This page exists so that nobody has to take the article on trust. Every source it rests on is listed here with what it actually measured, who paid for it, and which way it cuts.
248 sources were catalogued for this article and added to the site's source library. This page covers the ones that carry weight, grouped by the question they answer. The complete machine-readable catalogue, including the ones that only confirmed a detail, is in the library index under the topics noahs-ark-logistics, ark-naval-architecture, ark-kinds-and-counts, ark-feed-water-metabolism, ark-construction-methods, ark-critics, and ark-outside-the-ark.
How to read an entry
Measurement or assumption. The most important distinction on this subject. A manure production rate from ASAE D384.1 is a measurement made on real animals. A "safety index" computed from a finite-element model whose element types cannot represent the failure mode is not. Both get used in this debate as though they were the same kind of thing, and separating them is most of the work.
Who funded it. Stated wherever known. The Korean seaworthiness study is real naval architecture by real naval architects and it was fully supported by a creationist society. Both halves of that sentence matter and neither is a reason to dismiss it.
Direction. Whether the source supports, undercuts, or complicates the Ark-feasibility case. Assigned by what the source shows, not by what its authors believe. Several mainstream papers are tagged supports and several creationist works are tagged undercuts.
The biblical specification
A Hebrew and English Lexicon of the Old Testament
Brown, Driver and Briggs. Oxford: Clarendon Press, 1906.
type: lexicon camp: mainstream direction: complicates
The entries that matter are gopher (p. 172), kopher (p. 498), tsohar (p. 844) and neshamah (p. 675). BDB says gopher is dubious and reports a proposal that it is a copyist's corruption. Its gloss of kol neshamah as "every breathing thing" is what undermines the argument that insects are excluded because they lack nostrils — the phrase is an idiom for living creatures, not an anatomical claim.
Verified by grepping the full Westminster Leningrad Codex (23,213 verses) rather than by repeating the hapax claims: gopher occurs once, kopher once in this sense, singular tsohar exactly once, and nishmat ruach chayyim once.
The Book of Genesis, with Introduction and Notes
S. R. Driver. Westminster Commentaries, 10th ed. London: Methuen, 1916.
type: commentary camp: mainstream direction: mixed
The source of the continuous-opening reading of the tsohar: "a kind of casement running round the sides of the ark… a little below the roof" (p. 88). This is the reading the ventilation calculation requires, and it matters a great deal that it comes from a mainstream nineteenth-century critical commentator rather than from creationists looking for air. Driver also supplies the concession that kopher is probably a Babylonian loanword (p. 87).
A Critical and Exegetical Commentary on Genesis
John Skinner. International Critical Commentary. Edinburgh: T. & T. Clark, 1910.
type: commentary camp: mainstream direction: undercuts
Skinner calls the tsohar clause "unintelligible as it stands." He also confirms BDB's distributive reading of qinnim qinnim ("cell upon cell") against Philo's loculos loculos and a Palestinian Syriac lectionary. Cite him for the honest statement that the key clause is obscure.
Genesis 1–15
Gordon J. Wenham. Word Biblical Commentary 1. 1987; Zondervan reissue 2014.
type: commentary camp: evangelical direction: undercuts
The most damaging single sentence in the exegetical literature, and it is from a conservative: the instructions are "much too short to serve as a practical blueprint for construction" (p. 173). Wenham also translates qinnim as "reeds", following B. Jacob, G. R. Driver, McEvenue and the NEB — which, if right, removes internal compartments from the text altogether. He calls the tsohar clause "the most obscure remark in the flood story." And he relays Cassuto's observation that 300, 50 and 30 are all sexagesimally tidy numbers, which is a real alternative explanation for why the proportions are so round.
The Siloam Tunnel Inscription: Historical and Linguistic Perspectives
Gary A. Rendsburg and William M. Schniedewind. Israel Exploration Journal 60 (2010): 188–203.
type: measurement camp: mainstream direction: undercuts
The only Iron Age Hebrew length checkable against a surviving structure: 1,200 cubits stated for a tunnel measuring about 533 m, giving 0.4442 m. This is the defensible cubit, and adopting it costs the Ark 58% of its volume relative to the figure Answers in Genesis uses. Caveats to print: the numeral 1,200 is partly restored on the stone, is almost certainly rounded, and Ussishkin disputed the 533 m denominator in 1976.
