How Complete Is the Geologic Column?

"You cannot go out and see the geologic column because it does not exist anywhere."
Richard Bliss, Gary Parker and Duane Gish, Fossils:  Key to the Present, Institute for Creation Research, 1980¹

"You just saw the whole column piled up in one place where one oil well can drill through it."
Glenn Morton, on a North Dakota well, 2001²

"The stratigraphical record is a lot of holes tied together with sediment."
Derek Ager, The Nature of the Stratigraphical Record, 1981³

This article asks how much of the geologic column is present at one place, and how much of it is rock rather than gap.  It is about rock and time.  A sister article, How Complete Is the Fossil Record?, asks a different question:  how much of past life left fossils.  The two meet where gaps in the rock cut short the known ranges of fossils.  This article covers the rock side of that meeting.

Summary

"Completeness" means two different things, and they have to be kept apart.  The first is how many of the column's systems, such as the Cambrian or the Jurassic, are stacked at one place.  That count needs no ages at all.  The second is how much of the time that the mainstream timescale assigns to a place is represented by rock, and how much by gaps between layers.  That count is made in the mainstream's own years, and this article reports it only as the mainstream's figure.

Seven conclusions, stated first.

1. The claim is older than its best-known source. The line that the column "does not exist anywhere" appears in an Institute for Creation Research school text of 1980, by Richard Bliss, Gary Parker and Duane Gish.¹ Henry Morris repeated it in 1982: the only place to see the column is "in the textbook!"⁴ Scott Huse's book The Collapse of Evolution, first published in 1983, says in its 1986 printing that the column "exists only in the minds of evolutionary geologists."⁵ The often-quoted wording, "a fiction, existing on paper only", is a paraphrase by the website TalkOrigins, not Huse's words.
2. Most places do hold only a few systems. John Woodmorappe found in 1981 that under 1 % of the land holds all ten of the systems he counted, and two-thirds holds five or fewer.⁶ A recount on Macrostrat, a public database of 1,402 regional columns, gives 0.8 % and 70 %.⁷ On this point the creationists are right.
3. But the complete column does exist, in single boreholes. In 2001 Glenn Morton listed the formation tops of one North Dakota oil well that passes from Tertiary rock at the surface to Precambrian basement at 14,945 feet.² A public U.S. Geological Survey dataset of 16,200 Williston Basin wells confirms it independently. In 101 wells the survey picked tops in nine systems, from Cretaceous down to Cambrian, and 95 of those wells start in Tertiary bedrock. The tenth system, the Permian, lies in an interval the survey did not split.⁸ Woodmorappe himself conceded in 1999 that Morris and Parker "are not strictly accurate".⁹
4. One part of Woodmorappe's charge holds up at that very site. The Williston Basin's "Triassic" is a red-bed unit with no fossils to date it. A North Dakota state geologist wrote in 1992 that "Paleontological data is lacking" to settle whether those beds are Permian, Triassic or Jurassic.¹⁰ So at the best-known complete site the Triassic, and perhaps the Permian, is placed by position rather than by its own fossils. Morton lists fossils from most of the other systems in the same wells.²
5. The ocean floors hold only the top of the column. Of 132 deep-sea drill sites that Macrostrat records as cored to or near basement, none holds Triassic or older sediment except one on the continental margin off Ireland, which bottoms in Devonian sandstone. Two reach the Jurassic and 29 hold every system from the Cretaceous up.¹¹
6. By the mainstream's own clock, most time at any place left no rock. Joseph Barrell argued in 1917 that the breaks between beds add up to most of geologic time.¹² Peter Sadler's 1981 compilation of nearly 25,000 measured rates showed why the figure depends on the scale of measurement.¹³ Andrew Miall finds that less than 10 % of the 4.86 million years assigned to one Utah rock group is represented by sediment.¹⁴ Chris Paola and colleagues put one sandstone at no more than two millionths of its assigned time.¹⁵ These are the mainstream's figures, on its own timescale.
7. Neither kind of completeness tests the order of the column or a young earth by itself. A gap is not a reversal. Both models expect most places to be incomplete, and both expect deposition to be continuous somewhere at any moment. What separates them is how long each gap lasted, and that is decided inside the gaps: by soils, hardgrounds, caves and minerals. Those tests belong to Flat Gaps and its sister articles.

