M. 1. Goldman — Catahoula Sandstone of Texas. 279 



blown away. The flaky micas are of course still more readily 

 carried off by the winds. This elimination of ferromagnesian 

 minerals and micas is the additional factor, alluded to under 

 the previous heading, to account in part for the low proportion 

 of heavy minerals in general. Of course it can not be 

 definitely asserted that these minerals have actually been 

 eliminated in the process of formation of the sandstone since 

 they may not have been present in the source rocks from 

 which the grains of the sandstone were derived. Nor, revers- 

 ing the process, is it, I think, safe to try to reconstruct the 

 predominant source rock from those minerals that are present 

 in the sandstone. Thus the meaning of the predominance of 

 acid varieties among the feldspars and of the relative scarcity 

 of microcline can not be determined without knowledge of the 

 relation, under various geographic conditions, of their propor- 

 tion in particular sediments to their proportion in the specific 

 source rock or rocks from which the material of those sedi- 

 ments was derived. Moreover there is room for unlimited 

 confusion through the mixing of material from different 

 sources. In the case of the Catahoula sandstone there is per- 

 haps less probability of such mixing since the proportion of 

 feldspars present, as explained above, suggests the derivation 

 of the material from near the place where it was deposited, 

 but even here there is again the danger of reasoning in a 

 circle since the high proportion of feldspar may be due to 

 derivation from a very feldspathic rock at a greater distance. 

 Still in an attempt to draw the most probable conclusion we 

 have as a basis of deduction the order of weathering given .by 

 Mackie, and Thoulet's tables for the relative amounts of 

 orthoclase and plagioclase in the Gulf of Lyon. While the 

 wide variation in the proportion of feldspars in a region so 

 limited as that studied by Thoulet shows how local this feature 

 is, yet at the same time this very local variability under con- 

 ditions which must be regarded as uniform in all essentials 

 suggests a close dependence upon source rock. Moreover the 

 predominance, in several of the samples, of plagioclase over 

 orthoclase shows that the plagioclase does not weather so 

 much more readily as to destroy in all cases its original pre- 

 dominance, so that the high ratio of orthoclase to plagioclase 

 in the Catahoula sandstone may be assumed to be largely 

 original in the rocks from which the material was derived. 

 This is made still more probable by the relatively small amount 

 of microcline, for if this is the most resistant of the feldspars 

 then its seemingly low ratio to the orthoclase indicates that 

 there has not been enough weathering to obscure the original 

 relations of the feldspars. Therefore the indications in the 

 Catahoula sandstone favor slightly the assumption of an acid 



Am. Jour. Sci.— Fourth Series, Vol. XXXIX, No. 231. -March, 1915. 

 19 



