C. E. Tilley—Para-Gneisses wn South Australia. 805 
the two species mentioned (the identification of which is not definite) 
-may belong to the uppermost Lower Trias. In the nodule bed 
a doubtful Flemingites has been found, and the Anasibirites and 
Keyserlingites are characterized by simple sutures, so that horizon 
C, may very probably be somewhere near the border-line between 
the Flemingites beds below and the Stephanites zone above.t The 
latter probably includes the lower half or more of the Posidonomya 
shales, but an exact correlation is impossible in the present state 
of our knowledge. 
(To be continued.) 
Precambrian Para-Gneisses of Southern Eyre 
Peninsula, South Australia. 
By C. E. Tinney, Emmanuel College, Cambridge. 
(Concluded from p. 259.) 
4. ORIGIN OF THE GARNET GNEISSES. 
HE composition of the garnet gneisses already described allows 
them to be treated collectively in a discussion of their origin. It 
will be clear from the foregoing account that the mineralogical 
composition and the textures developed are those characteristic 
of the highest grades of metamorphism. They are, therefore, to be 
included amongst the Kata-zone gneisses of Grubenmann. The 
pronounced banding in these rocks must be interpreted as a result 
of initial differences in composition of the original sediments. This 
banding therefore represents the bedding and has been preserved 
through the various stages of metamorphism, eloquent testimony 
that the amplitude of diffusion in metamorphism is in general 
of quite limited dimensions. 
far too lowin the Lower Trias. On the other hand, the Sibirian Olenek Beds, 
probably, like the Stephanites zone of the Salt Range and the Himalayan 
Hedenstremia Beds, can be correlated with the Spitsbergen Posidonomya Shales 
or at least their lower part (see also Diener, ‘“‘ Das Alter der Olenekschichten 
Sibiriens’’: Centralbl. f. Min, ete., 1908, p. 233). 
1 The Lower Triassic fauna of Kéira in Albania, that Diener (Kashmir, 
1913, p. 121) considers to be homotaxia! with the Hedenstremia stage of India, 
and that, at least in part, corresponds with the Columbites fauna of Idako, 
shows a striking resemblance to the apparently contemporaneous fauna here 
described, but contains many higher (and later ?) types. Arthaber’s treatment 
of this fauna, as his whole classification, is not very fortunate (see L. F. Spath, 
“Notes on Ammonites’’: GrotoaicaL Macazine, Vol. LVII, 1919, p. 224); 
but the affinities of e.g. some of the forms that Arthaber includes in his 
heterogeneous Dagnoceras (Arctoceratine!) with certain Spitsbergen 
Ammonites here referred to Anasibirites, Goniodiscus, Prionites, etc., are most 
interesting. They show that, as mentioned above, the ‘‘ Meekoceras’’ and 
“ Xenodiscus” of the Stephanites zone probably are only heterochronous 
homceomorphs of the lower, true Meekoceras and Xenodiscus. In the case of 
the latter, the forms of the Ophiceras layer may have to be separated generically 
both from the Permian type and from the Upper Eotriassic forms, that is to 
say, there are at least three independent developments that successively take 
on the “‘ Xenodiscus’’ characters. 
VOL. LYIII.—NO. VII. ; 4 20 