The verdict in two sentences.  The claim that the column exists nowhere is false, and its young-earth authors have conceded it.  The claim that most places hold only part of the column is true, but it does not distinguish the models, and the mainstream itself says most of its time at any place is gap, which shifts the argument from how much rock exists to how long the gaps took.

Part 1: The young-earth claim, in its own words

The claim began in a school text.  The earliest printing found for this article is Fossils:  Key to the Present, a 1980 science text for schools from the Institute for Creation Research.  Its authors were the educator Richard Bliss, the biologist Gary Parker and the biochemist Duane Gish.  They wrote:  "You cannot go out and see the geologic column because it does not exist anywhere.  All real rock layers include gaps and even reversals from this perfect sequence."  Their example was the Grand Canyon, where "The Ordovician … and Silurian … systems are missing from this group."  They added a thickness argument:  "If the entire geologic column were found in one place, it would be as much as 210 kilometers (130 miles) deep."¹

The same page grants something important.  "The geologic column is not a column of rock; it is an idea.  But it is an important idea, because it does show a trend for rock layers or strata to be found in a vertical sequence."¹  So the claim, from the start, was about completeness, not about order.

Morris and Parker, 1982.  Parker repeated his wording in What Is Creation Science?:  "Now the geologic column is an idea, not an actual series of rock layers.  Nowhere do we find the complete sequence."⁴  Henry Morris, in his own chapter of the same book, gave it its best-known form:  "There is only one place in all the world to see the standard geologic column.  That's in the textbook!"⁴  He set a standard column "at least 100 miles thick" against a local average of "about one mile".⁴

Huse, 1983.  Scott Huse's The Collapse of Evolution drew on Bliss, Parker and Gish and pushed the claim further.  "It is important to realize that nowhere in the world does the geologic column actually occur.  It exists only in the minds of evolutionary geologists."⁵  The TalkOrigins Archive's index of creationist claims gives Huse's claim as "The geological column is a fiction, existing on paper only", citing page 15.¹⁶  Those words are not on page 15 of the printing read for this article, which carries the sentences quoted above.  The TalkOrigins wording is a paraphrase and should not be quoted as Huse's.

Woodmorappe, 1981:  the claim with numbers.  John Woodmorappe's paper in the Creation Research Society Quarterly is titled "The essential nonexistence of the evolutionary-uniformitarian geologic column."  He mapped rocks assigned to ten systems, Cambrian through Tertiary, on a grid over the continents, and counted how many systems each part of the land holds.⁶  His result is the core of the young-earth case and is examined in Part 2.

Morris, 1974:  the claim about time.  The second kind of completeness has its own young-earth statement.  In Scientific Creationism Morris wrote that the evolution model needs "extensive time gaps in the column, when no deposition was occurring", while "The creation model, on the other hand, must interpret the column in terms of essentially continuous deposition".¹⁷  His evidence was that "there is no worldwide unconformity!"  He concluded:  "In other words there are no time-breaks; the record is continuous."¹⁷  Morris accepted local gaps.  His argument was that every local gap, followed far enough sideways, passes into rock that fills it.

The modern young-earth geologists.  Later writers kept the observation and dropped the strong form.  Andrew Snelling of Answers in Genesis writes that the complete sequence "is not found in any one place or region on the earth's surface, even though much of it is present in some places, as can be seen in the northern Arizona-southern Utah area."¹⁸  Steven Austin's Grand Canyon book of 1994 describes the surface where the Ordovician and Silurian are missing as having "About 30 feet of channeling" and asks whether 100 million years is "justified by the physical evidence".¹⁹  Timothy Clarey and Davis Werner of the Institute for Creation Research explain the same missing rock as non-deposition:  "we suspect the Tippecanoe is missing here primarily due to non-deposition, with only minimal erosion marking the boundary between megasequences.  We infer that the tsunami waves of the Flood simply did not reach this point during deposition of the Tippecanoe, for some unknown reason."²⁰

So the young-earth literature makes four claims.  No place holds the whole column.  Most places hold little of it.  The column as a whole is an idea, not a rock body.  And the record is continuous, with no real time breaks.  The first two are about rock, the fourth is about time, and the third depends on the first.

Part 2: How many systems lie at one place?

