The History of Oxygenic Concentration in theEarth’s AtmosphereSouthwest Centre for Advanced Studies, P.O. Box 8478, Dallas 5, Texas, U.S.A.Receiued 30th January, 1964Geologic evidence points to the absence of primordial atmosphere upon the Earth’s formationand suggests that the present atmosphere is of secondary origin derived from sources containedwithin the Earth. Initially, the important probable constituents, H2, N2, C02 and H20 have beenderived from volcanic effluents. Quantitatively, these effluents can account for the present volumeof the oceans and the development of an atmosphere over a period of 3-4x 109 years.Since oxygen is absent from volcanic effluents, its origin and rise have been the subject of wide-spread speculation and uncertainty. The present calculations based on complete u.-v.absorptiondata show that the photodissociation of water is self-regulated by the subsequent formation of02, thus exercising a shadowing effect on H20 on account of the non-exponential distributionof water vapour. This plus the effect of any C02 places an upper limit on the oxygenic pressurein the primitive atmosphere at <lO-3 present atmospheric level (P.A.L.). It is further apparentthat sufficient u.-v. energy is available to dissociate 0 2 in the primitive atmosphere to the extentthat 0 3 will be maintained as an important constituent near the surface. From absorption datait is shown that sufficient oxygen is released by u.-v. dissociation of water vapour to account forthe crustal oxides at primitive oxygenic level.Subsequent rise of oxygenic level can only be attributed to photosynthesis in primitive cellularorganisms.Strong fluxes of lethal u.-v. in the range 2200-3OOOA severely restrict the ecologyto bottom dwelling organisms in shallow protected pools or seas at a depth of 5-10 metres and pre-vents spread of life to the oceans. Only when the spread of such organisms in this ecology is suf-ficient to produce 0 2 at a rate exceeding its photodissociation can the building of an oxygenicatmosphere begin.As oxygen rises to the level -10-2 P.A.L., 0 3 levels will rise to the point where the spread oflife in the oceans becomes possible and coincidentally, the oxygenic level reaches the “Pasteurpoint ’’ where organisms change from a mechanism of fermentation to respiration, raising theenergy available for chemosynthesis by a factor of -30-40.The attainment of this “ first critical level ” -10-2 P.A.L.-opens widespread new evolutionaryniches, still confined to the liquid hydrosphere, and suggests search for a widespread evolutionaryexplosion in the waters of the Earth.Only one such event is evident-the beginning of the Cambrianera -600 million years ago. Then evolution took place explosively and laid the foundation for allmodern phyla. By immediate inference we identify the oxygenic level, 10-2 P.A.L., with the openingof the Cambrian.Following this a rapid increase in marine photosynthesis raised the level of oxygen to -0.1P.A.L. At this “ second critical level ” lethal radiation would be shielded largely from dry landby 0 3 levels, thus opening a new ecological niche for evolution ashore.Search of geologic evidenceshows evolutionary development of several phyla ashore between mid- and late-Silurian, withgreat forests by early Devonian. By immediate inference we identify the oxygenic level, 4 - 1 P.A.L.,with the late Silurian, 420 million years ago.From this point, life would develop very rapidly on land. With rapid increase in total oxygenic-production by photosynthesis, the oxygen would build to its present level, and perhaps overswingas a consequence of the phase difference in production peaks of 0 2 and C02. Study of this balancesuggests that oxygen may have fluctuated since the Devonian era in a damped saw-tooth oscillationaround the present quasi-permanent level.The implications of this model suggest that the Cambrian was not preceded by a long period ofevolution of advanced organisms which somehow, in spite of diligent search of favourable sedi-ments for more than a century, have not been preserved in the fossil record.Likewise, this modelaccounts for the sudden evolutionary migration of several phyla of plants and animals ashore atthe late Silurian and the prior absence of evolutionary evidence on land. It suggests that thegeologic record should be read exactly as observed, without extrapolation, and raises a new problem12L. V. BERKNER AND L. C. MARSHALL 123on the limiting rate of evolutionary development as favourable, widespread evolutionary nichesare opened.The analysis of oxygen balance during the several eras also poses the problem ofthe present stability of the oxygenic level.This paper reports on a subject which might be classified under the generalheading of “ paleoatmospheres ” or of “ fossil atmospheres ”. This subject involvescritical quantitative study of the history of planetary atmospheres. Studies of thiskind are now feasible in view of the substantial data from space probes on solarradiation at all wavelengths, together with reasonably complete data on the ab-sorption of particular spectral regions by component atmospheric gases. There isalso a growing knowledge of the succession and rates of photochemical reactionsinvolved in the absorption processes.These studies are of value in understanding the composition and organizationof planetary atmospheres which are of increasing interest in view of their growingaccessability.Interpretation of the events of the physical and biological history ofa planet may be misleading, unless at the same time some reasonable estimates ofaccompanying atmospheres can be made.The analysis of the paleoatmosphere of Planet Earth is most convenient, sincean abundance of data are available. Pertinent data relate to the composition,distribution and photochemistry of the present Earth’s atmosphere ; to the be-haviour, composition, chemistry and geology of its surface and interior ; to relatedsubjects of molecular biology, biochemistry, and molecular genetics of its livingmaterials ; as well as to the paleontological and evolutionary evidence derived frommore than a century of study.Moreover, the implications apparent from a criticalstudy of the paleoatmosphere of the Earth open new and interesting lines of inquiryand suggest reasonable solutions to many of the so-called “puzzles” which, inthe formulation of the history of the Earth, have so far remained unsolved.In this paper, attention is directed only to the history of the growth of oxygenin the atmosphere of Planet Earth. This report is a prkcis outlining only theprincipal methods and conclusions of a more extended study from which a preliminarymodel has been formulated.