Correspondence. 361 
creeks near Cash City, and of Crooked creek near Odee, and other 
streams of this region where the ground-water is in or in communication 
with the upper part of the Red Bluff, including the great salt well near 
Meade, the porous, sunken mass of whose floor undoubtedly reaches 
down to a contact with Red Bluff walls. Indeed, it is not improbable that 
the cavity by whose collapse (in March 1879) this "well" was produced, 
was in the Red Bluff formation and owed its existence to the action of 
subterranean waters which literally mined the saliferous shale, dissolv- 
ing the salt, and (since the buoyancy, and so the transporting power, of 
brine is greater that of fresh water) probably carrying off with the brine 
minute particles of the shale itself; and it seems probable that St. Ja- 
cob's well and the other "drops" in Little and Big basins and in Coman- 
che county have been formed in a similar manner.* 
The Day Creek Dolomite. — In the western part of Clark and in 
Meade county, Kansas, the Day Creek formation losas its typical char- 
acter. It is there represented by a band of greenish-gray to red and 
gray sandstone with occasional streaks of dolomite. This band is not 
always well differentiated and is often c.mcealed by talus, but appears 
as a soft greenish-gray sandstone, twenty-five feet below the Big Basin 
sandstone, in a deep right-hand canyon of Big Sandy creek (containing 
a row of dugouts and a well) a little below the junction of Big Sandy 
and Gyp creeks. On Two-mile creek, farther southwest, it appears as 
a two-foot bed of red and gray sandstone similar to the Big Basin sand- 
stone and at the same distance below it. At Odee post-office, in Meade 
county, a thin local ledge of reddish-gray, pisolitic and stalagmitic-ap- 
pearing rock outcropping in a left-hand ravine of Crooked creek north- 
east of the house of Mr. Peteflsh, is apparently not far from the horizon 
of the Day Creek, and may possibly represent it in that quarter; but 
such an identification is very doubtful, and the true position of this 
ledge will require more exact observations than any I have yet had op- 
portunity te make. 
From this loss of typical character which befalls it in the most west- 
erly outcrops of the Kiger division in Kansas and fi-om its moderate 
thickness, one might be ready to infer that the Day Creek is an unim- 
portant formation. Such a conclusion would be erroneous, as the ex- 
tent of this dolomite in essentially its typical phase is very much greater 
in the general southwesterly course of the Cimarron outcrops than east 
and west, and it is probable that the horizon of its modified western ex- 
tension in southern Kansas can usually, if not always, be identified 
where its position is exposed and where due care is exercised in the ob- 
servations. The brow of the Red hills near Watonga, Oklahoma, is 
coped with the Day Creek dolomite, which there presents itself as a com- 
pact stratum of gray, somewhat pinkish or reddish-tinged, cherty-hard 
rock, little different from the ty[)ical ledge that skirts the flanks of 
Mount Lookout in Clark county, Kansas. The stratum has a thickness 
of three feet. How far the Day Creek dolomite can bo traced be- 
*Drops or sink-holes, are also formed by the collapsiuK of gjpsuiu-caves, but the 
gypsum is usually "in evidence" in such cases. 
