F. R. Cowper Reed—Fauna of the Bokkeveld Beds. 165 
monazite in small crystals in granites and gneisses is also of some 
importance, especially now that the radio-activity of rocks is being so 
much discussed. Considerations such as these may have an important 
bearing on the study of petrographic provinces. The possibility of 
using such minerals as supplementary magmatic indicators has not yet 
been considered ; but if ever the eruptive rocks of the earth’s crust 
are mapped out in a series of provinces, it is possible that such 
indicators may be important. These views give strength to the old- 
fashioned belief that natural mineral assemblages are more important 
than artificial norms. It may be, indeed, that future work will 
confirm the American method of making chemical analyses and 
artificial norms the primary basis of classification; but it 1s more 
likely that the working out of natural mineral assemblages on 
a comprehensive scale, and the characterisation of peculiarities in 
the minerals themselves, will be the ultimate goal of systematic 
petrography. 
V.—Tue Fauna oF THE BokKEVELD BEeEps. 
By F. R. Cowper Rezep, M.A., F.G.S. 
T is useful from time to time to take stock of our knowledge in 
connection with such a geological problem as that of the fauna of 
the Bokkeveld Beds, so that we may ascertain to what stage in the 
process of its elucidation we have arrived, and may note the direction 
in which the evidence is pointing. An indication of the principal gaps in 
our knowledge may also be obtained in this manner, and new suggestions 
as to the more promising lines of investigation. The conclusions 
arrived at from such a survey may be employed tentatively as working 
hypotheses, but it may be premature to base upon them confident 
generalisations. Approaching the question of the composition, 
relations, origin, and distribution of the Bokkeveld fauna in this 
spirit of caution and with the foregoing objects in view, we find 
firstly that a considerable advance has been made in our knowledge 
of it and allied faunas during the last few years. The researches of 
Ulrich, Kayser, Clarke, Katzer, Thomas, and others in South America 
have given us an immense amount of fresh information on the Devonian 
faunas of that continent, and the South African fossils themselves 
have been studied by Lake, Schwarz, and the author. Their general 
poor state of preservation is a matter of regret, but the increase in the 
material available has cleared away some of the difficulties met with 
by Salter and other previous workers. Much undoubtedly remains to 
be learnt from work in the field, and we may expect further yields of 
new species from different localities in the immense area still to be 
explored. 
‘The references given at the end of this paper do not aim at being 
a complete list of all recent literature dealing with the subject of the 
Devonian faunas of South Africa and South America, for only those 
works are quoted which appear to the writer to be of special interest 
or importance in connection with the Bokkeveld fossils. 