Naval architecture and the hull
Safety Investigation of Noah's Ark in a Seaway
S. W. Hong, S. S. Na, B. S. Hyun, S. Y. Hong, D. S. Gong, K. J. Kang, S. H. Suh, K. H. Lee and Y. G. Je. Creation Ex Nihilo Technical Journal 8(1):26–36 (1994). Translation of a 1993 Korean conference paper.
type: model camp: creationist funder: Korea Association of Creation Research direction: mixed
The single most-cited and most-misreported source in Ark apologetics. It is genuine work by naval architects at a genuine institution, and the seakeeping half of it is respectable. Its structural half is not usable for the purpose it is put to.
What it actually did: compared thirteen hull forms, generated by scaling the Ark itself, on structural safety, overturning stability, and seakeeping. What it did not do: name a wood species, state an elastic modulus, state a numerical allowable stress, or publish the wave bending moment it computed. Its structural finite-element model represents frames as truss elements and planking as membrane elements at rigid nodes (p. 31), which cannot represent joint slip, fastener working, plank sliding, or seam failure. Its entire structural conclusion is the unfalsifiable sentence on p. 32.
The "30 metre waves" headline is a linear extrapolation roughly four times beyond its own data, all of which stops at a significant wave height of about 11 m. And the Ark placed second, not first, on the combined index.
Feedback: Was Noah's Ark Seaworthy, or Is That Impossible?
Tim Lovett. Answers in Genesis, 2013.
type: reanalysis camp: creationist direction: undercuts
Valuable because it is Answers in Genesis correcting the Answers in Genesis line. Re-running Hong's numbers, Lovett places the Ark seventh of thirteen in seakeeping, sixth in strength, and fifth in stability. Cite this whenever anyone claims the Ark had "optimal" proportions.
Lovett's 2007 piece is also where the useful concession lives: "The Ark is near the maximum size that is known to be possible for a wooden vessel." That article additionally contains an error worth not repeating — it calls the tessarakonteres a "trireme," which is a contradiction in terms.
Ancient Ship Structures: Ultimate Strength Analysis of Wooden Joints
Albert Zamarin, Smiljko Rudan, Davor Bolf, Katarina Lucchini and Irena Radić Rossi. Journal of Marine Science and Engineering 13(8):1392 (2025).
type: measurement camp: mainstream direction: mixed
The best empirical support the Ark's hull case has, and the source of its sharpest problem. Finite-element analysis of reconstructed ancient joints measures an Uluburun-type mortise-and-tenon seam at 84% of the bending strength and 80% of the tensile strength of an unjointed plank. That is a genuinely strong joint.
The same paper measures ancient keel scarf joints at 10% to 40%. A 137 m keel needs at least six scarfs and each is a hinge. The excellent seam number is attached to the wrong joint.
Wood Handbook: Wood as an Engineering Material
Forest Products Laboratory, USDA Forest Service. FPL-GTR-190, 2010.
type: measurement camp: unaffiliated direction: supports
The material property source for every stress calculation in the article. Table 5-3a: green white oak, modulus of rupture 57 MPa, shear parallel to grain 8.6 MPa. The ratio between those two is why shear, not bending, is the binding check.
The Age of Titans: The Rise and Fall of the Great Hellenistic Navies
William M. Murray. Oxford University Press, 2012.
type: review camp: mainstream direction: undercuts
The most damaging mainstream source in the file, and it comes from the leading scholar of exactly the giant ships creationists cite as precedent. Murray puts the safe limit for an all-wood seagoing hull at 70–75 m (p. 173, citing Sleeswyk and Meijer 1994), which is about half the Ark's length, and he states it knowing all about the "forty." His escape hatch for the Hellenistic giants is that they were double-hulled and were siege platforms rather than seagoing warships (p. 176). He also reserves judgement on Athenaeus's reliability altogether (p. 181).