Woodmorappe's count.  Woodmorappe drew rocks assigned to each of ten systems on an equal-area grid of the continents, mostly from a Soviet atlas.  He found that "slightly less than 1% has all 10 periods simultaneously in place", that "66% has 5 or less of the 10 present", and that "only 14% has 8 or more geologic periods represented at all."  The Cretaceous covered the most land, "just over 51%", and the Triassic the least, 33 %.⁶

The same count on modern data.  Macrostrat, a public database kept at the University of Wisconsin, holds 1,402 regional stratigraphic columns on land.  The parent article scored each column for Woodmorappe's ten systems, counting only layered rock and leaving out granites and other intrusions, as Woodmorappe did.⁷

The two counts agree within a few points, forty-five years apart, on different data.  Most of the land holds only a few systems.  On this the young-earth writers are right, and nothing below changes it.

What a regional count cannot show.  A Macrostrat column is a composite for a region of several thousand square kilometres.  Its units may lie side by side within the region rather than stacked at one point.  Woodmorappe's grid squares have the same weakness.  So a regional count can overstate completeness at a point, and it says nothing about whether the systems in a region lie in order.  To test "nowhere", one needs a single vertical section:  a cliff, a core or a well.

Woodmorappe's own concessions.  Woodmorappe never claimed that no place holds all ten.  His 1981 paper says:  "There apparently are regions on earth where all ten geologic periods can be found superposed", and names the Holy Cross Mountains of Poland, where "All ten geologic periods are undoubtedly represented".^{6 6}  In 1999, answering Morton, he went further.  "Yes, Morris and Parker, whom Morton attacks, are not strictly accurate when they say there is no place on earth where all ten geologic systems are superposed."⁹  He listed the places his 1981 map already showed, the Williston Basin among them, and wrote that he had never set out to prove the column absent:  "The truth is that I already knew it didn't!"⁹

What the count assumed.  Woodmorappe's 1981 study did not test the order of the systems.  He explained in 1999:  "I assumed that, where the two systems occur in the same geographic locality, Ordovician strata always overlie Cambrian strata (and never the reverse)."⁹  His Table 2 lists only how often each system rests on each older one, with no column for a system resting on a younger one.⁶  And every square on his maps was assigned to systems by the standard correlation;  his Polish column counts "sedimentary lithologies that are unambiguously biostratigraphically dated".⁶  A study that counts systems has already granted that fossils can tell one system from another.

The thickness argument.  In 1999 Woodmorappe shifted the claim from systems to thickness.  Even where all ten are present, "the column, as represented by sedimentary-thickness, is mostly missing."⁹  The 100 to 200 miles of Morris and Parker is, in his words, "the sum of the thickest sections from the field of each of the ten Phanerozoic systems".⁹  Morton answered with an analogy.  Adding the snowiest place's snowfall for every day of the year gives a total no single town receives, yet every town still had a full year of weather.²  Thick sections pile up where the crust sinks, and different basins sank in different systems.  No geologist predicts that one place holds every system at its greatest thickness.  Its absence tests nothing.

Part 3: One well at a time in the Williston Basin

Morton's well.  Glenn Morton was an oil-industry geophysicist who had published more than twenty papers in the Creation Research Society Quarterly before he gave up young-earth creationism.²¹  His 2001 essay on the TalkOrigins Archive answers Morris and Parker with one well in McKenzie County, North Dakota:  the W. H. Hunt Trust Estate Larson #1, "drilled to 15,064 feet deep."  It "begins in the Tertiary Fort Union Formation."²  He printed its formation tops.  The well passes the Cretaceous Greenhorn at 4,910 feet, the Jurassic Rierdon at 6,690, the Triassic Spearfish at 7,325, the Permian Opeche at 7,740, the Pennsylvanian and Mississippian between 7,990 and 10,255, the Devonian from 11,085, the Silurian Interlake at 12,539, the Ordovician from 13,250 and the Cambrian Deadwood at 14,445, and reaches the Precambrian at 14,945.²  "You just saw the whole column piled up in one place where one oil well can drill through it", he concluded.²

Morton's tops are his own report, and the state's well file was not read for this article, because North Dakota sells its formation tops by subscription.  So the claim was tested on a different and public body of data.