*Out of the work of Goldschmidt,l Harrison Brown,29 3 Spencer-Jones,4 Kuiper,5Urey,69 7 Alfven,8 Fesenkov,g Vinogradov,lo and many others, the initial premiseis adopted that upon its agglomeration, the Earth was without a primordial atmo-sphere.The relative abundance of the rare gases, illustrated in table 1, after Brown,3shows that their abundance on Earth ranges from one-millionth to one ten-billionthof their cosmic abundance.TABLE ~.-FRACTLONATION FACTORS OF RARE GASESelement atomic weight fractionation factorneon 20 - 1010argon (36) 40 - 1 0 8krypton 83 - 2x 106xenon 130 - 106Likewise, the relative abundance of the lighter elements, H and He, on the Sunand the more massive planets, and the paucity of these lighter elements on the innerplanets, shows that during their agglomeration, these inner planets lost the greaterproportion of mass attributable to the usual abundance of the gaseous elements.* The more complete study will be published in an early issue of the J. Atm.Research ( h e r .Meteorological SOC.)124 OXYGENIC CONCENTRATION IN EARTH’S ATMOSPHEREThe accumulation of Ar40, from decay of K40, corresponds to the estimated age ofthe solid Earth of about 5 billion years.Thus, all lines of evidence point to the absence of any primordial atmosphere,and to subsequent growth of an atmosphere primarily from secondary sourcesself-contained in the Earth. This view is consistent with the agglomeration ofthe Earth from planetisimals whose gravitational fields were too small to retaina primordial atmosphere.According to Urey,6P 7 Vinogradov,lo and others,s~ 8 the physical chemistry ofcompounds comprising the major portion of the Earth indicate that since its agglom-eration the Earth has never been substantially molten.The Earth consequentlyretains large quantities of gases chemically bound in various ways. These are suf-ficient to provide for the source of the secondary atmosphere.The volcanic origin of this secondary atmosphere is developed by Rubey,119 12Urey,g Holland,l3 Vinogradov,lo and others. The continents have been built atan average rate of 1-3 km3 per year from volcanic effluents, based on estimates ofSapper,l4 Verhoogen,ls Bullard,l6 and Wilson.17-19. In evaluating these estimates,on one hand following Vinogradov,lo volcanic activity should have been consider-ably greater in earlier eras, precedent to long decay of radioactive elements.Onthe other hand, following Verhoogen,ls the andesitic volcanoes, particularly in thePacific andesite ring, represent to some extent merely reworked continental materials.Wilson’s 17 estimate of a rate of average continent building of 1.3 km per yearthroughout geological time, and down to present times, seems compatible withinan order of magnitude with present estimates of total volume of continental materials.Accompanying these solid effluents are corresponding volumes of gases-primarilyprimitive water vapour, presumably released from water of crystallization, togetherwith C02, Nz, SOz, Hz, C12 in substantial quantities, accompanied by traces ofmany other gases.11920 No oxygen is released directly from volcanic effluents.The absence of a significant content of oxygen in the primitive secondary atmo-sphere is confirmed by several lines of evidence.First, there is no suitable source,as will be shown later. Secondly, the incomplete oxidation of early sedimentarymaterials (- 3 billion years of age) as demonstrated by Rankama,zl Ramdohr,22Lepp and Goldich,23 and others, and summarized by Rutten,24 suggests very earlylithospheric sedimentation in a reducing atmosphere. This evidence is in concord-ance with conclusions of extensive studies by Holland.13 Finally, the rapidly grow-ing evidence on the origin of life on Planet Earth appears to fQrbid a significantoxygen concentration until photosynthesis has been achieved.The work of Oparin25 and Bernal26 directed attention to the steps leading tothe organization of the simple biological cell, which is now recognized as a veryadvanced evolutionary entity.Oparin 25 visualized a logical series of evolutionary syntheses starting frominorganic, and simple organic materials of non-biological origin (i.e., the simplestcompounds of H, C, 0 and N) together with some traces of other elements (suchas S, P and Fe), finally ending with the organized cell replete with living functionwhich Bernal26 defines pragmatically as “ the embodiment within a certain volumeof self-maintaining chemical processes ”.Oparin 25 suggested that each step in theachievement of the whole process was the consequence of natural experimentationon a large scale guided by natural selection.The extensive literature on evolution leading to the simple cell is developed inwork by Wald,27 Rabinowitch,28 Calvin,29-31 Anfinson,32 and many others.33~ 34Synthesis of amino acids and other complex elements of cell structure is demonstrablein a reducing atmosphere, in the presence of ultra-violet light, which provides thL.V. BERKNBR AND L. C . MARSHALL 125energy for chemo-synthesis through photo-excitation (cf. Miller).33 Viable chemicalprecursors to the living cell were selected step-by-step from a thin soup of ever morecomplex organic compounds.During this phase of primitive evolution of pre-living compounds, oxygen is apowerful poison, acting to break them down as they are formed. Abelson 35 hasshown, e.g., that organic materials such as amino acids which are normally stablefor long intervals in anoxygenic atmospheres, are quickly degraded in presence ofatmospheres that are significantly oxygenic, particularly in the presence of a catalyticenergy source such as visible light.(AFTER BERNAL, CALVIN, HOERYG, OPARIN.RABINOWITCH. WALD P OTHERS)I SIWIFICANT OUANTlTlES OFOXYGEN FORBIDWN 0,cOOOl PRESENT ATMOSPHEhlC LEVELIPA L llw 1200-3003 AT 0-10 km ALTITUDEFOUMTION OFLQUm HZO)POLTMERtZATlON OF MACROMOLECULESWITH NATURIL SELECTION 1 PRDTEINS,NUCLEIC ACIDS, ENZYMES, GENES, HORMONES)YY PHOTO ASSOCIATION OFINTERMEDIATE ORGANICMATERIALS IAMINO ACIDS, ETC)AUkERWIC MCT~BOUSU OF ORDANIC UATERIALSTHROUQfl FfRMENTATON I15 CALORIES)FORMATTION OF CELL MEMBRANECbLLODAL COAZLRVATSS ASSOCIATING SYMBIOTICUACROUOLECULES WITH SURVIVAL DETERUINED81 NATURAL SELECTIONABOUT -2 7 x 109 YEARSTIME-FIG.1 .