foot bed of yellowish limestone, briskly effervescing with dilute acid, which was struck at a depth of 100 feet from the surface, or about 300 feet from the base of the Dundee. This seems to correspond in position with a group of limestones, 276 feet in thickness, containing corals, chert, and dolomite beds, reported by Hunt as occurring at Goderich, at the foot of Lake Huron. The series was struck at a depth of 357 feet from the surface and is overlain by 278 feet of dolomite, with thin limestone layers.* Hunt regarded this as an intercalation of Corniferous (Dundee) amongst the beds of Lower Helderberg (Monroe). The overlying dolomite was supposed to be continuous with the Lake Erie beds containing Waterlime fossils. The limestone itself is of the gray non-magnesian variety, rich in fossils, corals and chert. 'From fragments in the cores Hall identified Favosites Winchelli Rom., and Favosites Emmonsii, Hall, these being characteristically Devonian. This peculiar limestone series is underlain by dolomite containing

gypsum.

§ 5. Lithological history.

It is evident from the variety of rock material represented as mak ing up the extensive Monroe series that its formation must have taken place under widely varying conditions. The sea bottom was subject to repeated oscillations and the shore line advanced and receded, probably several times. Geologically it was a period of unrest, these disturbances in Michigan being but distant ripples from the agitation which brought the Silurian to a close and caused the advance in plant and animal life. The beds of sandrock and the conglomerate of Ohio and Mackinac were formed while shore conditions prevailed. The beds of shale, which occur toward the bottom of the series, represent the muddy accumulations of the off-shore. The beds of dolomite, which comprise the main bulk of the series in southeastern Michigan, seem to have been formed in a shallow, open sea, sufficiently distant from land not to get much sand and mud.† The paucity of fossil remains in such beds led Cordier, Hunt, Ramsey and others, to assign to them a chemical origin. If such a theory is accepted in whole, or in part, then this dolomite dough must be regarded as a chemical precipitate, to be described below. The oscillations of the earth's crust occasionally exposed patches of this

*Geological Survey of Canada, Report of Progress, 1876-77, pp. 221-243.

+Or the climate may have been dry and the land low and the sea shallow so that not much sediment was formed. L.

sea-bottom to the air and sun, which upon drying shrank and gave the mud cracks mentioned as occurring about Monroe, as well as elsewhere. These were later submerged, covered with soft slime and indefinitely preserved. Pressure, carbon dioxide gas and probably a low degree of heat have compacted these soft layers into firm rock. The conversion of calcium carbonate into the double carbonate of calcium and magnesium, known as dolomite, has as yet received no entirely satisfactory explanation. The geological occurrence of dolomite gives no warrant for the theory of Von Buch that it represents limestone, in which some of the calcium has been replaced by magnesium, through the agency of magnesia vapors rising from beneath. Neither can it be maintained that thousands of square miles of heavily bedded dolomite have resulted from the action of percolating spring waters carrying magnesium salts in solution. Still less probable is the theory that the dolomitization is due to the action of molten magmas from the interior of the earth which have come in contact with limestone beds. In sea water there occur considerable quantities of magnesium chloride (10.878 parts) and magnesium sulphate (4.323 parts), in every 100 parts of the total salts, according to the analyses of Dittmar. By experiment it has been found that these salts, under high pressure, and above the boiling point of water, can convert limestone into genuine dolomite. An analysis of coralline limestone from the elevated coral island Matia, one of the Society group, according to Dana (Corals and Coral Is lands, 1872, p. 357), gave 38.07% of magnesium carbonate, showing it to be in reality dolomite. Ordinary coral contains from 95 to 98% of calcium carbonate, with only a slight amount of magnesium carbonate. Dana concludes that this introduction of magnesium into the composition of the coral has taken place apparently in the sea. water at ordinary temperatures and without the agency of any mineral waters except those of the ocean. "But the sand or mud may have been that of a contracting and evaporating lagoon, in which the magnesian and other salts of the ocean were in a concentrated state." These conclusions of Dana are accepted by Neumayr* and probably represent the consensus of geological opinion at the present time.

October 28th, 1844, Cordier deposited with the French Academy of Sciences a sealed packet, which after his death was opened and

*Erdgeschichte, 1886, p. 590.

found to contain a paper entitled "De l'origine des roches calcaires qui n'appartiennent pas au sol primordial."* In this the chemical origin of the great bulk of our limestones and dolomites is maintained, the carbonates being derived from the reaction of carbonate of soda upon the chlorides of calcium and magnesium in sea water. The supply of sodium carbonate he believed to be derived from the decomposition of feldspars, from alkaline springs and from plutonic emanations. Some three years before this information was made public, Hunt had published an elaborate account of his experiments on salts of lime and magnesio, under the title "On Some Reactions. of the Salts of Lime and Magnesia, and on the Formation of Gypsums and Magnesian Rocks," in which he had arrived at about the same. conclusions. The source of the carbonate of soda, in all cases, he considered to be decomposing feldspathic minerals of the sedimentary rocks and the argillaceous beds are the correlatives of the limestones. The conclusion reached in this paper (p. 383) is that

"Dolomites, magnesites and magnesian marls, have had their origin in sediments of magnesian carbonate formed by the evaporation of solutions of bicarbonate of magnesia. These solutions have been produced by the action of bicarbonate of lime upon solutions of sulphate of magnesia, in which case gypsum is a subsidiary product; or by the decomposition of solutions of sulphate or chloride of magnesium by the waters of rivers or springs containing bicarbonate of soda. The subsequent action of heat upon such magnesian sediments, either alone or mingled with carbonate of lime, has changed them into magnesite or dolomite."

