270 M. 1. Goldman — Catahoula Standsone of Texas. 



good sizing. In the analyses of dune sands published by 

 Udden (cf. average presented in fig. 5) an average of 65 per 

 cent is in the size between \ and -J mm , with the maximum, 

 in almost all the numerous analyses presented, in that grade. 

 Atterberg, on the contrary, believes that 0*6 to # 2 mm (nearly 

 \ to -J-) is so characteristic and prevalent a size in dune sands 

 that in his classification of sand sizes he would actually call 

 that size " dune." But he admits in his answer* to Keilhack's 

 criticism that he considers only the dune ridges, where the 

 material being least protected is naturally coarsest, and 

 expresses the belief, moreover, that if derived from a more 

 mixed and finer source material than was supplied to his 

 coastal dunes the dunes themselves might contain more fine 

 material. Lehmann's results, too, while not directly compar- 

 able on account of differences in sieve mesh, rather support 

 Udden's results, though in one of his analyses there is 36 per 

 cent even in a size so coarse as 1— §- mm , and in general there is 

 much greater variation between different samples analyzed by 

 him. Oldham's analyses, again (cf. average presented in fig. 7), 

 as far as comparison is possible on account of the differences of 

 sieve mesh, show a marked tendency to conform in the size of 

 grain of the maximum to that of the- Catahoula sandstone, 

 with in fact even a larger portion of material coarser than J mm 

 tban in the Catahoula sandstone. The same is true, to even a 

 greater extent, of Thoulet's analysis of Sahara sand (fig. 6), of 

 which only 5 per cent is finer than 0'21 mm . Now it is note- 

 worthy that while all the other analyses are of material from 

 the temperate regions, these last two authors were dealing with 

 samples from the tropics, so that their results tend to support 

 Keilhack's contention that much depends on the prevalent 

 force of the winds. 



If we consider another factor brought out by Udden, that 

 outside the maximum portion the greater part of the material 

 in seolian sands tends to be in the next coarsest size, we find 

 again a divergence from the composition of the Catahoula sand- 

 stone, but in view of the difference in the position of the maxi- 

 mum in his results and mine this fact may really be interpreted 

 as well for or against similarity. 



Comparison with any of the subaqueous deposits (figs. 8 to 

 12) shows at once two very significant differences; i. e., the 

 presence of a considerable portion of " clay" in all of these (a 

 constituent which is negligible in the Catahoula and practi- 

 cally absent in the beach and seolian sands) and the greater 

 predominance of the finer sizes in them. Thus the typical 

 analysis from 25 feet depth of water off the shore of the Gulf 

 of Lyon (fig. 8) has all the characteristics of a beach sand 



* Atterberg, Chem. Ztg., xxix, No. 80, p. 1074, 1905. 