Ships and Seamanship in the Ancient World
Lionel Casson. Johns Hopkins reprint of Princeton 1971, with addenda.
type: review camp: mainstream direction: undercuts
Kills two of the standard precedents outright. Of the Syracusia: "Moschion, unfortunately, gives no dimensions." Of the largest wooden ship ever built, the Great Republic at 334 ft 6 in: "the size was made possible by the use of iron braces" (p. 185, n. 8). Casson also reads the "forty" as "a mammoth version of what today is called a catamaran" (p. 110).
Also relevant: the free Yonge 1854 translation of Athenaeus that everyone quotes has "gunwale" for akrostolion and "triremes" for "fours," and both errors inflate the creationist case. Use Casson's renderings or the Loeb.
The large ships of antiquity
Larry Pierce. Creation 22(3):46–48 (2000).
type: popular camp: creationist direction: undercuts
The origin of the entire creationist precedent list, and it does not hold up. Pierce himself writes "We are not given the dimensions of this ship" about the Leontifera, then derives 120–150 m by multiplying rowers per tier by an airline-seat spacing of three feet. Downstream repetitions kept his number and dropped his caveat. Pierce is also the translator of the Master Books edition of Ussher, which is where the "Leontifera" entry entered creationist circulation with its date garbled from 280 BC to 280 AD.
The Six-Masted Schooner Wyoming
International Marine Engineering 15(1):1–2 (January 1910). With the 1910/1921/1923 Record of American and Foreign Shipping (American Bureau of Shipping) and contemporary newspaper coverage of her loss via Chronicling America.
type: measurement camp: unaffiliated direction: mixed
The ship that is supposed to be the strongest single counter-example to the Ark, and the primary record does not support the way it is usually told.
Her hull is 350 ft over all, 329.5 ft on the register, 304 ft on the keel, 50.1 ft beam, 30.4 ft depth of hold. The 450 ft in circulation — including in this site's own notes until now — is jib-boom tip to spanker-boom tip. "Between perpendiculars" is also the wrong name for 329.5 ft; that is registered length under the Act of 6 May 1864, a different definition.
She was not cheaply built. ABS-supervised, iron-strapped and salted, classed A1 for sixteen years — the longest term the Bureau granted any wooden vessel — opened for survey at Bath in 1916 and passed, still classed in 1923. She was lost at anchor on Nantucket Shoals in thirty feet of water, loaded with 6,000 tons of coal, in a March 1924 north-easter. That is a stranding, not a hull failure at sea.
The "leaked continuously, ran pumps" claim about this ship traces to an uncited Wikipedia sentence and should never have been printed. What is sourced is the class-level statement from Parker: "all the large schooners leaked badly and were forced to depend heavily upon their steam pumps." The real measured hog belongs to the Edward J. Lawrence, "an exceptionally well-built schooner… crowned a full thirty inches on her keel after only nine years in service."
Do not cite HAER ME-1 for any Wyoming construction detail. It is a seven-page shipyard nomination form containing one sentence about the ship, and it misnames a second vessel elsewhere.
The six-masted schooners as a class
W. J. Lewis Parker, The Great Coal Schooners of New England (1948), and Paul C. Morris, The Four-Masted Schooners of the East Coast and its companion appendices.
type: compilation camp: unaffiliated direction: supports
The class survival record is better evidence than any single anecdote, and it favours us. Of the nine wooden six-masters: five lost by grounding, two by fire, two foundered — and the two that foundered were the two oldest, at 19 and 20 years. Average life 13.2 years. The steel seven-master Thomas W. Lawson lasted five years. What killed the type was lee shores, shoals, collision and fire, not the hull girder, and Parker — no friend of the type — says so himself.
Parker also supplies the mechanism that does not transfer to a box: the schooners hogged because of "length without corresponding depth of hold," fine and heavily-weighted ends, ground tackle forward and overhanging sterns aft, so that "the ends therefore lacked the buoyancy which supported their long flat midsections." A rectangular unrigged hull has none of those features.
