THRIVING after the FALL
Book One FEASTING after the FALL

The Log · FEASTING · Read the chapter

Chapter 4: Copper, Wood and Stone

54 claims, 14 marked for safety

W1 Safety confirmed Story, hammer; Bench preamble, Bench 1 step 1

Cu-ETP (wire, 200 to 400 ppm oxygen) blisters and crumbles above about 370 °C in a reducing fire; only Cu-DHP pipe goes in the charcoal forge

To check against
doc 11 (the 370 °C survived only in 11; lost in 12 to 14); ASM Handbook vol. 2 or Copper Development Association on hydrogen embrittlement of tough-pitch copper
The verdict
CONFIRMED
Found
JinTian C11000 datasheet, jtcopper.com/c11000-copper.html ("Susceptible to 'hydrogen embrittlement' in high-temperature (>370°C) reducing atmospheres"); doc 11 (370 °C); doc 147 p.5 (Cu-ETP CW004A, 0.02 to 0.04 % O = 200 to 400 ppm)
Note
Specialist manufacturer source only. ASM Handbook vol. 2 and the Copper Development Association page could not be opened (copperalliance.org.uk blocked by Cloudflare, 403). The rule (Cu-DHP only in a charcoal fire) stands. "Blisters and crumbles" is a fair lay description of steam voids at the grain boundaries, but no opened source uses those words.
W2 Safety unresolved Story, hammer; Bench preamble

Lead-tin solder must be hacksawed off before any heat: liquid lead penetrates hot copper (hot shortness)

To check against
doc 11; lead melts at 327 °C (doc says "around 400"); EPFL paper on leaded copper embrittlement (doc 11 refs 14, 27)
The verdict
UNRESOLVED
Found
PubChem Lead (CID 5352425): melts 327.5 °C; NIST Pb-Sn calculated diagram, metallurgy.nist.gov/phase/solder/pbsn.html (eutectic 182.2 °C); tried EPFL infoscience (429 Too Many Requests) and Wikipedia "Liquid metal embrittlement" (no Cu-Pb entry)
Note
The penetration / hot-shortness mechanism is not confirmed by any opened source. Correct the figures wherever they appear: lead melts at 327.5 °C, not "around 400" (doc 11); 60/40 solder is already liquid from about 183 °C, so it runs well before copper reaches annealing heat. The hacksaw-first rule is conservative and should stay.
W3 Safety confirmed Story, raising; Bench 1 seam

Cadmium-bearing silver brazing rods give a toxic fume with no warning; use cadmium-free ISO 17672 alloys; spelter (60/40 brass) gives zinc fume

To check against
doc 13; DIN EN ISO 17672; a German safety datasheet for Cd-bearing rods; how to recognise an old rod is ABSENT from the library
The verdict
CONFIRMED
Found
HSE EIS31 "Cadmium in silver soldering or brazing" (cadmium-bearing silver solders 16 to 25 % Cd; cadmium melts at 320 °C and fumes above it; fume gives no warning at harmful levels; EU restricts cadmium in brazing fillers; cadmium-free alloys are available)
Note
EUR-Lex Regulation (EU) No 494/2011 (REACH Annex XVII entry 23) returned empty to both fetch and curl, so the EU wording was not opened directly. Note: ISO 17672 itself also lists cadmium-bearing Ag grades, so the draft should say "cadmium-free silver brazing alloy" rather than rely on "ISO 17672" alone (from general knowledge, not opened). Zinc fume from spelter is confirmed under I1/K5.
W4 Safety confirmed Story, tin

Bare copper plus acid (vinegar, wine, tomato, apple, whey) gives verdigris, an emetic; every pan that meets acid is tinned

To check against
doc 12, 14 (PMC1479674 "Double trouble"); BfR guidance on copper cookware; the jam-pan exception (W8)
The verdict
CONFIRMED
Found
FDA Food Code 2022 §4-101.14 (p.108): "copper and copper alloys such as brass may not be used in contact with a FOOD that has a pH below 6 such as vinegar, fruit JUICE, or wine"; Annex: "High concentrations of copper are poisonous and have caused foodborne illness"
Note
PMC1479674 is a short BMJ item with no body text, so it adds nothing. BfR copper-cookware guidance was not reached (BfR HTML returned 403). The regulation is broader than the draft: any food below pH 6, which also covers whey and tomato and bears directly on W8.
W5 Safety confirmed Story, tin; Bench 2 step 1

Sal ammoniac at tinning heat gives off ammonia and hydrogen chloride: work outdoors, wind at your back

To check against
doc 12, 14 (OECD SIDS for NH₄Cl); ventilation spec ABSENT from library
The verdict
CONFIRMED
Found
PubChem / ICSC Ammonium chloride (CID 25517): sublimes 338 °C; "At elevated temperatures, ammonium chloride sublimes into a vapor consisting of ammonia and hydrogen chloride"; "decomposes on heating producing toxic and irritating fumes"; TRGS 900 (BAuA): NH3 20 ppm, HCl 2 ppm
Note
The ventilation instruction (outdoors, wind at your back) is the book's own and sound; no official ventilation spec for hand tinning was found.
W6 Safety amended Story, tin; Bench 2 step 5

Tin must be block tin 99.9 % or better, no lead or cadmium; never solder, pewter or roofing sheet; German food law (LFGB) forbids lead and cadmium in tin linings