The U.S. Geological Survey's picks.  In 2023 S. E. Gelman and B. G. Johnson of the Survey released the data behind a three-dimensional model of the Williston Basin.  It includes formation tops "picked in 16,200 wells from the Williston Basin of North Dakota, South Dakota, and Montana", 141,096 picks in all, made on the wells' geophysical logs.²²  The release lists which system each pick belongs to.  It was counted for this article well by well.⁸

The result:

So 95 single boreholes pass from Tertiary bedrock through Cretaceous, Jurassic, Triassic, Carboniferous, Devonian, Silurian, Ordovician and Cambrian rock.  The Carboniferous is present as both its Mississippian and Pennsylvanian parts.  The Permian is the one system the Survey did not pick on its own.  It lies inside the interval from the top of the Spearfish to the top of the Pennsylvanian Tyler, which also holds the lower Spearfish and the Pennsylvanian Amsden.  That interval is 134 to 1,331 feet thick in the 101 wells, median 892 feet.  The Survey's own model names the Permian layers in it, the "Lower Permian Opeche Shale" and the "Permian Broom Creek Formation," and Morton's well lists the Opeche at 7,740 feet.²²  The tenth system is therefore present in these wells too, though this dataset cannot show it pick by pick.

Morton's 2001 claim stands on public data that he did not use.  The complete column, as Woodmorappe and Morris and Parker defined it, lies under western North Dakota and eastern Montana, in single wells, in the standard order.

Three limits on the count.  First, a missing pick is not a missing rock unit.  The Survey did not pick every top in every well, and "in many other cases the shallow intervals were not logged."²²  The count of complete wells is therefore a floor.  Second, the Tertiary is read from the state geologic map at each wellhead, not from the well.  Third, the count does not test order.  Picks on a log are made from the top down by tracing each marker from well to well, so a list of picks is in order by how it was made.  The Survey's table has 698 neighbouring picks out of depth order, none by more than 16 feet, all where a formation thins to nothing.  That tests the table, not the rocks.

Where Woodmorappe's narrower charge holds.  Woodmorappe's 1999 paper made a second, narrower point:  "Locally, many of the systems have not been assigned by the index fossils contained in the strata but by indirect methods that take the column for granted".⁹  At the Williston Basin that charge is partly right.  The Spearfish Formation is a body of red sandstone and shale with gypsum and salt.  The first geologist to name it, N. H. Darton in 1899, assigned it to the Triassic with no stated reason, and an early survey of it reported "No fossils found."²³  Clarence Carlson of the North Dakota Industrial Commission wrote in 1992 that the same red beds have been called "Triassic or Permian-Triassic" in North Dakota and "Jurassic-Triassic or Jurassic" in Saskatchewan and Manitoba, and that "Paleontological data is lacking for resolution of this problem."¹⁰  The Survey's 2023 release calls the Spearfish "Permian-Triassic".²²  The Permian beds below it are red beds and salt too.  Morton reports Permian pollen in the Opeche salt, citing a 1984 study that was not read for this article.²

So at the best-known complete site, the Triassic is placed by its position, under marine Jurassic beds and over the Permian, and by tracing the red beds to their outcrops in the Black Hills.  Its exact age is unsettled.  That should be said plainly.  Most of the other systems in these wells are marine carbonates, shales and sandstones, and Morton lists fossils from many of them, from Ordovician graptolites and corals to Jurassic clams and forams.²  Woodmorappe's narrow charge touches one or two systems out of ten here.  It does not turn the column into an idea.

The basin is not complete everywhere.  Of the 152 wells that reach the Deadwood, 23 lack a Pennsylvanian Tyler pick and others lack Jurassic or Silurian picks.  Some of that is the Survey's choice of what to pick, and some may be real thinning toward the basin's edges.  The same basin holds the whole column at its centre and less of it at its rim.  Completeness is a property of a place, and it changes within a single basin.

The other basins.  Morton also listed 25 more basins, from Libya to China and Australia, where "the entire geologic column is found".²  His source was a proprietary industry compilation, the Robertson Group's Stratigraphic Database of Major Sedimentary Basins of the World (1989).  Woodmorappe could not obtain it, and neither could this article.  The list is Morton's report and is not relied on here.  The TalkOrigins index adds the Bonaparte Basin of northern Australia, citing a Western Australian geological memoir that was not read for this article.¹⁶

Part 4: Under the oceans

Woodmorappe's 1999 paper points out that if the ocean floors are counted, "the global figure falls to less than 0.4%."⁹  That is right, and ocean drilling shows why.

