SCIENCE 



NEW YORK, JUNE 9, 1893. 



NOTE ON THE "AGE OF THE EARTH."' 



BY W. J. MCGBE, U. S. GEOLOGICAL SURVEY, WASHINGTON, D.C. 



Roughly classified, there are four principal ways of estimating 

 the duration of geologic time, of which two are geologic and two 

 non-geologic. The first of these is based on sedimentation, and 

 the second on erosion ; the third is based on terrestrial temperature 

 and supposed rate of cooling, and may be designated physical; 

 the fourth method rests on inferences as to the cooling of the sun 

 and other cosmic changes and conditions, and may be called as- 

 tronomical. 



The sedimentation estimate, as commonly applied, depends on 

 an erosion estimate for its unit; usually it is assumed that the 

 rate of degradation of the land is a foot in 3,000 to 7,000 years, and 

 that in the long run the rate of sedimentation on the sea-bottoms 

 of the globe is the same. The unit rate commonly accepted is 

 that determined by Humphreys and Abbot from measurements 

 of the matter transported by the Mississippi River, or one foot in 

 6,000 years = 1 mile in about 80,000,000 years. In earlier decades 

 the aggregate thickness of sediments was usually placed at a few 

 miles only, but probably no modern geologist acquainted with 

 the results of researches in the Algonkian rocks of the Lake 

 Superior and other regions would venture to estimate the total 

 thickness at less than 50 miles; and this rate and thickness of 

 sediments would indicate a period of 1,500,000,000 years for the 

 deposition of the stratified rocks of the earth. It is probable that 

 geologic process was more active in the earlier ages than at 

 present; on the other hand, the deposition of the stratified rocks 

 represents only the closing episode in the history of the earth — 

 ages must have been required for the antecedent cooling and en- 

 crusting of the planet before the transfer of materials by hydric 

 agency began. 



Until recently the erosion estimate has seldom been applied, 

 except as a unit for the sedimentation estimate ; and even now 

 it is applicable only to the later eons without the introduction of 

 so many unknown quantities as to vitiate its results. Perhaps 

 the most favorably conditioned region for the use of this method 

 thus far studied is that found in eastern United States. Here 

 rate-units have been determined from the measurements of re- 

 cession of Niagara Falls and other cataracts, and these have been 

 corroborated by measurements of the recession of Saint Anthony 

 and other cataracts in the Mississippi Valley. These rate-measures 

 and the measures of the volume of material removed from the 

 gorges since the disappearance of the latest ice-sheet of the Pleis- 

 tocene yield a fairly consistent chronometer for the post-glacial 

 period, giving a value ranging from a few thousand to as many 

 as 50,000 years. Toward the margin of the glaciated area in 

 eastern United States there is another series of gorges, of which 

 that of the Potomac, between Great Falls and Georgetown, may 

 be taken as the type, representing erosion since the close of the 

 Columbia period, or since the end of the first ice-invasion of the 

 Pleistocene. A number of these gorges have been studied by 

 Chamberlin, Salisbury, Gilbert, and others, as well as by the 

 writer; and all geologists familiar with them aie agreed that, if 

 the post-glacial erosion is represented by unity, the post-Columbia 

 erosion must be represented by two figures. These post-Columbia 

 gorges themselves are excavated in the bottoms of many times 

 longer and wider gorges recently shown to have been cut since 



1 The meeting of the Geological Society of Washington, held April 12, 1893, 

 was devoted to a symposium on the age of the earth, based chiefly on the re- 

 cent article on that subject by Mr. Clarence King (Amer. Jour. Scl., vol. xlv., 

 1893, pp. 1-20); and these paragraphs are a, revised abstract of remarks made 

 on that occasion. 



the deposition of the Lafayette formation — eg., the post-Colum- 

 bia gorge of the Potomac is not over 13 miles long, a quarter of a 

 mile in mean width, and 100 feet in mean depth, or j\ of a cubic 

 mile in content; while the post-Lafayette gorge of the upper 

 river with its principal tributaries may fairly be put at 600 miles 

 in aggregate length, 1 mile in mean width, and 350 feet in mean 

 depth, or 30 cubic miles in content, or more than 500 times greater 

 than the post-Columbia gorge. 



The recent estimates for the post-glacial period derived from 

 the Niagara and other gorges are shorter than of old, ranging 

 from 5,000 to 10,000 years; it is not conservative to estimate the 

 post-Columbia gorges at less than 20 or 30 times as old (assuming 

 erosion to be uniform), or, say, 300,000 years. Now the post- 

 Lafayette period is represented not only by the gorges trenching 

 the Piedmont and Appalachian regions, but also by the widespread 

 ravining and hill-sculpturing of the pre-Lafayette base-level pene- 

 plain of these regions, by great estuaries hundreds of feet deep 

 and scores of miles broad traversing the coastal plain, and by the 

 entire removal of two-thirds of the volume of the Lafayette for- 

 mation (nearly all along and north of the Potomac) and extensive 

 degradation of subjacent rocks, i.e., by erosion fully 500 times 

 greater in the gorges and many thousand times greater over the 

 general surface than the post-Columbia erosion ; and it is accord- 

 ingly hardly conservative to estimate this period at less that 20. 

 50, or 100 times longer than the post-Columbia period, or say 

 5,000,000 or 10,000,000 years. In this estimate allowance is made 

 for the discrepancy between the figures based directly on relative 

 erosion and those derived from the Humphreys and Abbot co- 

 eflflcient (1,000,000 or 3,000,000 yfeirs for the 200 or 300 feet of 

 post-Lafayette degradation^ It is, of course, to be remembered 

 that erosion, per se, does not give a trustworthy time-measure, 

 since stream-work is a function of declivity rather than time; but 

 in eastern United States the physiographic conditions affecting 

 stream-declivity have varied so little as to render the erosion-rate 

 here exceptionally uniform. 



In brief, while the erosion estimate of geologic time is subject 

 to a large probable error, even to a considerable "factor of safety," 

 the phenomena of eastern United States indicate an enormous 

 lapse of time, probably reaching into millions of years, since the 

 deposition of the late Neocene-Lafayette formation ; yet this is one 

 of the latest episodes in the development of the continent. 



Combining the erosion estimate and the sedimentation estimate 

 by employing the former so far as applicable and then using 

 sediment ratios beyond, with a •' factor of safety " beginning at 4 

 for the last and shortest period and raised to successively higher 

 powers with each successive period and age counted backward 

 into the past, the following values are obtained" : — 



' American Anthropologist, vol. v., 1892, pp. 329-310. Throush a simple and 

 evident arithmetic error in this paper, Dana's ratios of 1, 3, and 12 for the 

 Cenozoic, Mesozoic, and post-Cambrian Paleozoic are compuled as 1, 3, and 36. 