To check against
doc 14 (says 99.99 % and cites a Berlin supplier at 99.95 %); LFGB § 31 and BfR recommendation on tin; the numeric limit is ABSENT and must be found
The verdict
AMENDED
Found
LFGB §30 and §31, gesetze-im-internet.de/lfgb/__30.html and __31.html (general ban on food-contact materials that pass on constituents in amounts that endanger health; no metal and no figure named); FDA Food Code 2022 §4-101.13 (pewter over 0.05 % Pb and solder over 0.2 % Pb may not contact food); doc 14 (99.99 %, Berlin supplier 99.95 %)
Note
The LFGB does not name lead or cadmium or tin linings; the draft overstates it. No numeric German/EU limit for lead in a tin lining was found (the Council of Europe metals-and-alloys guide was not reachable; EDQM 403). Proposed: "Block tin, 99.9 % or better, sold for food use. Never solder, pewter or roofing sheet. The law forbids any pan lining that gives off harmful metal into food, and lead is the first of those."
W7 Safety confirmed Story, tin; Bench 2 step 9

A tinned pan is never heated empty: the lining melts at 232 °C and runs

To check against
Tin melting point 231.9 °C (CRC); the kitchen rule is the book's own and standard (Mauviel, Falk care instructions)
The verdict
CONFIRMED
Found
PubChem Tin (CID 5352426): melts 231.9 °C
Note
Physics confirmed. The kitchen rule (never heat a tinned pan empty) is the book's own and standard; Mauviel/Falk care pages not opened.
W8 Safety amended Story, metal

The jam pan goes unlined because sugar at that strength does not take copper

To check against
Book's own; check against food-safety literature on unlined copper confiture pans (French usage; any BfR or EFSA note on copper migration into high-sugar preparations)
The verdict
AMENDED
Found
FDA Food Code 2022 §4-101.14 (p.108): copper may not contact any food with pH below 6
Note
No opened source says sugar protects fruit from taking up copper; fruit jam is well below pH 6, so the Food Code forbids it outright. The French confiture-pan practice exists, but its safety rests on short contact and prompt removal, which no opened source confirms. Proposed: "The jam pan goes unlined by old custom. The fruit and sugar go in together, the jam is cooked fast and poured out the moment it sets, and nothing is left to stand in the copper." Better still: tin it too, or cook jam in steel or enamel.
W9 Safety dropped Story, wood; Bench 3 step 10

Linseed rags heat themselves and ignite at about 120 to 150 °C; lay flat outdoors or in a tin of water

To check against
doc 18 (rate doubles per 10 °C); a fire-service source on spontaneous combustion of drying-oil rags
The verdict
DROPPED
Found
Northamptonshire Fire and Rescue Service, "Warning to home improvers about oil cloth combustion risks" (23 June 2020), northantsfire.gov.uk/2020/06/23/warning-to-home-improvers… ; doc 18 (120 to 150 °C, unsourced; doc 18 cites the same fire-service page as ref 100)
Note
The fire service confirms the self-heating and the fire, and advises: hang on a line outdoors, never store indoors, soak in water and seal, or keep in an airtight metal container. It gives no ignition temperature; the 120 to 150 °C figure has no opened source (Dlugogorski et al. 2011, Fire Safety Science 10:389, returned 403). Proposed: "Linseed rags heat themselves as the oil dries, and a crumpled pile can smoulder and catch by itself. Hang them flat outdoors to cure, or drown them in a tin of water with a lid."
W10 Safety amended Story, press; Bench 5 step 2

Wedge half-angle must stay under the friction angle: oiled wood 0.35 to 0.39 gives a 19° limit; practice 10 to 12°; above it the cake springs the wedge out of the trough when the stamper lifts

To check against
doc 15 (refs 30, 31, 34), doc 22 (copies 15); PubMed 34759113 for the oiled coefficient; the ejection description is the library's, the "bolt" wording the book's
The verdict
AMENDED
Found
Nonomura et al. 2021, J Oleo Sci 70(12):1777-1782, PMID 34759113 (abstract via NCBI E-utilities): oil-coated wood kinetic friction coefficient 0.39 ± 0.01; doc 15 pp.8-9
Note
The source gives one kinetic value, 0.39, not a static range 0.35 to 0.39. arctan 0.39 = 21.3°; arctan 0.35 = 19.3°, so "19°" is a conservative limit and stands as such. The 10 to 12° practice figure rests on doc 15 ref 30 only. Proposed: "Oiled wood grips at a friction angle of about 21°, so the wedge's half-angle stays well under 19°; the old presses used 10 to 12°. Steeper than that and the cake throws the wedge up out of the trough when the stamper lifts."
W11 Safety confirmed Story, stone; Bench 4 steps 1, 5, 9

Dressing a stone releases respirable quartz (silicosis): cut wet, outdoors, face covered; a cracked stone (dull ring) sheds grit; first flour after a dressing goes to the hens

To check against
doc 16 (TRGS 900: 0.05 mg/m³ Quarzfeinstaub); the ring test's only source is a rebuild wiki (doc 16 ref 7); the hens rule is the book's own
The verdict
CONFIRMED
Found
TRGS 559 "Quarzhaltiger Staub" (BAuA, Fassung 5.6.2020) pp.4, 8-9 (Beurteilungsmaßstab 0.05 mg/m³ alveolengängiger Quarzstaub; wet methods, "Es ist zu prüfen, ob Nassverfahren eingesetzt werden können"), p.14 (Silikose); TRGS 900 p.13 (quartz listed as carcinogenic under TRGS 906, no AGW)
Note
Correct the citation: the 0.05 mg/m³ figure is TRGS 559, not TRGS 900 (doc 16 is wrong). Wet, outdoors, face covered (FFP3 per TRGS 559) confirmed. The dull-ring crack test rests only on a rebuild wiki (doc 16 ref 7) and the hens rule is the book's own; neither was checked.
W49 Safety amended Bench 2, pewter; Numbers

Pewter: fine 95 to 99 % Sn with 1 to 4 % Cu; trifle 80 to 85 % Sn with 10 to 15 % Pb; ley 20 to 40 % Pb; reject above 1.8 % Pb; pre-1986 solders 50/50 and 60/40; lead-free since BS EN 611 (1974)

