20 



SCIENCE. 



[N. S. Vol. XVUI. No. 444. 



seem, therefore, that Saccharomyces niger is to 

 be added to the small list of fungi which are 

 thus capable of assimilating free nitrogen 

 from the air. Mention was also made of cer- 

 tain other of the nutrition phenomena of this 

 fungus. 



Dr. George T. Moore gave a very interesting 

 address on a new method of artificially in- 

 oculating soils for leg-umes with the nitrogen- 

 assimilating, tubercle-forming bacteria. The 

 practical application of the method is very 

 simple and was fully described. The paper 

 will be published in full as a bulletin of the 

 Bureau of Plant Industry, U. S. Department 

 of Agriculture. H. J. Webber. 



DISCUSSION AND CORRESPONDENCE. 



THE GRAND GULF FORMATION. 



In Science of December 12, 1902, Professor 

 Dall, in commenting on our note on the Grand 

 Gulf Formation, published in the number for 

 November 21, 1902, calls our attention to 

 two errors, which we now acknowledge and 

 are very glad to correct. We made the state- 

 ment that Dr. Hilgard had considered the 

 Grand Gulf as of Eocene age. This is a mis- 

 take which escaped us both in the manuscript 

 and in the proof-reading. Since Dr. Hil- 

 gard's work forms the basis of all our knowl- 

 edge of the Gulf Coastal Plain, we knew from 

 long-continued study thereof that there was 

 not a line in all his writings which could be 

 interpreted as even suggesting this age for 

 the Grand Gulf. So also we were mistaken 

 in saying that Professor Dall had regarded 

 it as of Eocene age. 



It is, furthermore, evident from Professor 

 Dall's criticisms that we have not stated our 

 case with sufficient clearness to prevent mis- 

 understanding of our position. Inasmuch as 

 to us the facts in our possession seem to 

 afford absolute proof of the correctness of 

 our conclusions, we beg to submit the evi- 

 dence somewhat more fully to the considera- 

 tion of the geologists interested. 



At the outset it seems necessary to define 

 clearly what we mean by Grand Gulf, and 

 we can do no better than to follow Hilgard, 

 who has so well described these beds, and who 



has correctly mapped them as covering the 

 lower part of the state of Mississippi from 

 the southern limit of the Vicksburg down to 

 within a few miles of the Gulf of Mexico. 



The materials of the formation are sand- 

 stones, sands and clays, with silicified trunks 

 of trees and beds of lignite, and lignitic clays 

 containing leaf impressions, badly preserved 

 and incapable of determination. Concern- 

 ing these Grand Gulf beds Dr. Hilgard re- 

 marks : " Two ijoints confront us in the dis- 

 cussion of the relations of the formation to 

 the sea; the great rarity of the calcareous 

 feature in the main body of the formation, 

 and the utterly ' unmarine ' character of the 

 materials generally, in the constant recur- 

 rence of the lignito-gypseous facies." And 

 again, " Of the sweep of 900 miles thus out- 

 lined as the known extent of this formation, 

 400 may be considered as having been ex- 

 amined sufficiently in detail to prove the 

 absence of marine fossils from the formation; 

 the portion so examined embracing, moreover, 

 its widest part and fully two thirds of the 

 area of the outcrop." * 



By the characters thus outlined, this great 

 fresh-water formation has been recognized 

 and described by the geologists in Georgia, 

 Florida, Alabama, Mississippi, Louisiana and 

 Texas. No one has had any serious difficulty 

 in distinguishing it in the field; but every one 

 has had difficulty in reconciling the known 

 facts of its surface distribution with any satis- 

 factory assignment of it to a definite place 

 in the stratigraphic column. The only forma- 

 tion with which it is at all likely to be con- 

 founded is the Lafayette, which everywhere, 

 according to Hilgard and other geologists, 

 directly overlies it, and of which the materials 

 are often quite similar; and we have conclu- 

 sive evidence that parts of the Grand Gulf 

 have by several authors been included in the 

 Lafayette. 



In their relations also to the underlying 

 older strata, these two formations have much 

 in common ; for instance, they both ' blanket ' 

 a number of older formations, but the Grand 

 Gulf, so far as yet known, overlaps only Mio- 

 cene, Oligocene and Eocene as far down as 



* Am. Jour. Sci., Vol. XXII., July, 1881. 