Report on the Ship-Building Industry of the United States
Henry Hall. Tenth Census of the United States, 1880. Washington: GPO, 1884.
type: measurement camp: unaffiliated direction: mixed access: public domain
The primary that should replace Casson as the citation for the Great Republic, and it is free. Hall gives the actual iron: "The frames on the inside were strapped with iron by braces 4 inches wide, 1 inch thick, and 36 feet long, there being 90 straps on each side of the ship" (p. 89).
That 90-straps figure circulates misattributed to the Wyoming. It is the Great Republic's. Hall also disposes of the "no tree is 137 m long" objection in one sentence: "There was little timber in the Great Republic longer than 50 and 60 feet, but the oak was in much shorter lengths."
On the great strength given to Ships of War by the application of Diagonal Braces
Robert Seppings. Philosophical Transactions of the Royal Society of London 108 (1818): 1–8. With his 1814 paper, Phil. Trans. 104: 285–302.
type: measurement camp: unaffiliated direction: supports access: public domain
A controlled full-scale experiment, and the best single datum in the naval file. The Nelson, St Vincent and Howe were 120-gun ships of the same form, dimensions, frames, beams and planking scantlings, differing only in diagonal bracing. Launch hog: 9½ in, 9¼ in, and 3⅝ in (p. 6). Construction method changed longitudinal behaviour by a factor of 2.6 at fixed length and fixed scantlings.
This also corrects an error this site made in its own retraction list. Seppings' riders were timber, not iron, and he credits a wooden bridge for the idea, so diagonal bracing is a geometry available to any carpenter rather than a nineteenth-century metallurgy. The claim to avoid is that Noah did it; the claim that survives is that "you cannot build a hull that long without iron" is false.
A Manual of Naval Architecture
W. H. White. 3rd ed. London: John Murray, 1894.
type: review camp: unaffiliated direction: mixed access: public domain
The Royal Navy's Director of Naval Construction, and the source that keeps the article honest in both directions.
Against us: White ranks the four classes of strain on a ship in order of importance and puts longitudinal hogging and sagging first, with strains from sail or steam propulsion third, calling the steady rig-induced racking moment "so small in amount as to be practically unimportant in its effect upon the ship considered as a whole." Removing the Ark's masts removes local damage, not the dominant load. He also states flatly that "in wood ships of large size the entire prevention of such working is practically impossible."
For us: on the Royal Navy's diagonally-planked wooden ships, three thicknesses with the two inner layers laid diagonally in opposite directions, "the skin thus formed possesses such superior strength to the skin of an ordinary wood ship that there need be comparatively little transverse framing above the bilges" — and the Victoria and Albert, built that way, "has continued on service for forty years with complete exemption from signs of weakness." The system was abandoned for cost, not performance. That is a wooden solution to the seam-shear problem this article identifies as binding.
How many animals
Mammal Diversity Database v2.5; IOC World Bird List 15.2; the Reptile Database; AmphibiaWeb
type: compilation camp: mainstream direction: background
Counted directly from the database exports in August 2026 rather than quoted: 39,586 living tetrapod species, 5,590 genera, 584 families. Every cell in the article's count table carries its database version and access date, because these numbers move.
Fossil quality and naming dinosaurs
Michael J. Benton. Biology Letters 4(6):729–732 (2008).
type: measurement camp: mainstream direction: complicates
51.8% of dinosaur species named through 2004 are invalid, with 24.3% nomina dubia. This is why extinct genus counts have to be printed as a range. It cuts both ways: a critic cannot demand the Ark carry every named genus, and a creationist cannot treat the fossil record as a clean inventory of pre-Flood life.
An amniote life-history database
Nathan P. Myhrvold, Elita Baldridge, Benjamin Chan, Dhileep Sivam, Daniel L. Freeman and S. K. Morgan Ernest. Ecology 96:3109 (2015).
type: compilation camp: mainstream direction: supports
21,322 amniote species with masses and life-history traits. The single most useful dataset in the article. Fitting juvenile mass against adult mass on this database is what produces the finding that mammals wean at a median 33% of adult mass, birds fledge at 83%, and reptiles hatch at 4% — retiring the unsourced "half of adult" rule in both directions at once.