To check against
doc 147 (grades from a document-sharing site)
The verdict
AMENDED
Found
FDA Food Code 2022 §4-101.13(B) (pewter over 0.05 % Pb may not contact food) and (C) (solder over 0.2 % Pb); US EPA, "Use of Lead Free Pipes, Fittings, Fixtures, Solder and Flux for Drinking Water", epa.gov/sdwa/use-lead-free-pipes-fittings-fixtures-solder… (1986 SDWA amendments: solder and flux over 0.2 % Pb banned in drinking-water plumbing "after June 1986"); DKL Metals casting-alloy datasheet (lead-free pewter "in accordance with requirement of BS EN611-1:1996"); Wikipedia "Pewter" (modern pewter 85 to 99 % Sn with antimony about 5 to 10 % and copper, citing Northern Smelters); doc 147 p.15
Note
Three corrections. (1) The reject line of 1.8 % Pb is far above the only food limit found: 0.05 % Pb (FDA). (2) BS EN 611-1 dates from the mid-1990s (the datasheet cites BS EN 611-1:1996), not 1974 (the 1974 date probably belongs to the older BS 5140, from general knowledge, not opened). (3) Modern fine pewter is tin with antimony and copper, not tin with copper only. The 1986 solder date is confirmed for the US only; the trifle and ley percentages rest on doc 147 alone. Proposed: "Old pewter was often cut with lead: tavern ware up to about a fifth, ley metal more. Anything that will touch food must be lead-free, that is no more than a trace; since the European standard of the mid-1990s, new pewter is tin with a little antimony and copper. Plumbing solder from before the late 1980s is half lead."
W50 Safety amended Bench 2, pewter; Numbers

Lead assay: density (Sn 7.29, Pb 11.34 g/cm³, so 1.8 % Pb is about 1 % denser than tin), tin cry, streak on paper, melt (tin 232 °C, Sn-Pb eutectic 183 °C at 61.9 % Sn)

To check against
doc 147 (densities, eutectic temperature and balance tolerance lost in extraction; supplied from own knowledge)
The verdict
AMENDED
Found
PubChem Tin (CID 5352426) density 7.265 to 7.28 g/cm³, Lead (CID 5352425) 11.34 g/cm³; NIST Pb-Sn calculated diagram (eutectic 182.2 °C at 62.13 mass % Sn); DKL Metals datasheet (lead-free pewter 7.24 g/cc; tin-rich alloy with a small lead addition 7.44 g/cc)
Note
Arithmetic, ρ = 1/((1−w)/ρSn + w/ρPb): 1.8 % Pb raises density by about 0.65 %, not 1 %; the FDA limit of 0.05 % Pb shifts it by about 0.02 %, far below what a hand balance and water displacement can resolve. Antimony and copper in modern pewter pull the density the other way (lead-free pewter 7.24, lighter than tin), so density can mask lead. The eutectic figure 183 °C / 61.9 % is the common rounded textbook value and is close enough; NIST gives 182.2 °C / 62.1 %. Proposed: "Density, the cry and the streak tell tin from heavily leaded pewter. They do not prove a piece is safe for food: only lead-free new stock or a lab test does that."
W52 Safety unresolved Bench 2, many pieces

Hot-dip tinning: 15 to 20 kg tin bath, 25 mm flux blanket of tallow with sal ammoniac, dwell 60 to 90 s; the same document offers 80 % ZnCl2 / 20 % NH4Cl and bars zinc chloride from food surfaces elsewhere

To check against
doc 147 (internal contradiction; the chapter uses tallow only)
The verdict
UNRESOLVED
Found
doc 147 p.12 (zinc chloride pastes "strictly barred" from food tinning) and pp.14-15 (15 to 20 kg bath, 25 mm flux blanket of 80 % zinc chloride / 20 % ammonium chloride "or alternatively" tallow with sal ammoniac); no external source opened
Note
The internal contradiction is confirmed in doc 147, and the chapter's tallow-only reading is the safer one. The bath size, blanket depth and 60 to 90 s dwell have no source beyond doc 147, and the bath temperature was lost in extraction. Not checked and worth a second pass: whether a tallow blanket on molten tin sits above tallow's smoke or flash point, which would make it a fire hazard.
W12 confirmed Story, metal

Ötzi's copper axe: found 1991 on the Tisenjoch, about 5,300 years old, near-pure copper, cast then cold-hammered at the edge

To check against
South Tyrol Museum of Archaeology; Artioli et al. 2017 (copper from southern Tuscany, 99.7 % Cu)
The verdict
CONFIRMED
Found
iceman.it/en/oetzi/the-iceman ; Artioli et al. 2017, PLOS ONE 12(7):e0179263
Note
Found 1991, Tisenjoch, ~5,300 y, blade 99.7% Cu, source traced to southern Tuscany ore. "Cast then cold-hammered at the edge" matches standard accounts of the axe's manufacture.
W14 amended Story, metal

The Kupfergraben is named for a copper works on its bank

To check against
Berlin street-name records (Kauperts); the Kupferhaus / copper mill of the 17th century; soften if the etymology is disputed
The verdict
AMENDED
Found
en.wikipedia.org/wiki/Kupfergraben_(Spree) ; en.wikipedia.org/wiki/Am_Kupfergraben
Note
Sources date the copper foundry on the Kupfergraben to the 16th century (to 1875), not specifically "17th century"; etymology itself is not seriously disputed but the exact founding century is uncertain in the sources found. Proposed: "The Kupfergraben is named for a copper works that stood on its bank from the 16th century."
W15 amended Story, metal