Hunt insists that calcium chloride must be absent, whatever the reaction, and that isolated and evaporating basins are indispensable conditions for the formation and deposition of magnesian carbonate.‡ An analysis of waters from the equivalent of these Monroe beds, in Western Ontario, at Chatham, Petrolea and Bothwell shows them to be derived from a bittern, "the result of the evaporation of the waters of an ancient sea."

One of the most recent discussions of the question is contributed by Dr. E. Philippi, of Berlin, in a paper entitled, "Ueber einen Dolomitisirungs vorgang an südalpinen Conchodondolomit." The author believes that dolomites in many instances, at least, result from a leaching of a lime rock which already contains a certain quantity of magnesium carbonate. The theory is based upon the supposition that the carbonate of lime is more readily dissolved than the carbonate of magnesium, and hence that under proper conditions

*C. R., Feb. 1862, pp. 293 to 299.

American Journal of Science, 2d series, Vol. XXVIII, 1859, pp. 170-187, 365-383. Chemical and Geological Essays, 1891, p. 92.

Neues Johrbuch für Mineralogie, Geologie und Palæontologie, I Band. Erstes Heft, 1899, pages 32 to 16.

more of the former would be removed. The result of this would be to produce a rock which would show relatively greater and greater quantities of magnesia until the normal proportion of lime and magnesia might be attained. There is no reason to think that this theory can have any application in the case of our Monroe beds, where 900 to 1,000 feet of practically horizontal layers are evenly dolomitized. It would necessitate the destruction of unsupposably great masses of limestone, unless there is assumed to be present, from the first, considerable quantities of the magnesium carbonate. If such supposition is made then the difficulty of accounting for this carbonate is as great as is that of the original question. In ordinary corals which contribute so largely to the formation of limestones, there is present but a very small quantity of magnesia, Isis hippuris yielding Forchhammer the relatively large amount of 6.36%. Upon the supposition, however, that the magnesium carbonate is but one-tenth as soluble as the calcium carbonate, it is evident that in the process of leaching such a stone, the magnesium would have practically disappeared before the normal proportions for the two had been attained. There is no apparent agreement upon the part of supposed authorities however, in regard to the relative solubility of these two carbonates. Bischof found that a prolonged action of water, charged with carbonic acid, upon a limestone containing 11.54% of magnesium carbonate, dissolved out 4.29% of its calcium carbonate, but without appreciably affecting the magne sium.* With water containing three to four grams of magnesium sulphate to the liter, Hunt found that there would be dissolved: 1.2 grams of carbonate of lime and 1 gram of carbonate of magnesia to the liter.

From his researches Dittmar concluded that in sea water carbonate of magnesia is far more freely soluble than is carbonate of lime. In all the above cases the water contained different ingredients but nothing which might not be present in the waters percolating through a bed of limestone. Additional evidence against the application of this leaching process to our Monroe beds is furnished by the complete conversion into dolomite of the granules of the bed of oölite. These were originally calcareous and, judging from the Great Salt Lake oölite now in process of formation, have not been appreciably reduced in size. There has taken place here a partial re

*Lehrbuch, Vol. II, p. 1176. See also Treadwell and Reuter in Zeits. für Anorganische Chemie, Vol. XVII, 1898, pp. 170-204.

placement of the calcium carbonate with the carbonate of magnesium, this action occurring at the time of the formation of the beds and in sea water strongly charged with magnesium salts.

It seems necessary to conclude from these and still other considerations that a great inland basin had been formed in which the waters were greatly concentrated, while the open sea retained its normal composition. Hunt explains the intercalation of Corniferous amongst the beds of the Salina, by assuming that in the ocean the life had advanced into the Corniferous age, while that in the basin itself was still Silurian. A temporary influx of the sea into a part of this basin would bring in the higher forms and establish a new set of conditions, favorable for the growth of corals and the production of limestone. The cutting off of the direct supply from the ocean would in time exterminate the life in this particular region and by the opening of direct communication with the remaining interior basins, former and older geological and palæontological conditions would again prevail. Such basins would supply the necessary conditions for the deposition of gypsum, or anhydrite, and for the formation of rock salt found further north. Through the continued concentration of the water contained, after the point of saturation had been passed, there would be deposited a layer of calcium carbonate followed by one of calcium sulphate and later one of rock salt.* As evaporation progressed the water would be withdrawn from the margins of the basin and the deposits of salt would not extend shoreward as far as those of the lime carbonate and calcium sulphate. For this reason we find beneath Monroe more or less anhydrite, but no rock salt until we pass further north, in which direction numerous beds are developed, some of them attaining surprising thickness beneath Detroit. The deeper portions of the original evaporating basin then lay to the north of Monroe county, and the shore to the south. After the deposition of most of the calcium carbonate, calcium sulphate and salt there would remain in the central and deeper portions of the basin a solution, rich in chlorides of calcium, magnesium and sodium, and consequently intensely bitter to the taste. If complete evaporation took place these salts would crystallize, but if not the bittern containing them would impregnate thoroughly the beds formed later. Percolating waters would subsequently take up these salts again and give rise *See Vol. V, Geol. Survey of Mich., Pt. II, pp. ix to xix. "The origin of Salt, Gypsum and Petroleum," Lucius L. Hubbard.