part 2] or souTHEEisr eyee peninsula. 115 



into the supercrust, which cooled and consohdated prior to the 

 irruption of the more acid and so more latel}^ evolved granite- 

 gneisses. With the intrusion of the latter into the gabbroid rocks 

 already crj^stallized, these basic rocks would be metamorphosed, 

 and under a stoping action would be incorporated within the flowing 

 gneiss. This metamorphism, as might be expected under such 

 conditions, has not been of the pure thermal type, but one in which 

 shearing stress has also played a part. In this way only is the 

 unidirectional foliation of the amphibolites to be explained, and 

 this metamorphism was therefore accomplished prior to the rifting- 

 oft of the amphibolite-bands, for with the subsequent engulfment, 

 the inclusion is subject to pure thermal metamorphism under 

 hydrostatic pressure. 



So far we have considered that large proportion of amphibolites 

 which show indubitable evidence of intrusion by the gneisses. It 

 is, of course, not possible to prove intrusive relationships in every 

 basic amphibolitic band. In several the junction with the gneiss 

 cannot be seen, owing to some accidental circumstance. Again, it 

 is noteworth}^ that some of the massive tj^pes show no visible 

 injection by the gneiss, and the border-zone may be quite sharp, or 

 the parallelism of the elongated bands with the foliation (which is 

 very generall}^ observed) may be somewhat departed from. From 

 the point of view ©f inclusions showing these characters, it may well 

 be that a more massive type of inclusion shall resist pegmatitic 

 invasion to a greater degree than one already foliated ; and, so far 

 as parallelism of gneissic foliation and elongation is concerned, a 

 departure may indicate simply that complete orientation had not 

 been effected. 



The point to be remarked is, that absence of evidence of invasion, 

 a not strictlj^-parallel arrangement of inclusion and foliation, and 

 the presence of a sharp, Avell-defined border-contact, such features 

 being shown by a number of amphibolites, cannot be necessarily 

 taken as evidence of a posterior dyke-origin. This is more especially 

 the case where the large majority of amphibolites give evidence of 

 their anterior origin by those proj)erties which have already been 

 discussed. It would be idle to deny, hoAvever, in a region of such 

 complexity as this, the absence of posterior dyke-origin. 



With regard to these particular bands in the gneiss, the question 

 of an anterior or posterior origin must of necessity at present be 

 left an open question, for the field-evidence adduced is capable of 

 a dual explanation ; nor does the microscopic evidence shed further 

 light on the origin of such a band, which may be identical in 

 composition and texture with a band in wdiich all the requirements 

 for an anterior origin are fulfilled. 



A posterior dyke-origin is rendered the more probable, when the 

 trend of the band is violently opposed to the flow-structure, and 

 when the contact is definitely an intrusive one as seen by branching 

 veins or other similar structures entering the surrounding gneiss. 

 There are a number of bands in the gneiss of Boston Island which 

 seem to fulfil these conditions ; but in no other locality have there 