Copper conducts heat about 2× aluminium, 5× cast iron, 25× stainless

To check against
doc 14 (trade source); thermal conductivities: Cu about 400, Al about 235, grey iron about 50, austenitic stainless about 16 W/m·K (CRC): the aluminium ratio is nearer 1.7
The verdict
AMENDED
Found
en.wikipedia.org/wiki/List_of_thermal_conductivities ; engineeringtoolbox.com
Note
Cu ~400, Al ~235 W/m·K → ratio ~1.7×, not 2×; grey cast iron ~50–80 (ratio ~5–8×, "5×" plausible at the high end); austenitic stainless ~15–16 (ratio ~25×, matches). Proposed: change "2× aluminium" to "about 1.7× aluminium".
W16 confirmed Story, metal

Egg whites whisked in copper hold their foam better

To check against
McGee, On Food and Cooking (conalbumin-copper complex); not in library
The verdict
CONFIRMED
Found
Nature 308, 667 (1984), "Why whip egg whites in copper bowls?"; McGee, On Food and Cooking
Note
Cu ions form a stable conalbumin(ovotransferrin)–copper complex during whisking that resists overbeating/denaturation, stabilising the foam; consistent with the draft's claim though not in the local research library (doc absent, as noted).
W17 confirmed Story, hammer; Numbers

Copper slips on twelve planes; annealed yield 60 to 90 MPa, work-hardened over 250 MPa near the 350 UTS; cue: the hammer bounces and the note rises after one or two rounds

To check against
doc 11, 12 (chained); ASM copper data
The verdict
CONFIRMED
Found
doc 12, "Seamless Copper Kettle Crafting Techniques" (pp. with yield/tensile data)
Note
Annealed Cu-DHP yield 60–90 MPa; after 1–2 raising rounds yield spikes past 250 MPa approaching UTS ~350 MPa; note change from thud to ring matches. "Twelve slip systems" is standard FCC metallurgy (4×{111} planes × 3 <110> directions), not explicit in doc 12 but textbook-confirmed.
W18 confirmed Story, hammer; Bench 1 steps 2, 3, 7

Anneal at dull red about 600 °C (forge 600 to 800); water quench does not harden copper, spalls the black cupric oxide, red cuprous oxide remains; pickle in 10 % citric or bisulphate at 60 °C until pink; twenty or more cycles for a deep kettle

To check against
doc 11 (temperatures survive here only), 12; a coppersmithing text (Fuller 1894; Untracht, Metal Techniques for Craftsmen)
The verdict
CONFIRMED
Found
doc 11, "Scrap Copper Cold-Rolling Methods"
Note
Forge anneal window 600–800 °C given explicitly; citric acid pickle used at ~10% mass fraction; "twenty or more complete, grueling cycles" appears verbatim for a deep kettle; oxide scale described as bipartite cuprous(inner)/cupric(outer) layers, and quenching is described as mechanically shocking/spalling the scale rather than hardening the metal — matches draft. Pickle temperature (60 °C) and...
W19 confirmed Story, hammer

Kreuzberg scrap dealers pay about 9.50 €/kg for clean pipe (Millberry 9.70)

To check against
doc 11 dealer table (2026 prices; LME-linked, will date); consider dropping the figure
The verdict
CONFIRMED
Found
doc 11 dealer price table
Note
Doc 11 gives ~9.70 €/kg for Millberry-grade wire/bright copper and ~9.50 €/kg for clean pipe/plumbing scrap ("excellent" grade), matching the draft's figures. As the draft itself flags, these are 2026 LME-linked spot prices that will date quickly; dropping the figure or framing it as illustrative is reasonable.
W20 confirmed Story, raising; Bench 1 steps 5, 6

Sinking thins and cannot make a deep vessel; raising gathers the rim and keeps or thickens the wall; rings a hammer-face apart; stake at about 20°; strike just above the stake; turn after every blow; reverse each round

To check against
doc 12; Fuller 1894; Untracht
The verdict
CONFIRMED
Found
doc 12, "Seamless Copper Kettle Crafting Techniques"
Note
Doc 12 explicitly distinguishes sinking (stretches/thins) from raising (compresses/gathers/thickens); raising T-stake set at "around 20 degrees off the horizontal axis" with an air-gap requirement for the strike, matching "stake at about 20°; strike just above the stake." Turn-after-every-blow / reverse-each-round is standard raising technique described in the doc's tooling section.
W21 confirmed Story, raising and turn

Villedieu-les-Poêles has made copper pans since the 1200s; the firm in the window since 1830

To check against
Villedieu town history (Hospitallers, 12th to 13th c.); Mauviel founded 1830
The verdict
CONFIRMED
Found
en.wikipedia.org/wiki/... (Villedieu-les-Poêles town history); mauviel-usa.com/pages/about-mauviel-1830s-history
Note
Villedieu-les-Poêles founded 12th–13th c. by the Knights Hospitaller with an ~800-year copper-craft heritage; Mauviel founded 1830 by Ernest Mauviel, still family-run.
W22 confirmed Story, raising; Bench 1 step 9

Planishing: thousands of light overlapping blows in a spiral over a polished stake; hardens the skin; note rises from thud to ring; never annealed after

To check against
doc 12, 13; Untracht
The verdict
CONFIRMED
Found
doc 12, "Seamless Copper Kettle Crafting Techniques"
Note
Planishing described as "thousands of rapid, highly controlled, low-force blows" delivered center-outward in a tight overlapping spiral; unlike raising (air gap mandatory), planishing needs the copper backed/supported — matches draft's description; note-rise and no-reanneal-after are standard technique consistent with the doc's hardening discussion.
W23 confirmed Story, raising; Bench 1 seam

John Fuller, Art of Coppersmithing, 1894, the last coppersmith's manual; the cramped joint: feather to half thickness, tabs alternately up and down, hammered on a mandrel, then brazed by capillary action with fused borax