-Diagrammatic visualization of evolution of the simple living cell.Photosynthesis can proceed in anoxygenic atmospheres as shown by Hill 36 andHill and Scarisbrick,37 and in the process the whole warehouse of oxygenic com-ponents required for growth and reproduction can be synthesized as shown byCalvin and his co-workers.29 The principal steps in evolution to the complete cell,replete with photosynthetic activity, are thus summarized in an elemental way infig. 1.Such evidence of an early anoxygenic atmosphere therefore leads to inquiryconcerning the origin and concentration of oxygen in this primitive, secondaryatmosphere of the Earth.Excellent data on solar radiation in the u.-v. spectrumare now available from the space probes of many workers, including Hall, Damonand Hinteregger ; 38 Detweiler, Garrett, Purcell and Tousey.39 Available data onsolar u.-v. radiation are summarized after Nawrocki and Papa,40 in fig. 2 and 3,after Johnson’s 41 new evaluation of the solar constant.Radiation down to 1400a arises in the upper 100 km of the solar photosphere.As shown from the studies of Wilson42 on the evolutionary history of stars of themain sequence, similar to our sun, this photospheric radiation above 1400A isextremely stable-probably self-regulating and unvarying over very long periodscomparable to the age of the Earth. Radiation below 1400A arises primarily inthe chromosphere and corona, and may have been as much as three times thepresent average level (thus involving total average fluctuations of as much as 2126 OXYGENIC CONCENTRATION IN EARTH'S ATMOSPHEREto 1 in this spectral region over shorter intervals). From the point of view of totalultra-violet energy, however, by far the major bulk (99.9 %) arises from the stablephotospheric radiation above 1400 a.wavelength (A)FIG.2.Colar intensity 700-1400 A.COMPOSITE DATA FROM NAWROCKI 0 WPAI/'O-'HOO 1660 ' l8bO ' 2obO ' 22bO ' 24b0 ' 26b0 ' 28bO 'wavelength (A)FIG. 3.-Solar intensity 1400-3000 A.Absorption of this ultra-violet spectrum by various possible component gasesabove 1000 A is summarized in fig. 4 after the rather complete work of Watanabe 439 44and the contributions of many others.45-5L.V. BERKNER AND L. C. MARSHALL 127In the spectral region above l400A absorption by H2 and N2, and by the raregases He, Ne, Ar, Kr and Xe is negligible. Of the important probable atmosphericconstituents, only H20, C02, 0 2 and O3 absorb radiation strongly in this redon,with only 0 3 important as an absorbing gas above 2200 A.I I I I I I I I I1000 1200 1400 1600 I800 2000 2 200wavelength (A)FIG. 4.-Composite of u.-v. absorption in atmospheric gases.II I I I I I I I I I I1800 2000 2200 2400 2600 2000 3000wavelength (A)FIG. 5.-U.-v. absorption in 0 3 (data from Watanabe and Vigroux).In order to interpret the relative contributions of different concentrations of theseveral possible constituent gases as absorbers over the u.-v.spectrum, these dataon radiation and on absorption must be combined at each wavelength over thespectrum. To do this, the path length expressed in cm s.t.p., required at each wave-length to reduce the incident radiation to a small selected arbitrary level is ascertainedfor each constituent gas in an atmosphere.The selection of the particular reference level is not important to the subsequentinterpretation, since translation to other reference levels can be made readily. Thereference level in this paper is determined by the path length of the constituentgas required to reduce the incident energy at each wave band to 1 ergcm-zsec-1(50A)-l. This reference energy level is 10-4 of the level in a 50A band at the pea128 OXYGENIC CONCENTRATION IN EARTH’S ATMOSPHEREof solar radiation at approximately 4500 A (i.e., the transmitted energy level, afterabsorption, which is 0.01 % of the peak radiation in a 50 A band, a transmitted levelequivalent to about 50 times the brightness of the full moon).In the present dis-cussion, the path length x of any constituent gas, required to reduce the radiationto 1 ergcm-2sec-1 (50A)-l, will be expressed as the path length required for its“ extinction ”. The results of combining radiation and absorption data are shownin fig. 6 for HzO vapour.n E3lo4103I020 PATH LENGTH N.T.P.10I10’’I 0-21400 1600 I800 2000 2200 2400wavelength (A)FIG. 6.Thickness of H20 required to absorb available u.-v.to “ extinction ” [(I erg cm-2 sec-1(50 A)-1)].Here it is apparent that the wavelength, at which significant penetration throughH20 vapour can occur, is not very sensitive to concentration. It drops only aboutlOOA for a decrease of three orders of magnitude of H20 vapour in the most sig-nificant range of pressures.” In absence of oxygen, and at low C02 pressures,the important photochemical reactions can be summarized by 509 51H20 +2h~+2H + 0which provides a source of atmospheric oxygen.Urey 7 has pointed out, however, that production of oxygen by photodissociationof H20 would be limited at some self-regulating concentration by the shadowingeffect of the 0 2 so produced. Since oxygen is distributed exponentially above thebase of the stratosphere, while water vapour is precipitated out to very low con-centrations at this same level, the presence of oxygen will shadow the P I 2 0 vapour* Note also that a change of reference level from 1 erg to 0.1 erg cm-2 sec-1(50 A)-1, for example,will simply shift the curve upward by one decade in fig.6L. V. BERKNER AND L. C. MARSHALL 129in its range of photodissociation as seen from fig. 7. In this figure it is evident thatwhen the path length of 0 2 above 10 km approaches 35 cm, water vapour is com-pletely shadowed from the H20 dissociative radiation-band up to 1950A. Thispath length of 35 cm of 0 2 above 10 km corresponds to a total path length of 0 2above the surface of about 100 cm s.t.p.This path length represents an oxygen level (see fig.7) of somewhat less than0.001 present atmospheric concentration. On the premises of the above discussion,the upper limit of oxygenic pressure in the primitive atmosphere is less than 0.1of present atmospheric level (P.A.L.).AT 0 01 PA.L._._ ~_._ 0 2 PATH LENGTH ~~ N T P AT PRESENT ATMOSPHERIC LEVELAT 01 P.A L /103n 102E a 35CMFc 101I10-110-2AT 0001 P A L .-JI I I , I # ,wavelength (A)1400 1600 I800 20004 IIIIIIIIIIIIIIIIIIIIIIII 1 ‘I /! I‘ I‘ II I 1 1 ,2200 2400FIG. 7.Thickness of 0 2 required to absorb available u.