Determining the Ark Kinds; Mammalian Ark Kinds; An Initial Estimate of Avian Ark Kinds
Jean K. Lightner et al. Answers Research Journal 4 (2011), 5 (2012), 6 (2013).
type: model camp: creationist direction: mixed
The methodological basis for the Ark Encounter's numbers, and it does not support the weight placed on it. The central rule — one documented hybrid merges two whole genera — is licensed in the 2011 paper by a coyote × red fox example given with no citation, unsupported by anything in the reference list, and for which no accepted hybrid exists (Canis 2n=78, Vulpes 2n=34).
The deeper problem is that the hybrid evidence the papers call their best evidence does not reach mammals. Hybridizable bird pairs average about 22 Myr of divergence and frogs 21 Myr, but placental mammals only 2–3 Myr. So AiG's 137 mammal kinds rest on the "cognitum" — visual similarity judgement — not on hybridization.
Noah's Ark: A Feasibility Study
John Woodmorappe. Institute for Creation Research, 1996. Not obtained; page references are relayed through Morton's review, and the book's own figures are taken from Woodmorappe's Impact summary.
type: model camp: creationist direction: mixed
Still the most detailed defense in existence and the source of nearly every number in circulation. Its weaknesses are structural rather than local: optimistic assumptions stacked in series, amphibians omitted as a class, and extinct Cenozoic mammals undercounted by about 2.5×. Redoing his own method with 2026 databases roughly doubles his answer, from about 16,000 animals to about 34,700.
ACQUIRE: the book is out of print on the second-hand market but still sold by ICR. Every page citation in this article's research files is secondary until a copy is in hand.
How Many Animals Were On Noah's Ark?
Ark Encounter / Answers in Genesis.
type: popular camp: creationist direction: undercuts
The headline 1,398 kinds and 6,744 animals has no published derivation. The extant Answers Research Journal papers sum to 628 kinds; the remaining 770 come from treating fossil families as kinds. The arithmetic implies exactly 329 "clean" kinds that nobody has listed anywhere. The spread between this estimate and Woodmorappe's is 27× on extant mammals alone.
Capacity, feed and water
Australian Standards for the Export of Livestock, version 3.3
Department of Agriculture, Fisheries and Forestry. Commonwealth of Australia, November 2023.
type: regulation camp: unaffiliated direction: supports
The most useful single source in the article and the closest real-world analogue to the Ark that exists: a binding regulation governing how much deck area an animal needs on a long sea voyage, with mortality data attached to the same regime.
Refitted, ASEL's own allowances come out at A = 0.03144 · W0.6601 with R² = 0.994 — that is, the regulation is to three decimal places the two-thirds surface-area law. Fitted across 234 rows drawn from ASEL, the NRC Guide, 9 CFR Part 3 and the EU directives, the transport regime gives A = 0.0263 · M0.673.
Food requirements of wild animals: predictive equations for free-living mammals, reptiles, and birds
Kenneth A. Nagy. Nutrition Abstracts and Reviews, Series B 71 (2001): 21R–31R.
type: measurement camp: mainstream direction: mixed
Field-measured feeding rates, and the correct replacement for the linear mass-scaling in this site's own notes. Solving Nagy's equations reproduces his stated result exactly: a 100 g mammal needs 12.0× and a 100 g bird 20.6× the daily energy of a 100 g reptile. That ectotherm advantage is a genuine strength of the Ark case and it is large.
Caveat to print: the equations carry 30–44% mean absolute error, so nothing downstream of them is good to better than a factor of two.
Aging, Maturation and Growth of Sauropodomorph Dinosaurs
Eva Maria Griebeler, Nicole Klein and P. Martin Sander. PLOS ONE 8(6):e67012 (2013).
type: measurement camp: mainstream direction: undercuts
Sauropod sexual maturity at 19 to 25 years. This is the price of the juvenile assumption: the animals that make the Ark fit are the animals least able to repopulate after it lands.