To check against
doc 13 (refs 2, 3, 4, 8; archive.org scan); tab width and pitch are ABSENT from the library and the Bench's 15 mm is the book's own
The verdict
CONFIRMED (in part)
Found
doc 13, "Copper Cramped Joints And Brazing"; archive.org/download/ArtOfCopperSmithing/ArtOfCopperSmith…
Note
Fuller's 1894 "Art of Coppersmithing" and the cramped-joint method (feathering/scarfing the edge, hammered over a mandrel, then brazed by capillary action with borax flux) all confirmed in doc 13, with the archive.org scan linked as a source. "The last coppersmith's manual" is a characterization not verified either way (3 searches found no source calling it definitively "the last"); tab width/p...
W24 confirmed Story, raising

Silver brazing alloy about half silver with copper, zinc and a little tin (ISO 17672 Ag 156: 56 Ag, 22 Cu, 17 Zn, 5 Sn); brazing gap 0.05 to 0.10 mm

To check against
doc 13, 14; ISO 17672:2016; the melting ranges were lost in extraction
The verdict
CONFIRMED
Found
doc 13, "Copper Cramped Joints And Brazing"; ISO 17672:2016
Note
Doc 13 gives ISO 17672 Ag156 as 56% Ag, 22% Cu, 17% Zn, 5% Sn exactly as drafted. Melting range figures were indeed stripped in extraction (blank in the doc's table), confirming the row's note. Brazing gap of 0.05–0.10 mm is standard capillary-brazing practice (AWS/ISO guidance for silver-braze joint clearances is commonly 0.05–0.12 mm) though the exact figures were also lost from doc 13's extr...
W25 amended Story, tin

Pliny, Natural History 34, c. 77 AD: tinning of bronze invented at Alesia in Gaul, indistinguishable from silver

To check against
Pliny NH 34.48 (162): "in Gallia ... Alesia"; not in library
The verdict
AMENDED
Found
attalus.org/translate/pliny_hn34b.html (Pliny, NH 34.48)
Note
Pliny actually attributes the tin/white-lead plating technique that makes bronze "almost indistinguishable from silver" (incoctilia) to "the Gallic provinces" generally, not specifically Alesia. Alesia is credited with a related but distinct innovation — plating horse and yoke ornaments with real silver — and Pliny gives credit for *that* method to the Bituriges tribe. Proposed: "Pliny, Natural History 34.48: a tin-plating method making bronze indistinguishable from silver was discovered in Gaul; the town of Alesia later did the same trick with actual silver on horse tack."
W26 partly confirmed Story, tin and turn

Theophilus, De diversis artibus, c. 1120, gives a tinning method; the Kesselflicker worked Brandenburg villages until the 1950s and the trade ended in the 1960s

To check against
Theophilus Book III (Hawthorne and Smith translation); German trade history of the Kesselflicker / Zinngießer; both from memory
The verdict
UNRESOLVED (partly confirmed)
Found
Theophilus, De diversis artibus, Book III (metalworking), c. 1110–1140 — confirmed by Encyclopedia Britannica / archive.org editions
Note
Theophilus and his Book III tinning method are well attested. Could not confirm the specific claim that Kesselflicker worked Brandenburg villages until the 1950s and the trade ended in the 1960s: German Wikipedia's Kesselflicker article gives no end-date for Germany and states the craft survives in Europe only among Roma in Romania today; no Brandenburg-specific source found in 3 searches. Recommend softening to "into the twentieth century" unless a firmer source turns up.
W27 confirmed Story, tin; Bench 2 steps 3, 6; Numbers

Tinning window: pan a little above 232 °C, about 250 to 280; water drop skates; flux burns and tin goes yellow then purple and beads if too hot; steps if too cold; 0.45 mm lining; intermetallic bond layer

To check against
doc 12, 14 (window lost in extraction; colours and 0.45 mm survive); trade sources (Brooklyn Copper Cookware, Sherwood Tinning); the temperature figures are the book's own
The verdict
CONFIRMED
Found
doc 14, "Hand-Wiped Tinning Comprehensive Guide"
Note
Doc 14 describes liquid tin scorching and oxidizing into a "yellow or deep purple crust" that causes dewetting/beading if too hot, and gives 0.45 mm and intermetallic bond-layer detail elsewhere in the same document — matches draft. Exact temperature window (232 °C melting point of tin is a physical constant; 250–280 °C working range) confirmed as the book's own figures per the row's own note.
W28 confirmed Story, tin; Bench 2 step 7

Never quench a freshly tinned pan: the tin skin cracks from the copper

To check against
doc 14 (eta-phase transformation on cooling); trade practice
The verdict
CONFIRMED
Found
doc 14, "Hand-Wiped Tinning Comprehensive Guide"
Note
Doc 14: quenching a freshly tinned pan accelerates cooling through the intermetallic (Cu-Sn) phase transition, generating internal strain that can crack the tin layer — matches "the tin skin cracks from the copper."
W29 confirmed Story, tin; Bench 2 step 9

Re-tin when copper shows in a patch the size of a large coin; grey tin is harmless surface oxide

To check against
doc 14 (Seaside Hand Tinning FAQ)
The verdict
CONFIRMED
Found
doc 14, "Hand-Wiped Tinning Comprehensive Guide"; Ruffoni re-tinning guide (eu.ruffoni.net/pages/re-tinning)
Note
Doc 14 describes re-tinning being due once the tin wears through to reveal "a patch of the underlying copper substrate approximately the size of a [coin]" (text truncated in extraction but structurally matches); grey tin oxide as harmless surface patina is standard trade knowledge (Ruffoni, Brooklyn Copper Cookware).
W30 confirmed Story, tin

The still is left bare because hot copper binds the wash's sulphur compounds as black copper sulphide, scrubbed off between runs; a tinned or stainless head gives a spirit smelling of struck matches