-v. to “ extinction ” [(l erg cm-2 sec-1(50 &-I)].The presence of significant C02 (see fig. 8) simply adds to this shadowing ofH20 vapour, and lowers this upper limit of oxygenic concentration in the primitiveatmosphere below this value, i.e., <lo-3 P.A.L.Geologists have heretofore assumed that because of the extensive oxides foundin the pre-Cambrian (Proterozoic) lithosphere, that atmosphere must have beenhighly oxygenic.This assumption seems unnecessary, and probably invalid, whenthe pertinent reactions are reviewed :0 + O+ M+02+ M02+O+M+03+M0 3 + M,+surface oxidescoupled with the additional reactions0 + 0 3 + 2 0 20 2 + hv-+O + 0.In the primitive atmosphere the supply of 0 3 is maintained close to the surfacewhere it is removed through surface oxidation at very high reaction rates.130 OXYGENIC CONCENTRATION IN EARTH’S ATMOSPHERESince oxygen appears in the primitive atmosphere in its most effective form forrapid oxidation of lithospheric materials, one is led to inquire into the availablesupply over geologic time.In this study the potential oxygen-supply has beencalculated from the total energy available for photodissociation. It is found thatthe available energy is of the order of 100 times (or more) that which is necessary toaccount for existing pre-Cambrian lithospheric oxides. (The calculations showthat for each kilometre of completely oxidized metals in the lithosphere, oxygenequivalent to the oxidation of about one-third of the water of the oceans wouldbe required.) It is further determined that sufficient u.-v. energy is available to-- AT 30 P.A.L.105104lo3t I I1400 1600 1800 2000~ //- .. .I t I I 1 12200 2400 2600wavelength (A)FIG. 8.Thickness of COz required to absorb available u.-v. to “ extinction ” [(l erg cm-2 sec-1(50 A)-I)],dissociate available 0 2 in the primitive atmosphere to the extent that 0 3 will bemaintained as a major constituent near the surface. These calculations lead to therealization that oxidation rate of lithospheric materials in the Archeozoic and inthe Proterozoic is dependent, not so much on the absolute concentration of oxygen,as on its chenlical form and the reaction rate in that form. Consequently, the classicassumption seems unnecessary that abundance of lithospheric oxides dictate highoxygenic levels in the primitive atmosphere.Thus in the primitive atmosphere, the oxygen balance is dominated by a rate ofloss which consumes oxygen promptly upon its production.The rapid removalof oxygen from the primitive atmosphere, and its inherent self-regulation by the“ Urey ” process,7 together with other geochemical and biological evidence, leadstherefore to the conclusion that the oxygen level in the primitive atmosphere of theEarth was <lO-3 (is., c0.1 %) of present atmospheric levelsL. V . BERKNER AND L. C. MARSHALL 131The subsequent rise of oxygenic level can be attributed only to photosynthesis.The summary of Rabinowitcli28 shows that in the present atmosphere, all oxygenpasses through the photosynthetic process in -2000 years, all C02 in -350 years,and all H20 in the oceans in -2 x 106 years. These intervals are very short com-pared to geological periods.Thus photosynthesis, in oxidizing liquid water and at the same time reducingcarbon dioxide to carbohydrate, is the overpowering source of oxygen in the presentatmosphere.At any time after the primitive, the oxygen balance will be determined by :PLUS : photochemical dissociation of H2O ; photosynthesisMINUS: 0 2 and 0 3 oxidation of surface materials; decay and respiration; and0 2 in H20 solution.As Holland13 points out, the rise of oxygen occurs as a consequence of a smalldifferential between two much larger opposing effects, which relatively speaking,increase together.As the initial oxygen from photosynthesis is released, moreover,it will simply substitute for oxygen from photodissociation of H20, because of theinherent " Urey " regulation to the level 0 2 < 10-3 P.A.L.This balance is illus-trated qualitatively in fig. 9.PHOTOCHEMICAL NG PHOTOSYNTHETICRODUCTION OF 0 2+ NTRlBUTlNGOXYGEN-SED OXIDATION. RESPIRATION,, AND H 2 0 SOLUTIONFIG. 9.-Factors in early oxygen balance.Not until the net rate of production of oxygen, primarily by photosynthesis,exceeds the net rate of 0 2 dissociation, and its consequent loss as an active oxidant,can equilibrium values of oxygen exceed the levels in the primitive atmosphere andbuilding of a stable oxygenic atmosphere begin.This leads to inquiry concerning the ecology for the rise of photosynthesis.Caspersson,s2 and Davidson,Q have shown that cell absorption of u.-v. arises fromabsorption by nucleic acids primarily between 2600 and 2700& and by proteinsbetween 2700 and 2900A.Cell absorption in these bands is highly lethal to cellfunction in all forms, degrading chemical function, and stopping growth, repro-duction and survival. Only atmospheric ozone can provide protection by shadowingthe lethal radiation in these bands (see fig. 10). The distribution of ozone is shownin idealized form in fig. 11. Ozone is distributed roughly uniformly in a coluinnbetween its level of maximum production and the surface, to which it is convectedand lost. The level of maximum production is lowered as the oxygenic coacen-tration diminishes132I023 ww7OXYGENIC CONCENTRATION IN EARTH’S ATMOSPHEREwavelength (A)[(l erg cm-2 sec-1 (50 &-I)].lO.-Thickness of 0 3 required to absorb available u.-v.to “ extinction ” (1800-3000dlo-’FIG. 11.Estimated distribution of ozone for various levels of oxygenL. V. BERKNBR AND L. C. MARSHALL 133Therefore, the total path length of ozone will diminish with oxygenic concentra-tions as shown in table 2. Combining these data on path lengths of ozone, pathlengths of oxygen previously derived, and the path lengths for absorption in liquidTABLE 2.*-ESTIMATED TOTAL PATHLENGTHS OF 0 3 FOR VARYING PRESSURES OF 0 2x, P.A.L. -H, cm % of atmos. (s.t.p.) -x cm10.0 6 5 ~ 105 7 x 10-8 0.51.0 4 7 ~ 105 7 x 10-8 0.330.1 2s x 105 7 x10-8 0.20.01 1 2 ~ 105 4 x 10-8 0.050*005 1 2 ~ 105 1 . 6 ~ 10-8 0.020~001 1 2 ~ 105 4 ~ 1 0 - 9 0.005*The crude integration of total path length established in this table is all that present datacoupled with basic assumptions will justify.water using the measurements of Dawson and Hulburtp yields fig.12, whichshows the penetration of u.-v. in liquid water, in the presence of various oxygenicatmospheres.wavelength (A)FIG. 12.