The Galapagos Tortoises in their Relation to the Whaling Industry
Charles Haskins Townsend. Zoologica 4(3):55–135 (1925).
type: measurement camp: mainstream direction: undercuts
The primary source behind the "tortoises survive a year without food or water" claim, and it says four to six months. Townsend's own word is "months." Correct the claim rather than repeating it — and stop offering tortoises as carnivore food, since they are Ark passengers.
Waste, air and labour
ASAE D384.1: Manure Production and Characteristics
American Society of Agricultural Engineers, 2003.
type: measurement camp: unaffiliated direction: undercuts
The canonical excretion table: 40–110 kg of fresh manure per tonne of live mass per day, by species. Fitting its eleven rows gives waste ∝ M0.96, not M0.75, because every animal in it is a production animal on a high ration. The instinct to discount to maintenance is checked and killed by ASEL's own shipboard feed standard, which implies about 55 kg per tonne per day for an animal merely standing in a pen.
At 55–65 kg/t/day, 800 t of animals over 370 days produce about 19,000 m³ — 45% of the hull.
Livestock Housing Ventilation: Natural Ventilation Design and Management for Dairy Housing
Tim Mescher and Michael Veenhuizen. AEX-113-95, Ohio State University Extension.
type: standard camp: unaffiliated direction: undercuts
The design rule that decides the tsohar question: 3.33% of floor area as the cold-weather minimum opening for a naturally ventilated single-storey barn. The Ark's continuous perimeter slot is 1.56% of its three-deck floor area — 0.47× the winter minimum. A single window is not in the same universe, failing by about 270×.
Marine Order 43 (Cargo and cargo handling — livestock) 2006
Australian Maritime Safety Authority.
type: regulation camp: unaffiliated direction: undercuts
20 air changes per hour required in enclosed livestock spaces. For the Ark that is 860,000 m³/h against 223,000 available from the perimeter slot at a 2 m/s breeze. This is one of three independent routes — the others being measured live-export practice at 50 m³/h per m² of pen, and a CIGR heat balance — that agree within a factor of two. Convergence of three unrelated methods is the best evidence the ventilation arithmetic is sound.
Crop and Livestock Labor Standards
Michael Langemeier and Kevin Dhuyvetter. Kansas State University Farm Management Guide MF-802, rev. 2005.
type: measurement camp: unaffiliated direction: undercuts
36 seconds per head per day for feedlot backgrounding with tractors; 54 for beef cows. Eight people on fourteen-hour days have 25 seconds per animal per day at 16,000 animals.
Broiler Production and Management
William A. Dozier, Michael P. Lacy and Larry R. Vest. University of Georgia Poultry Science Department, 2004.
type: measurement camp: unaffiliated direction: supports
The only published husbandry rate that clears the Ark's labour bar: "approximately 4 to 6 minutes per day per 1,000 broilers," or 0.3 s per bird per day. It exists because a broiler grower never touches an individual bird. The Ark's labour budget closes if and only if it is run on this model.
A review of Shipboard Mortality Data 1988–2017
R. Sadler et al. W.LIV.0295, Meat & Livestock Australia (Ausvet), January 2021.
type: measurement camp: unaffiliated direction: mixed
The mortality benchmark: 1.3% for Australian sheep over voyages averaging 15.6 days, falling to 0.17% by 2024; cattle 0.4% falling to 0.06%; and a clear duration effect, cattle rising from 0.071% at ten days or less to 0.338% past thirty. The Cormo Express lost 9.82% of 57,937 sheep over 80 days on a purpose-built ventilated carrier with a veterinarian aboard.
The most useful finding is the cause of death: inanition and salmonellosis account for 60–76%, not heat, ammonia or trampling. Good news for the engineering, bad news for the assumption that wild animals will eat.