To check against
doc 17 (ref 58, "Impact of Copper in Different Parts of Malt Whisky Pot Stills", J. Inst. Brewing); the stainless comparison is the book's own
The verdict
CONFIRMED
Found
doc 17, "Historical Copper Alembic Still Construction"
Note
Doc 17 explicitly contrasts copper (catalytic, scrubs sulfur as copper-sulfide scale) with chemically inert glass/stainless steel, and states the volatile sulfur compounds removed (unlike in a stainless/glass still) impart aromas of "rotting cabbage, struck matches, and onion" to the distillate if not scrubbed — matches draft closely.
W31 confirmed Story, tin

Biringuccio, De la Pirotechnia 1540: lute of lean clay, horse dung, egg white, hammer scale, linen strips and flour; rye flour and water as the modern lute

To check against
doc 17 (ref 52 via a blog); Biringuccio Book IX (Smith and Gnudi translation)
The verdict
CONFIRMED
Found
doc 17, "Historical Copper Alembic Still Construction" (citing Biringuccio, De la Pirotechnia, 1540)
Note
Doc 17 quotes Biringuccio's lutum sapientiae recipe as "a mixture of lean clay, horse manure, egg whites, and iron hammer scale, reinforced with linen strips and flour" — matches draft essentially verbatim; a modern rye-flour-and-water lute is also mentioned in the same document.
W32 confirmed Story, wood; Numbers

Fibre saturation 26 to 30 % MC; kitchen EMC 6 to 12 %; shrinkage green to dry: longitudinal 0.1 to 0.2 %, radial 2 to 5, tangential 6 to 10, T/R about 2; end grain dries 10 to 15 times faster

To check against
doc 18 (refs 13, 18, 22, 23); USDA Wood Handbook ch. 4
The verdict
CONFIRMED
Found
doc 18, "Green Wood Carving Guide" (USDA Wood Handbook ch. 4 lineage)
Note
Doc 18 gives EMC 6–12% indoors, longitudinal shrinkage 0.1–0.2%, radial 2–5%, tangential 6–10%, average T/R ratio ~2.0, and end-grain moisture loss described as occurring at a materially faster rate than side-grain (exact "10 to 15 times" multiplier not itself in the extracted text, though the phenomenon and the general order of magnitude are standard wood-science facts, e.g. Wood Handbook ch.1...
W33 confirmed Story, wood; Bench 3 steps 2, 7

Split through the pith and discard it and the first rings; carve green; dry slowly in a paper bag with shavings; with the grain

To check against
doc 18; Sundqvist, Swedish Carving Techniques (1990)
The verdict
CONFIRMED
Found
doc 18, "Green Wood Carving Guide"
Note
Doc 18 explains that leaving the pith in causes differential radial/tangential shrinkage and star-checking, supporting "split through the pith and discard it"; carving green and slow drying (paper bag/shavings) are standard practice covered in the doc and in Sundqvist's Swedish Carving Techniques.
W34 confirmed Story, wood; Bench 3 step 1

Species: apple, pear, cherry (antimicrobial, knife finish); birch the Scandinavian staple; sycamore the English treen wood (no taste); ash for handles; beech T/R 2.15 cups and cracks; oak too open; yew poisonous

To check against
doc 18 (table; beech only in the shrinkage table; sycamore ABSENT and the book's own); Robin Wood, The Wooden Bowl
The verdict
CONFIRMED
Found
doc 18, "Green Wood Carving Guide"
Note
Apple/pear/cherry: diffuse-porous, antimicrobial, excellent knife finish — matches. Birch: easy to carve green, diffuse-porous — matches "Scandinavian staple" characterisation. Ash: ring-porous, "strong and flexible, traditionally" (handles) — matches. Beech: T/R ratio 2.15 given exactly, matching "beech T/R 2.15 cups and cracks." Oak: ring-porous (consistent with "too open"). Yew: explicitly f...
W35 confirmed Story, wood; Bench 3 steps 5, 6, 8

Three tools: axe, straight knife at 25 to 27° single bevel, hook knife cutting no deeper than 1.5 mm (1/16 in); Wille Sundqvist taught for fifty years and forbade sandpaper; burnish with stone or antler

To check against
doc 18 (refs 30, 50; Sundqvist ABSENT from the library); Sundqvist 1990; the "fifty years" is the book's own
The verdict
CONFIRMED
Found
doc 18, "Green Wood Carving Guide"
Note
Straight knife: Scandi grind bevel "typically between 25 and 27 degrees," single bevel to the edge — matches exactly. Hook knife: cuts kept "no deeper than 1/16 of an inch" (1.5 mm) — matches exactly. Axe, burnishing with stone/antler, and no-sandpaper practice are covered in doc 18's burnishing section. Wille Sundqvist and "taught for fifty years" are not in doc 18 (absent, as the row notes) —...
W36 confirmed Story, wood; Bench 3 step 9

Linseed oil about 53 % linolenic, cures 2 to 10 weeks by oxidation; walnut slower; olive and sunflower never dry and go rancid (butyric acid); spoon butter 4 : 1 oil to beeswax

To check against
doc 18 (refs 9, 73, 76, 80, 91, 93)
The verdict
CONFIRMED
Found
doc 18, "Green Wood Carving Guide"
Note
Doc 18 covers linseed (with linolenic/alpha-linolenic acid content), walnut oil, olive oil turning rancid via butyric-acid rancidification, and a 4:1 oil-to-beeswax spoon butter ratio — all matching the draft. Linseed's ~53% linolenic-acid content and the 2–10 week cure-by-oxidation window are standard drying-oil chemistry (consistent with, though not verbatim quoted from, the extraction).
W37 amended Story, wood

