31'8 NATURAL HISTORY BULLETIN. 



of material, and no loess will be formed; where these condi- 

 tions are best developed, there the deposit will be thickest. It 

 has long ago been observed that the loess is best developed 

 along our larger river-courses, and it is there that these con- 

 ditions are all most likely to be presented. A more detailed 

 reference to these conditions may be of interest: 



i. The source of supply. — It is quite generally conceded 

 that the finely comminuted particles which make up the loess 

 originated primarily in the drift. In the north it was loosened 

 and sifted out from the coarser material by rain, by wind, by 

 burrowing worms, insects, mammals, etc., and by scratching 

 birds, and to some extent by growing plants, while its volume 

 was increased by the chemical decomposition and disintegra- 

 tion of the coarser materials thus exposed.* 



Some of this material was washed into the streams, and 

 some of it was (and is) blown about by the winds. In the 

 southern Mississippi valley, where there is no glacial drift 

 and yet an abundance of loess, the material of the loess was 

 probably all brought down by the great river, and chiefly by 

 its Missouri branch. During the summer months our streams 

 reach their lowest levels. Great bars of sand and mud are 

 thereby exposed to the dessicating influence of sun and wind. 

 During the summer, too, strong winds sweep along their val- 

 leys and gather up the fine material so exposed, f 



With it is also mingled more or less of the finer material 

 gathered by the winds directly from higher grounds, and per- 

 haps with calcareous particles from fluviatile shells where pres- 

 ent in the streams, but the main supply in the south and per- 

 haps in large part elsewhere where loess was formed, was ob- 

 tained from the bars of the stream. 



*That finer material is thus removed and the coarse material so con- 

 centrated near the surface has already been observed. See Proc. la. 

 Acad. Sci., vol. IV, p. 70, 1897; etc. 



t For discussion of the probable amount of this material see Jour. 

 Geol., vol. VII, p. 135. — 1899; or Proc. la. Acad, Sci., vol. VI, pp. 109 

 and 110. — 1899. 