The nineteenth-century transatlantic cattle trade, from the US Bureau of Animal Industry annual reports, is the best pro-Ark datum in the file: 1.6% mortality in 1891 falling to 0.31% by 1899 on two-week crossings, achieved with naturally ventilated sailing and early steam vessels and no machinery.
The critics
The Impossible Voyage of Noah's Ark
Robert A. Moore. Creation/Evolution XI (vol. 4 no. 1), Winter 1983, pp. 1–43.
type: review camp: counter-apologetics direction: undercuts
The canonical critique. Everything written against the Ark since is downstream of it, and it contains 63 distinct arguments. Several are excellent and several have not aged well. Read it before writing anything on this subject.
Note a correction to a common confusion: talkorigins.org/faqs/faq-noahs-ark.html is not Moore. It is Mark Isaak, Problems with a Global Flood, 2nd ed. (1998), a separate work. Moore's full text is free at the National Center for Science Education.
Problems with a Global Flood, 2nd edition
Mark Isaak. TalkOrigins Archive, 1998.
type: review camp: counter-apologetics direction: undercuts
The most-read modern statement of the case. Its capacity argument is the one that matters: adding a corridor allowance to Woodmorappe's own three space figures takes the animals past 90% of the ship.
Index to Creationist Claims, CH500–CH542
Mark Isaak, ed. TalkOrigins Archive.
type: review camp: counter-apologetics direction: undercuts
Worth reading mainly for what it reveals about our own side. Every claim from CH500 to CH505.8 attacks Ark-sighting material — Durupınar, Ron Wyatt, the Ararat Anomaly, Navarra's wood — that Answers in Genesis and Creation Ministries International have already retired in print. Conceding this up front costs nothing.
Review of John Woodmorappe's Noah's Ark: A Feasibility Study
Glenn R. Morton. TalkOrigins Archive, 1996.
type: review camp: old-earth-christian direction: undercuts
The most useful critique of Woodmorappe because it is page-numbered and because Morton is a Christian geologist rather than an atheist, so the argument stays technical. Source of the breakdown 7,428 mammals + 4,602 birds + 3,724 reptiles = 15,754 animals. Woodmorappe replied in 1997.
To the Ark, and Back Again?
Chad Arment. Answers Research Journal 13 (2020): 1–22.
type: review camp: creationist direction: undercuts
The most important concession in the creationist literature. Writing in Answers in Genesis's own journal, Arment calls the required post-Flood marsupial dispersal "not only unlikely, it is highly improbable," notes "if anything, this is a conservative measure," and dismisses the usual escape as "untestable and… simply God-of-the-gaps theorizing." Cite this rather than pretending biogeography is an open question within our own camp.
Outside the Ark
On the action of sea-water on the germination of seeds
Charles Darwin. Journal of the Proceedings of the Linnean Society (Botany) 1 (1857): 130–140. With the 1855 Gardeners' Chronicle note.
type: measurement camp: mainstream direction: supports
64 of 87 seed kinds germinated after 28 days in artificial seawater, Capsicum at 137 days. Print the retraction with it: Darwin's 1855 "permit me to cry peccavi… all sink within a month." He rescued the result with dried material and settled on a conservative ten plants per hundred crossing 900 miles. Use the conservative figure, not the famous one.
The New Flora of the Volcanic Island of Krakatau
Alfred Ernst, trans. A. C. Seward. Cambridge University Press, 1908.
type: measurement camp: mainstream direction: supports
The field confirmation, and better than expected: "39—72% have been introduced by sea-currents, 10—19% by birds, and 16—30% by air-currents." Sea drift was the largest single source, measured by a mainstream botanist. Ernst also supplies the disanalogy himself: Sebesi, 19 km away, "was only partially deprived of its vegetation."
Tsunami-driven rafting
James T. Carlton et al. Science 357 (2017): 1402–1406.
type: measurement camp: mainstream direction: mixed
289 living species on 634 objects across 7,000 km of Pacific over six years. Rafting demonstrably works. But the wooden debris lasted only two to three years because shipworms ate it, and no Western Pacific wood has arrived in North America with a living community in 150 years of observation.