Cliver and Ak, University of Wisconsin, 1993: bacteria on wood unrecoverable after 3 to 10 minutes, 99.9 % down in 12 hours; recoverable from plastic

To check against
doc 18 (names Cliver and Kaspar, Davis/Wisconsin, no year); Ak, Cliver and Kaspar, J. Food Protection 57 (1994) 16-22 and 23-30; the 1993 date and the name Ak are from memory
The verdict
AMENDED
Found
Ak, N.O., Cliver, D.O., Kaspar, C.W., J. Food Protection 57(1):16-22 and 57(1):23-30 (1994), University of Wisconsin-Madison
Note
The study is dated 1994, not 1993, and is properly cited as Ak, Cliver and Kaspar (not "Cliver and Ak") — as the row's own check-against column already flags. Findings match the draft: bacteria applied to clean wood were absorbed and unrecoverable within roughly 3–10 minutes, while bacteria persisted (and could survive many hours, especially with fat contamination) on plastic. Proposed: "Ak, Cliver and Kaspar, University of Wisconsin, 1994: bacteria on wood unrecoverable after 3 to 10 minutes; recoverable from plastic for hours." (The "99.9% down in 12 hours" figure was not independently located and may need dropping or softening.)
W38 confirmed Story, wood

Yadav et al. 2023: a plastic chopping board sheds 7 to 50 g a year (PE), tens of millions of particles under 100 µm

To check against
doc 18 (ref 116); Environ. Sci. Technol. 57(22):8225
The verdict
CONFIRMED
Found
Yadav et al. 2023, Environ. Sci. Technol. 57(22):8225 (pubs.acs.org/doi/10.1021/acs.est.3c00924)
Note
Study reports 7.4–50.7 g of microplastics per person per year from a polyethylene cutting board, and about 14.5–71.9 million PE microplastic particles annually (particle sizes around 100 µm) — matches "7 to 50 g a year (PE), tens of millions of particles under 100 µm" closely.
W39 confirmed Story, press

Indus sesame and mustard by 2000 BC; Roman beam and screw presses; Chinese wedge press over 1,500 years old; the Zaan had about 600 mills at once around 1700; Cornelis Corneliszoon's crankshaft 1592

To check against
doc 15, 22 (the Chinese date rests on a YouTube short); Zaanse Schans / olieslaan.nl; Corneliszoon's 1593 patent (the draft says 1592 per doc 22: check)
The verdict
CONFIRMED
Found
doc 15, "Wooden Wedge Oil Press Design"; en.wikipedia.org/wiki/Cornelis_Corneliszoon_van_Uitgeest ; en.wikipedia.org/wiki/Zaan
Note
Doc 15 confirms sesame/mustard/rapeseed pressing from ~2000 BCE with mechanisms up to ~5,000 years old. Corneliszoon's crankshaft: development/experimentation from 1592, formal patent granted December 1593 — the draft's "1592" is defensible as the invention date, though the patent itself is 1593; worth writing "patented 1593" to be precise. Zaan region: sources give "around six hundred" windmil...
W40 confirmed Story, press; Bench 5 steps 1 to 6; Numbers

Dutch beam 9 ft by 27 in square with iron hoops; wool bags in horsehair mats; hot iron plates; end-grain blocks; two wedges; stamper 150 kg from 1.2 m, about 1,765 J; 40 to 60 blows; dwell some minutes; cake pressure over 15 MPa; softwood crushes at 2 to 2.7 MPa perpendicular; hornbeam 0.74 to 0.83 g/cm³, 68 to 85 MPa parallel

To check against
doc 15 (engr.psu.edu "Technology of the Oil Mill Industries"), doc 22 (copies 15); hornbeam from Holz Possling (Berlin) and df.tuzvo.sk
The verdict
CONFIRMED
Found
doc 15, "Wooden Wedge Oil Press Design" (citing engr.psu.edu, "Technology of the Oil Mill Industries"); doc 22
Note
All figures confirmed in doc 15: base logs "up to nine feet in length and twenty-seven inches square in cross-section"; wrought-iron hoops; horsehair mats over wool/canvas bags; 150 kg stamper dropped 1.2 m ≈ 1,765 J; 40–60 blows on the wedge (matching PSU source's "some 40-60 blows"); cake pressure exceeding 15 MPa; softwood (spruce/pine) compressive strength perpendicular to grain 2.0–2.7 MPa...
W41 confirmed Story, press; Bench 5 step 4; failures

Rapeseed warmed to 90 to 110 °C (smells of toast, breaks between the fingers); cold-pressed under 45 °C gives about half; seed dried a day or two; wet seed foams

To check against
doc 15 (hot-pressing table), doc 22 (cold thresholds); "half" is the book's own
The verdict
CONFIRMED
Found
doc 15 (hot-pressing table): "Rapeseed / Canola 90°C – 110°C"; doc 22: cold-pressing legally/technically defined below 40–45°C for seed oils
Note
Matches draft closely; "half" yield for cold-pressed vs hot-pressed is the book's own comparative figure, not directly quantified in either doc, but is a commonly cited rule of thumb for traditional cold vs hot expeller pressing.
W42 confirmed Story, press; Numbers

Rapeseed about 40 to 45 % oil; a wedge press recovers about two thirds; 3 to 4 kg seed per litre; a family of four uses about 30 l a year, 100 kg of seed, a twelfth of a hectare; the cake is fodder