-Path length in liquid water in presence of 0 2 and 0 3 for various concentrations of 0 2to absorb available u.-v. to " extinction " [l erg cm-2 sec-1 (50 &-I].Here it is seen that in the primitive atmosphere, lethal radiation penetrates to adepth of 5 to 10m of water. Therefore, the requirements for primitive photo-synthesis are as follows :(a) a water depth more than about 10 m, sufficient to shadow lethal radiation but nodeeper than-needed, to permit a maximum of visible light for photo-synthesis 134 OXYGENIC CONCENTRATION IN EARTH'S ATMOSPHERE(b) a minimum of liquid convection to avoid circulation of primitive metazoatoward the lethal surface, but gentle convection to provide organic nutrientssynthesized photochemically at the surface, in the presence of u.-v., accordingto the processes described by Miller.32This rigidly restrictive ecology describes bottom dwelling organisms (greenalgae or their evolutionary precursors) in protected shallow lakes or seas.Inparticular, life in the oceans seeins unlikely.Warm pools associated with volcanic hot springs, rich in nutrient minerals andelemental compounds seem prime candidates for the origin of life and photosyn-thesis. The ancient bioherms, 2.7 billion years old, and known to have supportedphotosynthetic organisms, from the work of Hoering and Abelson,55 could wellhave been at the base of such pools, and must be very close to the seat of life.Therigid ecologic insulation between such pools suggests the possibility of multipleorigins of living organisms with natural selection among them only at a later erawhen the permissive ecologic environment became more general.Only as the continents grow with volcanic action can sufficient areas for photo-synthesis at suitable densities of activity be found to meet the criterion for a growingoxygenic atmosphere. With constzntly changing geographic areas, and corres-ponding fluctuations of climatology, the growth of the oxygenic atmosphere appearsto have awaited the proper combination of conditions to satisfy the critical criterion.Considering the lowered levels of light energy below lethal depths of water, we estim-ate that photosynthetic activity at about present densities must have covered between1 and 10 % of present continental areas before an oxygenic atniospherc could be built.As oxygen finally rises toward the level, 02-0.01 PAL.( k , 1 % present con-centration) several interesting potentialities arise :(i) the penetration of lethal u.-v. (replotted in fig. 13) diminishes to a few cmof water, opening the oceans to life;(ii) the oxygenic level reaches the " Pasteur point " where organisms change fromfermentation to respiration. Thus the energy available for chemo-synthesisjumps from - 20 to - 675 cal/mole (cf.Genevois 56 and Rabinowitch 28).(iii) At this same oxygenic concentration, many primitive organisms changeform anaerobic " photoreduction " to " photosynthesis " through rapidoxidation of an hydrogenase enzyme, thus broadening the base for evolu-tionary activity.28This leads to a search of paleontologic and geologic history for a radical andexplosive change in evolutionary forms, corresponding to the opening of entirelynew and far more widespread evolutionary opportunities as 02-+0.01 P.A.L. Thereis just one such evolutionary explosion-the Cambrian-beginning 600 millionyears ago." By immediate inference we therefore identify the oxygenic level,02--+0.01 PAL. as immediately preceding the opening of the Cambrian, followingthe earlier suggestion of one of the authors.58Prior to the Cambrian there is no evidence in the fossil record of any form oflife advanced beyond the elementary algae, fungi and bacteria, i.e., the simplestforms of thallophyta (cf.Rutten 24). Since Proterozoic sediments favourable tofossil preservation have been diligently studied for more than a century, the completeabsence of fossils representing more advanced forms prior to the Cambrian evolu-tionary explosion has been considered heretofore as a scientific " puzzle " (cf.Kummel 57).* This dating follows the most recent geologic and geo-chronologic conclusions (cf. Kummel57)135The usual assumption is that evolutionary pre-Cambrian precursors could havehad only " soft " parts that were unfavourable for fossilization (although in sub-sequent ages, fossils of this general kind are not infrequent) 57.Under the interpretation dictated by our present model, no advanced precursorsto the Cambrian evolutionary explosion should be expected until sufficient oxygenicconcentrations presaged the opening of the Cambrian.Thus, according to themodel developed in this discussion, the geologic record should be read exactly aspresented in nature.L. V. BERKNER AND L. C . MARSHALL1400 r10-3 10-2 10-1 I 10fraction of P.A.L. of oxygenFIG. 13.Penetration of u.-v. in liquid water with various combinations of oxygen and ozoneatmospheres [(intensity at extinction = 1 erg cm-2 sec-1 (50 %.)-')I.Following the opening of the Cambrian, the complexity of life is known to havemultiplied rapidly.In the next few million years more than 1200 species of differentcreatures appeared, many of very considerable size and variety of characters.During this time the foundations for all modern phyla were laid. In particular,complex and efficient forms of respiratory apparatus were evolved independentlyamong various phyla as increasing oxygen levels presented favourable opportunitiesfor selection of such evolutionary advances. These advanced respiratory systemsprovided the mechanistic basis for the concurrent development of circulatory systems,digestive tracts, central nervous systems, bisexual modes of reproduction, andsimilar characters associated with advanced biological organisms through effectivesupply of oxygen and removal of oxidized carbon.As oxygen rises with widespread marine photosynthesis to the level of 0.1 P.A.L.