Macrostrat includes 132 active deep-sea columns, each a drill site that the database describes as "completely cored from sea floor into or near basement rocks."¹¹  Counted with the same rule as the land columns:

Twenty-nine sites hold every system from the Cretaceous up, and two hold every system from the Jurassic up.  The one Devonian entry is Deep Sea Drilling Project Site 548 on the Goban Spur, a piece of continental crust off southwest Ireland, where Macrostrat records Late Devonian quartzite at the bottom of the hole.  No site on true ocean floor holds anything older than the Jurassic, and none holds a Paleozoic sequence.  The ocean floors carry only the top of the column.

Even the drilled oceanic record has gaps.  Ager noted in 1981 that "we now know from the Deep Sea Drilling Project that there are vast gaps in the record of the oceans, too."³  On Macrostrat's own compiled ages, the sedimentary units at the median deep-sea site fill 79 % of the time between the site's oldest and youngest sediment, and a quarter of the sites fill 55 % or less.²⁴

This pattern is shared ground.  The mainstream reads it as ocean floor that is continually made at ridges and destroyed at trenches, so that none is older than the Jurassic.  Catastrophic Plate Tectonics, the leading young-earth Flood model, accepts the same observation and makes all of today's ocean floor during the Flood.  Either way, an incomplete column under the oceans is expected, and it does not test the order of the systems on land.

Part 5: How much of the time is rock?

The second kind of completeness asks a different question.  Take a place whose rocks the mainstream timescale spreads over some span of years.  How much of that span is represented by rock, and how much passed with no rock left behind?  Every number in this Part is the mainstream's own, worked out on its own timescale.  None of them is evidence for that timescale.  They are reported because creationists quote them, and because they show what the mainstream itself expects a column to look like.

Barrell, 1917:  the gaps outweigh the beds.  Joseph Barrell of Yale argued that besides the obvious breaks, "the breaks of smaller time interval are still more numerous and may add up to equally large measures of time unrecorded by sedimentation."  He named them:  "The name selected is diastem.  Diastems range of all values from seasonal cessations of sedimentation to those which approach geologic epochs in duration."¹²  His model section looked "not unlike those found in nature; yet only one-sixth of the time is recorded in it".¹²  Barrell also wrote the sentence that matters most for the Flood debate:  "a long time interval may be represented by a short columnar section, and yet the individual beds may be deposited rapidly."¹²  Fast beds and long gaps were part of the mainstream picture from the start.

Ager, 1973 and 1981:  "More Gaps than Record".  The British geologist Derek Ager, often quoted by creationists for his catastrophism, titled a chapter "More Gaps than Record".³  "We are only kidding ourselves if we think that we have anything like a complete succession for any part of the stratigraphical column in any one place", he wrote.³  His picture of the record was "one long gap with only very occasional sedimentation", and "a lot of holes tied together with sediment."³  The popular phrase "more gap than record" is a paraphrase of that chapter title.

Sadler, 1981:  why the answer depends on the ruler.  Peter Sadler compiled "nearly 25,000 rates of sediment accumulation" and found them "spanning at least 11 orders of magnitude", with "a systematic trend of falling mean rate with increasing time span."  He traced that trend mainly to "unsteady, discontinuous sedimentation."¹³  The trend is now called the Sadler effect.  Its lesson is that completeness has no single value:  "since sedimentation is unsteady, its continuity can be meaningfully specified only in terms of some stated time scale", and "all sedimentary stratigraphic sections must contain hiatuses at some scale."¹³  A column can look complete at a resolution of millions of years and be mostly gap at a resolution of days.

The same effect in Macrostrat.  The point can be seen in the database used in Part 2.  Each Macrostrat unit carries the ages its compilers assigned to it.  At that coarse resolution, the median land column's layered units fill 57 % of the span from its oldest unit to its youngest, and the middle half of columns fill between 31 % and 83 %.  Measured against the whole of the time from the Cambrian on, the median column fills 15 %.²⁴  These are upper limits:  any gap shorter than one compiled unit is invisible at that resolution.

