large. We must also remember that these two 

 classes are not each all of one kind and wholly distinct from the 

 other; it may he, and indeed almost certainly is, the case that 

 botli nonnal and reversed faults comprise more than one group 

 wholly distinct in origin and mode of formation, and that in some 

 cases there is a closer relationship between normal and reversed 

 faults than l)etween them and others of nominally tlie same class. 



There is, hoAvever, one very definite difference between normal 

 iind reversed faults, in that the former necessitate an increase in 

 the horizontal distance of two points situated on opposite sides 

 of the fault and the latter a decrease. In other words, normal 

 faulting indicates an extension of the country affected by it. and 

 reversed faulting a compression. This distinction has long been 

 recognized, and more than half a century has passed since there 

 appeared a paper by the Rev. J. M. Wilson ^ — whom I revere as 

 my rirst teacher in geology — on the cause of contortion and faults. 

 In this it was pointed out that the elevation of a tract of country 

 would increase the length of the measurement across the elevated 

 tract, while depression would give rise to a decrease, and calculation 

 was made of the amount of the extension or compression which 

 would be produced in this way. There can be no question that 

 the cause assigned is a true one and that both extension and com- 

 pression can be produced in this way; but, in the light of our present 

 knowledge of the extent of contortion and faultino- it is evident 

 that the cause is quantitatively inadequate, and that some other 

 must be invoked to account for the amount of change in the original 

 dimensions which is indicated by existing structure. This is not 

 a matter to be dealt with here : but it may be noticed that, although 

 extension of the horizontal dimensions of a fanltcd region would 

 sufficiently account for the facts of normal faulting, provided that 

 the necessary fractures were in existence, the case of reversed 

 faults is much less simple, for, while they imply a reduction in 

 the horizontal dimensions of the faulted region, it is easily 

 demonstrable that, in many instances, they could not have been 

 produced merelv by compression in a horizontal direction. It is 

 true that reversed faulting has been imitated in experiments on 

 a small scale, and produced in those instances by compression ; it 

 is equally true that on the scale met with in Xature they might 



^ ' On the Cause of Contortion & Faults ' Geol. Mag. 1868, pp. 205-208. 