(10 % present concentration), fig.13 shows that lethal radiation will for the firsttime be largely shielded from dry land. This will open a new ecological niche forevolution ashore. Recent evidence indicates the possibility of microscopic palyno -logic organisms on land as early as mid-Silurian, but the geological record showsno evidence of any " advanced " form of life ashore until the late-Silurian age, 420million years ago. Then a number of different phyla of plants and animals ex-ploded on dry land. By the Early Devonian, 30 million years later, great forestshad appeared, and soon after amphibian vertebrates were found ashore136 OXYGENIC CONCENTRATION IN BARTH’S ATMOSPHEREThe late Silurian is interpreted by immediate inference as the earliest stage atwhich plants could emerge above the surface without danger of lethal “ sunburn ”.Therefore we identify the period 420 million years ago with an oxygenic level,0 2 N 0.01 P.A.L.The explosion of evolution ashore increases photosynthesis in astep function by some 20-25 %, again tilting the oxygen balance radically towardthe plus side.In examining the oxygen balance in light of all factors studied to date, therelation between production rate and final equilibrium may be represented crudely,by fig. 14. Much more refinement of this balance is necessary in future studies.Sx 10-4 E-lt : i I1441 I I I I II o9 1010 1011 1012 1d30 2 rate of production by photosynthesis(mol.cm-2 sec-1)FIG. 14.-Estimated atmospheric levels of oxygen as a function of production rates.Following the late Silurian, hgh rates of photosynthesis are induced withoutcorresponding quantities of organic materials immediately available ashore fordecay and replenishment of C02. This suggests that oxygen may have “over-swung ” the present level to a somewhat higher value as the lush life of the Carbon-iferous developed. Then, with reduction of CO2, the Earth would cool, due to lossof the “ greenhouse ” effect of CO2, leading to the ice ages of the Permian Period.As the Earth cooled, photosynthesis would sharply fall, leading to a radical lossof atmospheric oxygen.Thus, the phase difference of production of 0 2 and of 6 0 2suggests that the levels of these two atmospheric components in the post-Silurianatmosphere niust have been unstable, the instability being damped by the ever-improving adaptation of organisms to wider environmental ecologies. Pendingmore analytical study, a preliminary estimate seems justified of the order of 108years for a complete oscillation above and below the present quasi-permanent levelL. V. BERKNBR AND L. C. MARSHALL 137We must not expect complete symmetry of this function, however, since any dropof oxygen is likely to arise from a series of inter-related events involving adaptationof oxygen-producing organisms, events whose mutual interaction would force anunstable and precipitous drop of oxygenic levels. A model may therefore be devisedfrom the studies to date of oxygenic levels over geologic history as shown in fig.15.FIG.15.-Tentative qualitative model of growth of oxygen in atmosphere.The standard assumptions made in historical geology and paleontology con-cerning rates of evolutionary development are modified so profoundly, by theconclusions drawn from this model, as to require further examination of the evolu-tionary process.In the conduct of this study, it has been deemed necessary to consider evolutionin two separate roles :(0(ii)the character and geographic extent of living organisms which must be themajor contributors to the rise of the oxygenic atmosphere at any periodbeyond the primitive ;the identification of a period of explosive evolutionary change as an indicatorof the timing of critical oxygenic levels in the atmosphere, e.g., as an in-dicator of major and appropriate physical change of the environment.In connecting these two roles, the evolutionary process during any geologicalperiod has been interpreted as an intricate interaction between the oxygenic levelgenerated by living organisms, and the natural exploitation of that level for the newevolutionary niches which that oxygen level opens to evolution.As such evolu-tionary opportunities are captured, the oxygen level is then raised as a consequence,thereby permitting a new round of evolutionary development. From time to time,critical levels are reached that permit vast physical ecologic opportunities, due tocritical responses of organisms generally.There appears implicit in the literature an assumption that very long periodsof time (that is long compared to 10 million years) are required to account for theevolutionary rise of organisms from the simple and microscopic metazoa of Hoeringand Abelson 55 and of Tyler and Barghoorn 59 to the large and somewhat morehighly organized creatures suddenly appearing at the opening of the Cambrian138 OXYGENIC CONCENTRATION IN EARTH’S ATMOSPHEREIn the absence of other evidence, such an assumption might appear reasonable,though as previously observed, the absence of any Proterozoan links in the geologicrecord (in spite of favourable conditions and diligent search) has long been recog-nized as a “ puzzle ”.* Early in the current studies it became apparent that thelevels of oxygen preceding the critical levels were not suficient to permit a smoothrise in evolutionary character, but that “ quantum jumps ” in ecological activityfollowing achievement of critical oxygenic levels seemed more likely.Thus, thesudden opening of the whole oceans to population as a consequence of sufficient shadow-ing of lethal u.-v. represented such a jump. As an interesting circumstance, theoxygenic level for opening the oceans to life seems to coincide approximately withsignijicant biological processes intimately related to the same oxygenic level, thusmultiplying the probability of revolutionary opportunity for the change in the charactersof life observed at the opening of the Cambrian.Likewise, the opening of the dry-land areas to land-dwelling plants and animals at oxygenic levels sufficient to shadowthe dry land from lethal u.-v. represents a similar ecologic jump during late Silurian.The questions then arise: how fast can evolution develop in complexity in itscapture of the new environmental opportunity? Are the great evolutionary dis-continuities to be properly interpreted as clear signals of great physical ecologicchange?The fact is that large changes in evolutionary character of living organisms arefrequently observed in geologic history as rather sudden jumps without a “ long ”history of visible precursors.60 This suggests that the absence of visible precursorsarises from the very speed of change-a change occurring in a small and limitedpopulation poorly adapted to the revised ecology, and proceeding rapidly from gener-ation to generation under the encouragement of natural selection in a vast newlyopened physical-eco1ogy.t In developing this interpretation it would seem that theadvance in the evolutionary character of life does not necessarily await some*The idea of a long