Finer rulers give smaller numbers.  Andrew Miall of the University of Toronto studied the Mesaverde Group of Utah, whose span the mainstream fixes at 4.86 million years by its ammonite zones.  Applying modern rates for the same settings, "less than 10% of the 4.86 m.y. elapsed time represented by the group can be accounted for."  The rest "is 'missing' at bedding planes and other bounding surfaces of all types".¹⁴  Many of those surfaces show nothing:  "The lengthy intervals between each sheet have not left any identifiable signature, such as mature paleosoils, or evidence or early cementation … or of deep erosion."¹⁴  Chris Paola and four colleagues worked one example to the scale of single dune crossings.  A 20 m river sandstone of 100 cross-bed sets, laid down by dunes whose faces each passed "under 4 h," represents under 400 hours, so "the temporal completeness is, at most, 2 × 10−6."¹⁵  Their summary:  "recorded time is best thought of as a kind of dust".¹⁵

Where the mainstream puts the missing time.  Paola and his colleagues say that "most of the missing time has not been eroded out but rather represents periods of inactivity or stasis."  They also reject the idea that the record is mostly giant catastrophes:  "paleotransport estimates suggest that this is not generally the case: The stratigraphic record is strangely ordinary."¹⁵  And they note that time is always being recorded somewhere:  "as long as net sediment supply to the basin continues, deposition must be occurring somewhere, and so at the basin scale, time must be recorded somewhere as well".¹⁵

The argument among sedimentologists.  In 2023 Angela Hessler called the whole idea of incompleteness a fallacy:  "Stratigraphy is not missing anything."²⁵  Miall replied that "Incompleteness is a fact, but it is measurable, quantifiable, and explainable."²⁶  This is a quarrel about words.  Neither writer doubts that the Mesaverde took millions of years, and neither proposes less time.  Both stand about six to seven orders of magnitude from a one-year Flood.

How creationists use these figures.  Michael Oard read Miall's study as support for the Flood, concluding that "the claimed millions of years are imaginary."²⁷  Miall concluded the reverse.  He kept the 4.86 million years, fixed by ammonite zones, and put more of it into the gaps.  The figures in this Part say that, on the mainstream's own account, rock is a small fraction of time.  They do not say the time did not pass.  Whether it passed has to be read from the gaps themselves, which is the subject of Part 6.

Part 6: What each model predicts, and which fits better

Completeness by itself does not decide between the models.  Both expect most places to hold only part of the column, and both expect some places to hold all of it.  But the two models give different reasons for the gaps, and some of those reasons can be checked.

1.  Where should the column be most complete?  The mainstream answer is in Barrell.  In the "rapidly subsiding center of a geosyncline", his old word for a deep basin, "a larger proportion of the time would be recorded than on its landward margin."¹²  The young-earth answer is Clarey and Werner's:  rock is missing where "the tsunami waves of the Flood simply did not reach".²⁰  Both point to the same places, the deep basins.  The Williston result in Part 3 fits both.  A basin centre that holds all ten systems while its rim holds fewer is what each model expects.

2.  How widespread should each system be?  On the Macrostrat land columns the Cretaceous and Paleogene are the most widespread systems, each in about half the columns, and the Permian and Triassic the least, each in about a fifth.⁷  Clarey's model has the water rise through the early Flood, cover "the highest hills" by Day 150 and fall afterwards, so the packages laid down near the peak, such as the Zuni with its Cretaceous rock, should cover the most ground.²⁸  The mainstream reads the same pattern as high sea levels in the Cretaceous.  Again both fit.  Clarey and Werner add that "the Tippecanoe has the least surface coverage and the least sediment volume across the globe compared to the other five megasequences".²⁰  A megasequence is not a system, so the Macrostrat system counts cannot test that directly.  In North America the Ordovician and Silurian, which lie mostly in the Tippecanoe, are each present in more columns than the Permian, Triassic or Jurassic.⁷  Whether that conflicts with Clarey's ranking depends on his unpublished per-package counts.

3.  Are the missing systems eroded away, or never laid down?  Here the young-earth side has a fair point from mainstream data.  Shanan Peters and Jon Husson found that most surviving sedimentary rock, about 75 % of it, was laid down in shallow seas on the continents and "does not exhibit any steady decrease in quantity with increasing age."²⁹  So the missing systems mostly mark places where little was deposited, not the remains of a once-complete column worn away.  Clarey's "non-deposition" fits that, and so does the mainstream account of rising and falling seas.  The difference lies in how long the non-deposition lasted.

