NOVEUBEE 21, 1902. J 



SCIENCE. 



835 



count for 33 atmospheres pressure in a nor- 

 mal solution of cane sugar, nor does it, in our 

 opinion, account for the ascent of sap in the 

 tube attached to the Begonia. 



Van't Hoff's law is based upon Pfefier's 

 researches, and Pf effer states : ' The same 

 pressure (22.4 atmospheres) must be exerted 

 by a solution of 342 grams of cane sugar in 

 one liter of water ;' and further : ' Hence it 

 follows that osmotic values may be calcu- 

 lated directly with perfect safety and ac- 

 curacy.' It is certain that, if Morse and 

 Frazer's results are reliable, Pfeffer's osmotic 

 conclusions and van't Hoff's theory collapse, 

 and the true osmotic pressures are not yet 

 known. 



The experiment with the Begonia plant, in 

 the light of the results of Morse and Frazer, 

 leads one to suppose that the actual osmotic 

 pressure — or that force producing pressure — 

 is far in excess of that indicated in van't 

 Hoff's law. James B. Dandeno. 



Agkicdxtukax College, Mich. 

 October 30, 1902. 



THE GRAND GULF FORMATION. 



The classification of the formations of the 

 gulf coastal plain more recent than the Vicks- 

 burg Limestone, has long presented difficulties 

 to the geologist. 



One of the most important of these forma- 

 tions, as regards at least extent of surface 

 outcrop, is the Grand Gulf, classed as Eocene 

 by Dr. Hilgard and by Mr. Kennedy of the 

 Texas Survey; as Oligocene by Mr. Gilbert 

 Harris and Miss Maury; as Miocene by the 

 Alabama Geological Survey ('Coastal Plain 

 Eeport'). Dr. Wm. H. Hall, who has pub- 

 lished much concerning the formation, has at 

 different times referred it to the Eocene, the 

 Oligocene and the Miocene. 



These classifications, in the absence of char- 

 acteristic fossils, have been based largely, if 

 not solely, on the stratigraphical position of 

 the beds, heretofore supposed to be unconform- 

 ably overlying, and chronologically next suc- 

 ceeding the Vicksburg Limestone, and many 

 sections have been published showing these 

 beds and the Vicksburg Limestone in imme- 

 diate contact. 



Our observations made during the past sum- 



mer, of the surface distribution of the Grand 

 Gulf beds in Washington, Mobile, Baldwin, 

 Escambia and Covington Counties; an inter- 

 pretation, in the light of these observations, 

 of some sections recorded in the ' Coastal Plain 

 Report ' ; and our identifications of some shells 

 brought up from borings recently made at the 

 Bascom Well near Mobile, and at Alabama 

 Port in the southeastern part of Mobile 

 County will, it is believed, help to clear up 

 some of the obscurities which have hereto- 

 fore beclouded the classification of the coastal 

 plain formations of the Gulf States. 



1. From Healing Springs in Washington 

 County southward to within three miles of the 

 coast near Bayou La Batre in Mobile County, 

 the surface formations are Lafayette sands 

 and pebbles, resting directly on Grand Gulf 

 mottled clays, overlying cross-bedded sands of 

 the same formation. Along bay, river and 

 gulf margins the more recent Port Hudson 

 strata occur. 



2. In Baldwin County, from its northern 

 border down to the Gulf coast, a distance of 

 seventy miles or more, the surface is in like 

 manner formed by the Grand Gulf beds with 

 overlying mantle of Lafayette. 



Southward of the line of the L. & N. rail- 

 road, this county is a high plateau, 200 feet 

 above tide near the line of the railroad, de- 

 clining to 75 feet or more on Perdido Bay; 

 with surface, away from the immediate vi- 

 cinity of the streams and bays, almost per- 

 fectly flat, but for the slight sinks or depres- 

 sions of the hundreds of ponds and savannas 

 which characterize the Grand Gulf in the 

 lower parts of the two coast counties of Ala- 

 bama, and contiguous parts of Florida. The 

 original plain in Baldwin has been far less 

 modified by erosion than that of Mobile. 



The high land in places extends to the 

 water's edge, terminating in high bluffs along 

 Mobile Bay from Daphne down below Mon- 

 trose, and along Perdido Bay from above 

 Suarez's landing dovsm to Soldier Creek. 



These bluffs, Y5 feet and upwards sheer 

 height, show in most characteristic exposures 

 the thin capping of Lafayette resting on the 

 clays and cross-beds sands of the Grand Gulf. 



3. While in most cases the Grand Gulf along 