pre-Cambrian history (i.e., long compared to a few tens of millions ofyears) of advanced evolution of organisms to account for the diversification found at the base ofthe Cambrian, a history which has been thought not preserved due to the imperfection of the fossilrecord, is deeply imbedded in the whole literature of geology and paleontology. Consequentlydirect inferences from new evidence that lead to contrary views are not easy to accept.Never-theless, we believe these inferences should be faced squarely, since the classic views are purelyintuitive, and therefore must be taken advisedly, however Ptolemaic their authority. Critical ap-proach to the present interpretation in light of the oxygenic evidence can focus sharper attention onsearch for pre-Cambrian and lower Cambrian evidence, and may clarify and quantify the evolutionaryprocess.t While pre-storage of evolutionary characters has long been known, it has recently come sharplyto the attention of science as a major factor in evolution through the effects of the widespread useof antibiotics and industrial chemical poisons.Except for the most extreme poisons, organismsfind preadaptation to widely varied ecologic changes in some small fraction of their population,which permits their evolved propagation in the new ecology. The molecular basis for geneticpreadaptation is mentioned by Huxley 61 and typically discussed by Anfinsen.32 The wholesubject of the molecular basis for evolution has undergone such drastic expansion and revision duringthe past decade, as a consequence of the scientific emphasis on molecular biology, that many previousconceptions in evolution bear re-evaluation.The term “ preadaptation ” has unfortunately been misused, during the early discussion ofnatural selection, to refer to some form of supernatural or mythical, non-scientific predestination,whereby the organism was “ divined ” toward an ever higher, preordained “ design ”.This viewwas imposed by some who could not conceive of natural selection from natural variability as thedirecting mechanism in producing the delicately balanced features of evolutionary change, a viewthat is now wholly discredited (cf. Simpson ; 62 Romer 63). Preadaptation does, however, have auseful connotation in the strictly scientific sense, and in complete accord with current evolutionarytheory, when used to refer to storage and reproduction of mutational events, some of which becomeuseful to natural selection when the environment is modified. The problem is : how large a rangL. V. BERKNER AND L. C. MARSHALL 139mutational accident while the ecological opportunity lays fallow. Rather, in theface of continual mutation (indeed, rapid mutation in terms of geologic periods),a wide variety of evolutionary characters are continually pre-stored among theseveral members of the population, and await selection by appropriate ecologicopportunities that are provided by the opening of new evolutionary niches.Geneticvariation of the individual in the population involves many unexpressed physiolo-gical traits. A few of these have the opportunity to respond in a few members ofthe population through their favourable articulation when a new and favourableecology permits or encourages that response through natural selection. Thenevolution proceeds as rapidly as combination, selection and adaptation permit,since current rates of mutation are not a limiting factor.Evolution through naturalselection appears as a process of rapid scanning in times compared to a few millionyears for selection of ecologies offering improved adaptation, and selection of suchecologies will occur quickly after they are offered. Thus, explosive changes in com-plexity of similar evolutionary characters simultaneously among many differentphyla appears as positive evidence for the opening of new and major ecologicopportunities. Large physical-ecologic changes are then most favourable for" quantum jumps " in evolution since not only larger opportunities are opened,but the pre-existing competition may be reduced or eliminated.This reasoning does not contend that a metazoan could pre-store all of thespecialized characteristics of " a man ".Clearly, in achieving such levels of com-plexity, natural selection must proceed through many millions of steps in develop-ing suitable opportunities upon which mutation can act and from which subsequentselection must proceed. The suggestion here is merely that bccause of pre-storageof characters, the speed of evolution can be enhanced sufficiently to account formajor jumps in levels of complexity such as occurred at the opening of the Cambrianor in the Late Silurian of the order of a few units of 10 million years. Precedingsuch jumps, widespread evidence of intermediate precursors would not and couldnot exist. Long series of precedent evolutionary steps should not be expected orassumed since they would be forbidden by the precedent environment.In conclusion, a comment on the atmosphere of Mars seems in order.Becauseof the low gravitational field, coupled with its smaller diameter and mass, Mars hasprobably lost all H2, He and should be losing atomic oxygen at a rate determinedby its temperature and rates of production and of diffusion or convection. In absenceof oceans, the atmosphere of Mars would otherwise appear to be somewhat similarto the primitive atmosphere of the Earth, well prior to the Cambrian. Life onMars therefore may be subject to the same restrictive ecology found for the primi-tive atmosphere of the Earth. As seems probable for the primitive Earth, some-what different forms of life, arising from different initial precursors in separate andof variability, unexpressed in the current environment, can be prestored as a result of recessivemutations, to be released as a useful evolutionary change in a sharply modified environment?It can be readily shown from the mathematics of genetics, for example, that a " lethal recessive "gene inhabiting 10-6 of a population will be reduced only 50 % in 106 generations when the proba-bility of mating among the population is purely random.Not all, nor even a large proportion ofrecessive genes are lethal, nor are many " recessive " genes purely recessive. Consequently, anyindividual may contain a relatively large number of hidden or only partly evident characters, usefulfor selection in a modified ecology. If some number n of independent inherited traits can bearticulated in