4.  What do the surfaces at the missing systems look like?  This is the question that can separate the models, and it is argued at length elsewhere.  At the Grand Canyon, where the Ordovician and Silurian are missing, Austin himself describes "About 30 feet of channeling" cut into the Muav Limestone below.¹⁹  His Flood reading is that "the Flood waters got shallower and bottom-current velocity increased, allowing soft sediments to be eroded in channels as sedimentation stopped."¹⁹  Other gaps are flat, like the base of the Chattanooga Shale, which in the Tennessee area alone rests on 23 different formations, at an angle "rarely as much as 1°."³⁰  Flat Gaps weighs the clocks that can be read on such surfaces, from limestone dissolution to glauconite growth, and finds that every surface priced so far with a measured rate needed far more than a year.  Buried Soils, Bioturbation and Trace Fossils and Evaporites and Salt Domes test the other features found inside a column.  This article does not repeat those arguments.

5.  Is the record continuous, as Morris said?  Morris predicted that there are "no time-breaks; the record is continuous", because no unconformity is worldwide and every local gap, followed far enough, passes into rock.¹⁷  Part of this is right.  Paola and colleagues agree that at the scale of a basin "time must be recorded somewhere".¹⁵  Megasequences and the Flood Sequence finds that the boundaries between Sloss's great rock packages spread over tens of millions of years of the mainstream timescale and show no resolvable gap at the median column for three of the five boundaries.  So Morris was right that the big gaps are not single worldwide events.  But rock that fills a gap somewhere else does not fill it here.  Where a surface at one place lacks a system that lies on it in the next basin, that surface still has to be explained, and how long it lay open is the question Flat Gaps takes up.

6.  Pricing a complete column in one Flood calendar.  A complete column in one well means that one place received every system's rock within the Flood year, in order.  Clarey and Werner place the Sauk, Tippecanoe and Kaskaskia packages "in the first 40 Days of the Flood".²⁸  In the 101 complete Williston wells, the rock from the top of the Cambrian Deadwood to the top of the Mississippian, which is all Tippecanoe and Kaskaskia, has a median thickness of 5,393 feet, about 1,644 m.  On Clarey's calendar that is at least 41 m of rock a day, or 1.7 m an hour, before counting the Deadwood.⁸  That rate alone is not impossible for a flood.  The difficulty is what the 1,644 m contains:  the evaporites of the Devonian Prairie Formation, carbonates with burrows and mud cracks, and more than a dozen named formations, each with its own top surface.  Morton made the same kind of argument in 2001.²  Those features are priced in the articles linked above, not here.

Which fits better?  On the count of systems, neither model is favoured:  both predict the pattern, and both explain a full Williston Basin and a bare shield.  On whether the gaps are mostly never-deposited rather than eroded, the data favour a reading both sides can hold.  On the duration of the gaps, the evidence found so far, examined in the linked articles, favours the mainstream, and that is where the debate actually lies.  Completeness moves the argument there.  It does not settle it.

Objections and responses

"Morton's well proves the creationists wrong, so the Flood is refuted."  It refutes one sentence, that the column exists nowhere, and the young-earth authors who count conceded that sentence in 1981 and 1999.  A complete column at one place does not show how long it took.  The well's value is narrower:  it shows that the column's order is a physical stack at a point, not only a chart assembled from scattered pieces.

"The well picks assume the column, so counting them is circular."  The picks are formations traced on logs from well to well.  Their system labels come from the fossils of the same formations where they are cored or crop out, and Woodmorappe's own 1981 count relied on labels of exactly this kind.  The charge does fit one or two systems in the Williston:  the Spearfish has no fossils to date it, and its age is disputed.  That is conceded in Part 3.  It has not been shown for the fossil-bearing marine formations of the same wells, several of whose fossils Morton lists.

"Even where all ten systems are present, most of the column's thickness is missing."  That compares one place with the sum of the thickest sections of every system from all over the world, a pile no geologist expects anywhere.  Morton's snowfall analogy makes the point:  "Antarctica received a full year's worth of snowfall--it is just a smaller amount than Vail, Colorado."²

"Most places hold only two to five systems, so the column is a construct."  The count is right and is printed in Part 2.  But a gap is not a reversal.  The column's claim is that where two systems are both present, the older lies below, and a count of missing systems does not test that.  Woodmorappe's study assumed it.