Mammalian biodiversity on Madagascar controlled by ocean currents
Jason R. Ali and Matthew Huber. Nature 463 (2010): 653–656.
type: model camp: mainstream direction: mixed
Argues for rafting to Madagascar against land bridges — helpful — and then concludes that under modern currents nothing rafts there any more.
Genomic signatures of extensive inbreeding in Isle Royale wolves
Jacqueline A. Robinson et al. Science Advances 5(5):eaau0757 (2019).
type: measurement camp: mainstream direction: undercuts
The Ark's exact experiment run in the wild: two to three founders in the 1940s, extinct as a breeding population by 2018, 33% vertebral malformations, up to 48% of the genome in runs of homozygosity. The northern elephant seal is the counter-case and it is genuinely better, but it still shows an eightfold permanent loss of heterozygosity.
What could not be obtained
Listed so that the gaps are visible rather than hidden.
- Woodmorappe 1996 itself — still the highest-value acquisition, but no longer the blocker it was. There is no digitised copy anywhere: four separate Internet Archive probes return nothing, and nothing exists under his real name, Jan Peczkis. What does work is Google Books in-volume snippet search (volume
c-19AAAAMAAJ), which returns three snippets per query carrying the printed folio, cross-checkable against HathiTrust search-only (inu.30000053006874). The retrieval script is tracked atbin/google-books-snippets.sh. Using it, most of what this article attributes to the book is now page-verified rather than relayed through Morton: the 15,754 animal breakdown at p. 10 Table 1 (and the mammal row sums exactly, settling the 15,754-versus-15,746 discrepancy in Morton's favour), the rat-sized median at p. 13, floor space at p. 16, the 1,990 tons of feed and the juvenile factors at pp. 18–19, water at p. 20, the 144 m² window at p. 42, the 12 t/day excreta at p. 78, and the labour figures at pp. 71–81. ICR still sells the book new for $19.99, and used copies start at about $6, which is the cheapest way to close the remainder permanently. A free physical copy is available to Indiana residents at IU Bloomington (BS658 .W64 1996), or by interlibrary loan through Indiana Share quoting OCLC 35397664. - Two Woodmorappe items that did not survive checking. The "569 railroad stock cars" figure widely attributed to him is not in the book — zero snippet hits for
569,stock cars,stock carorboxcars, and the onlyrailroadhit is p. 16, where he is rebutting Moore's railroad analogy. Cut it or disown it. Conversely, the apparent 11%-versus-15% contradiction about sheep-sized animals is not a contradiction: p. 13 says 11% were "substantially larger than sheep" while the Impact summary's 15% is "sheep-sized or larger," which includes the sheep-sized band. Different quantities, both correct. - WorldCat could not be reached at all — repeated 429s,
turnstile_required, a disabled API, and finally a hard IP block. So the number of holding libraries is unknown, not estimated. Three holders were confirmed directly: IU Bloomington, Liberty University, and the Library of Congress. - HALOT (Brill paywall) and TWOT. Every HALOT reference in the research files is relayed through the NET translators' notes and flagged.
- The Genesis Flood (1961), lending-restricted on archive.org.
- Fish salinity physiology. The family-level euryhaline tally (Schultz and McCormick 2013) is paywalled, and measured 96-hour LC50 salinity values were not obtained. This is why section 7 of the article is the weakest-sourced part of it and says so.
- Luttmer and Boon, RINA 2007, the real peer-reviewed version of the wooden-ship-limit paper our notes cite through a Cloudflare-locked academia.edu copy.
- Primary confirmation of the Wyoming's scantlings and bracing count. The HAER documentation confirms only that she was "the largest wooden merchant ship to carry the American flag." William Avery Baker's A Maritime History of Bath, Maine is the source to get.
- A pre-mechanisation hand-labour standard. The one usable thesis is behind an institutional wall, so the labour break-even in the article is a lower bound on the difficulty.
- Archaeological attestation for gravity watering, siphons and float valves. Not researched. Until it is, the article must not claim ancient attestation for automatic watering.