adapting to a new environmental opportunity, the probability of an individualpossessing such an articulated group will be 2-n.Thus, for example, the probability of articulationof any random group of 20 independent inherited characters with an individual would be 10-6.Since populations of micro-organisms are large, very considerable evolutionary changes can beexpected rather quickly under compulsion of a strong selective ecologic change (cf., Dobzhansky 64)1 40highly insulated pools, might be expected. The direct study of Mars and its atmo-sphere should open exciting new and powerful vistas to science.The study of paleoatmospheres involves a tremendous range of correlative in-formation, and this prCcis can only present the briefest account of much more ex-tensive analysis.The generation of a model for the Earth opens vistas to a widerange of studies relating to quantitative paleoclimatology. A wide range of cir-culatory and convective patterns is suggested as the level of ozone production riseswith increasing oxygen or concentration at different eras, and of paleoionospheresresulting from widely differing aeronomys. The purpose of this paper is to encouragemore critical studies and discussions of the subject, and to approach the muchgreater refinements that are immediately suggested.OXYGENIC CONCENTRATION IN EARTH’S ATMOSPHERE1 Goldschmidt, Norske Videnskaps-akad. Oslo, Skr., Mat.-Nat. KI., 1937, 4, 148.2 Brown, Rev. Mod.Physics, 1949, 21, 625.3Brown, The Atmospheres of the Earth and Planets, ed. Kuiper (The University of Chicago4 Spencer-Jones, Sci. Progress, 1950, 38, 417.5 Kuiper, Proc. Nat. Acad. Sci., 1951,37, 1.6 Urey, The Planets: Their Origin and Development (Yale University Press, 1952).7Urey, Symp. Int. Union Biochem. Moscow, 1957, (The Macmillan Company, New York,8 Alfvkn, On the Origin of the Solar System (Oxford Clarendon Press, 1954).9 Fesenkov, Symp. Int. Union Biochem., Moscow, 1957 (The Macmillan Company, New York),10 Vinogradov, Symp. Int. Union Biochem., Moscow, 1957 (The Macmillan Company, New11 Rubey, Bull. Geol. SOC. Arner., 1951, 62 (2), 111.12 Rubey, Geol. SOC. Amer. (special paper), 1955, 62, 631.13 Holland, Petrologic Studies: A Volume to Honour A.F. Buddington (Princeton University,14 Sapper, Vulkankunde (Stuttgart, Engelhorns, 1927).15Verhoogen, Amer. J. Sci., 1946, 244, no. 11, 745.16 Bullard, Volcanoes, in History, in Theory, in Eruption (University of Texas Press, 1962).17 Wilson, Amer. Sci., 1959, 47, 1.18 Wilson, in The Solar System, vol. I1 of The Earth as a Planet, ed. Kuiper (University of Chicago19 Wilson, Physics and Geology, with Jacobs and Russell (McGraw-Hill Book Company, New20 Tazieff and Fabre, Compt. Rend., 1960, 250, 2482.21 Rankama, Geol. SOC. Amer. (special paper), 1955, 62, 651.22 Ramdohr, Abhandl. deut. Akad. Wiss. Berlin KI. Chem., GeoIogie und Biologie, 1958, 35,23 Lepp and Goldich, Geol. SOC. Amer. Bull., 1959, 70, 1637.24 Rutten, The Geological Aspects of Origin of Life on Earth, Amsterdam, New York (Elsevier25 Oparin, Origin of Life (Dover Publications, Inc., New York), 1953, S213.26 Bernal, Proc. Physic.SOC. A , 1949, 62, 537.27 Wald, in The Phy.sics and Chemistry ofLife, ed. Simon and Schuster, New York, 1955, part 1,28 Rabinowitch, Photosynthesis and Related Processes (Interscience Publishers, Inc., New York,Press, 1952), p. 258.1959), 1, 16.1959, 1, 9.York, 1959), 1, 23.Princeton, New Jersey, 1962), p. 447.Press, 1954), p. 150.York, 1959).no. 3, p. 19.Publishing Co., 1962).chap. 1 .1951).29 Calvin and Bassham, The Photosynthesis of Carbon Compounds (Benjamin, Inc., New York,1962).30 Calvin, Symp. Int. Union Biochem. (Macmillan Company, New York), 1959, 1, 207.31 Calvin, Bull.Amer. Inst. Biol. Sci., 1962, 12, no. 5, 29.32 Anfinsen, The Molecular Basis of Evolution (John Wiley and Sons, 1961).33 Miller, Symp. Int. Union Biochem. (Macmillan Company, New York, 1959), 1, 123.34 Florkin, Symp. Int. Union Biochem. (Macmillan Company, New York), 1959, 1, 503, 578.35 Abelson, Ann. N. Y. Acad. Sci., 1957, 69, 276141 L. V . BERKNER A N D L . C. MARSHALL36 Hill, Proc. Roy. SOC. B, 1939, 127, 192.37 Hill and Scarisbrick, Proc. Roy. SOC. B, 1940, 129, 238.38 Hall, Damon and Hinteregger, 3rd Int. Space Sci. Symp. (Washington, 1962).39 Detwiler, Garrett, Purcell and Tousey, Ann. Geophysics, 1961, 17, 263.40 Nawrocki and Papa, Geophysics Corp. Amer. (Bedford, Massachusetts), A.F.C.R.L. Report,41 Johnson, J. Meteorol., 1954, 11,431.42 Wilson, Astrophys. J., 1963, 138, 832.43 Watanabe, Zelikoff and Inn, Geophysical Research Papers, no. 21 (June, 1953), A.F.C.R.C.44 Watanabe, Adv. Geophysics, 1958, 5, 153.45 Allen, Astrophysical Quantities (University of London, The Athlone Press, 2nd ed., 1963), p. 124.46 Herzberg, Molecular Spectra and Molecular Structure. I. Spectra of Diatomic Molecules(D. Van Nostrand Company, Inc., New York, 1961).47Pearse and Gaydon, The Identification of Molecular Spectra, 2nd ed. (John Wiley & Sons,Inc., New York, 1950).48 Pauling, The Nature of the Chemical Bond and the Structure of Molecules and CrystaZs (CornellUniversity Press, Ithaca, New York, 1960).491rzt. Crit. Tables (McGraw-Hill Book Company, Inc., New York, 1926-30).50 Nicolet and Mange, J. Geophys. Research, 1954, 59, no. 1, 15.51 Nicolet and Bates, J. Geophys. Res., 1950, 55, no. 3, 301.52 Caspersson, Cell Growth and Cell Function: A Cytochemical Study (Norton and Co., Inc.,53 Davidson, The Biochemistry of the Nucleic Acids, 4th ed., (John Wiley & Sons, New York,54 Dawson and Hulburt, J. Opt. SOC. Amer., 1934,24, 175.55 Hoering and Abelson, Proc. Nat. Acad. Sci. US., 1961, 47, 623.56 Genevois, Beochemisches Zeit., 1927, 186,461.57 Kummel, History of the Earth (W. H. Freeman and Co., 1961).58 Berkner, Proc. Conf. on Ionospheric Physics (July, 1950), part B. Published by GeophysicsRes. Div., Air Force Cambridge Research Centre, Cambridge, Massachusetts, pp. 13-20(Apr. 1952).Contract # AF 19(604)7405 (August, 1961).Technical Report # 52-23 (Cambridge, Mass.).New York).1960).59 Tyler and Barghoorn, Science, 1954, 119, 606.60 Romer, Vertebrate Paleontology (University of Chicago Press, 1962).61 Huxley, Evolution in Action (Harper and Brothers, New York, 1953).62 Simpson, The Meaning of Evolution (Yale University Press, 1960).63 Romer, The Vertebrate Story (University of Chicago Press, 1959).64 Dunn and Dobzhansky, Heredity, Race and Society (The New American Library, New York,1957)