"If most time is gap, the time never happened."  The mainstream figures in Part 5 are computed on the mainstream's own timescale and only describe its expectation.  Miall kept his 4.86 million years and moved more of it into the gaps.  Whether a gap lasted a day or a million years has to be read from the gap:  from weathering, hardgrounds, borings, caves, minerals and the contacts' geometry.  That evidence is weighed in Flat Gaps and does not favour a year.

"Ager was a catastrophist, so his gaps support the Flood."  Ager held that "the history of any one part of the earth, like the life of a soldier, consists of long periods of boredom and short periods of terror."³  The terror lays down the beds;  the boredom is the gaps.  His catastrophes are spread across the whole timescale.  His "one long gap with only very occasional sedimentation" is a statement about long time, not short.

"Even the sedimentologists disagree about incompleteness."  Hessler and Miall disagree about the word.  Both accept the same gaps and the same millions of years, and both stand six to seven orders of magnitude from a one-year Flood.

"The Macrostrat counts are regional and overstate completeness."  That objection comes from the other side, and it is fair.  A regional column can hold systems that lie side by side rather than stacked.  The well count in Part 3 answers it for the Williston Basin, where the stack is in single boreholes.  Elsewhere, the regional counts should be read as upper limits for any one point.

"Ocean drilling shows that most of the column is missing worldwide."  It shows that the ocean floors hold only the Jurassic and younger systems.  Both the mainstream and catastrophic plate tectonics expect that, since both make today's ocean floor young relative to the continents.

Issues and unanswered questions

1. Morton's well was not checked against the state's file. North Dakota sells its formation tops. The USGS picks confirm the pattern independently, but the Larson #1 tops as printed are Morton's. A North Dakota Industrial Commission subscription, or the well's scout ticket, would settle them.
2. The USGS picks do not split the Permian. The Permian lies in an unpicked interval in every one of the 101 wells. A per-well test of the Permian needs the state's Opeche and Broom Creek tops.
3. The Williston Permian and Triassic are poorly dated. Carlson's 1992 abstract says fossils are lacking to settle the red beds' age. Morton's report of Permian pollen in the Opeche salt rests on a 1984 study not read for this article. How each Williston system is fossil-dated, formation by formation, is not set out in one source found here.
4. Morton's 25 basins rest on an unread industry database. No one on either side has quoted the Robertson Group's compilation. Each basin would need its own published stratigraphy to be counted as a single-point complete column.
5. The deep-sea count rests on Macrostrat's compilation. The one Devonian entry, Site 548 on the Goban Spur, was traced to its Macrostrat unit but not to the Deep Sea Drilling Project's own site report.
6. No per-package global coverage is published by either side. Clarey and Werner say the Tippecanoe covers the least area of the six megasequences worldwide. Their counts are not public, and Macrostrat's system counts cannot test a claim made in megasequences.
7. No young-earth model has published where complete columns should be found. Clarey's "the tsunami waves of the Flood simply did not reach this point … for some unknown reason" is an explanation after the fact. A Flood model that predicted, before drilling, which basins hold all ten systems would be a real test.
8. The mainstream figures in Part 5 depend on its timescale. They cannot be used as evidence for long ages, only as a statement of what the mainstream expects. The independent test is inside the gaps, and that work is incomplete: no clock has yet been read on the Grand Canyon's Muav-Redwall contact itself, where Clarey's calendar allows days or weeks.

Verdict

The young-earth claim that the geologic column "does not exist anywhere" is false.  Woodmorappe conceded it, and public U.S. Geological Survey data show the full stack of systems in about a hundred single wells in western North Dakota, with the one caveat that the Williston Triassic is dated by position rather than fossils.  The claim that most places hold only a few systems is true, and the young-earth writers who measured it measured it well.  But neither claim tests the column's order or its age, and the mainstream itself holds that most of the time at any place is gap.  So the completeness argument, on both measures, moves the debate to a different question:  how long the gaps took.  That question is answered inside the gaps, and the evidence there is weighed in the articles linked above.

Sources