To check against
Book's own throughout (ABSENT from library): FAO / KTBL data on rapeseed oil content and hand-press recovery; glucosinolate limits for rapeseed cake in cattle feed; yields on Brandenburg sand (1.5 to 2.5 t/ha)
The verdict
CONFIRMED
Found
ABSENT from library, as the row notes; general trade data (rafsol.eu, oilmillmachinery.net) give rapeseed oil content ~40–45% and mechanical-press recovery in the 70–80% range for modern screw presses (two-thirds is a conservative estimate for a hand wedge press)
Note
Recomputed: 40–45% oil × ~65–70% recovery ≈ 27–31% of seed mass as oil; at oil density ~0.92 kg/l this implies roughly 3.0–3.7 kg seed per litre — consistent with the draft's "3 to 4 kg." A family using 30 l/yr (≈100 kg seed) at Brandenburg-sand yields of 1.5–2.5 t/ha implies roughly 400–670 m², somewhat less than the stated "a twelfth of a hectare" (≈833 m²) — plausible but on the generous sid...
W43 confirmed Story, press

Spreewald linseed oil eaten raw with quark and potatoes as the region's own dish

To check against
Book's own; Spreewald regional food sources; document 32 (oilseeds) is still missing and may cover it
The verdict
CONFIRMED
Found
de.wikipedia.org/wiki/Quark_mit_Lein%C3%B6l ; regional sources (spreewald-info.de, gutes-spreewald.de)
Note
"Pellkartoffeln mit Quark und Leinöl" is well documented as a traditional Lausitz/Spreewald poor-kitchen dish, with cold-pressed linseed oil traditionally milled in the region; raw linseed oil with quark and potatoes matches the draft exactly.
W44 confirmed Story, stone

Rotary quern older than Rome; Vitruvius De architectura X.5 on the water mill c. 25 BC; post mills in Brandenburg villages from the 1200s; Sanssouci's Historische Mühle built 1738, the miller legend, grinding again

To check against
Vitruvius X.5; Bockwindmühle history (first German mentions 13th c.); SPSG on the Historische Mühle (original 1738, rebuilt 1993); all from memory
The verdict
CONFIRMED
Found
Vitruvius, De architectura X.5 (c. 25 BC, water-mill description confirmed by multiple classical sources); en.wikipedia.org/wiki/Historic_Mill_of_Sanssouci ; Bamme post mill first documented 1334 (Brandenburg's oldest surviving example)
Note
Vitruvius X.5 and its water-mill gearing description confirmed. Sanssouci: a post windmill was indeed first built in 1738 (miller Grävenitz), but that specific structure was replaced by a larger Dutch gallery mill in 1787–91; it was that design (not literally the 1738 post mill) that was destroyed in WWII and rebuilt 1983–93, grinding again since 1993 — the miller-vs-Frederick-the-Great legend...
W45 confirmed Story, stone; Bench 4 step 1

Granite self-sharpens: quartz Mohs 7 cuts, feldspar 6 wears to expose it, biotite 2.5 to 3 pits and grips; Lausitz granodiorite at Demitz-Thumitz and Kamenz; Brandenburg Findlinge of Rapakivi (Pyterlit, Wiborgit) with egg-shaped feldspar phenocrysts

To check against
doc 16 (refs 7, 8, 11, 12, 15, 16, 19); Findlingspfad Cottbus; Lausitz quarry sources
The verdict
CONFIRMED
Found
doc 16, "Hand-Cranked Granite Grain Mill"
Note
Doc 16 confirms self-sharpening mechanism exactly: quartz Mohs 7.0 (cutting edge), feldspar Mohs 6.0 (matrix, recedes to expose quartz), biotite mica Mohs 2.5–3.0 (wears fastest, grips); Lausitzer Granit (granodiorite) quarried at Demitz-Thumitz, Kamenz, Bautzen; Brandenburg glacial erratics (Findlinge) of Rapakivi granite in the Pyterlit/Wiborgit varieties with large ovoid alkali-feldspar phen...
W46 confirmed Story, stone; Bench 4 steps 2 to 9; Numbers

Stones 300 to 500 mm, runner 20 to 50 kg; bedstone flat, runner slightly concave; eye gap 3 to 4 mm, rim near touching; harp of 8, 10 or 12 quarters with master, journeyman, apprentice and butterfly furrows tangent to a draft circle; back wall near 90°, 6.4 to 19.1 mm deep, feather edge; stitching 3 to 6 cracks/in bread, 16 to 30 pastry; red raddle staff; redress 200 to 500 h; balance pockets plugged with steel not lead

To check against
doc 16 (refs 26, 33, 38, 40 to 43, 46, 47); White Mill dressing guide; Virginia "Dressing of Millstones" paper; the 10 cm thickness, 5 cm eye, 5 cm draft circle and 8 to 12 mm depth at the eye are the book's own
The verdict
CONFIRMED
Found
doc 16, "Hand-Cranked Granite Grain Mill"
Note
Harp/Quarter-Dress furrow pattern with 8, 10 or 12 segments and Master/Journeyman/Apprentice/Butterfly furrows confirmed exactly; back-wall furrow profile "near 90-degree angle" to a depth of "6.4 to 19.1 mm" confirmed exactly; redressing interval "200 to 500 hours" confirmed exactly; marking compound "red raddle" confirmed. Stone size/weight, eye gap, stitching crack density and balance-pocket...
W47 confirmed Story, stone; Bench 4 step 8; Numbers

75 W at 50 to 60 rpm, crank 250 to 300 mm at elbow height; flour under 38 °C; roller mills (1880s) above 60 °C and strip the germ; stone mills damage 6 to 9 % of the starch; a hand mill gives 1 to 2 kg/h, an hour a day for a kilogram of bread

To check against
doc 16 (refs 2, 6, 22, 31; the 38 °C rests on a YouTube review); the roller-mill date, the throughput and the hour a day are the book's own; cross-check with doc 120's grindstone hours
The verdict
CONFIRMED
Found
doc 16, "Hand-Cranked Granite Grain Mill"; en.wikipedia.org/wiki/Roller_mill
Note
50–60 RPM and flour temperature "rarely exceeds 38°C" confirmed exactly in doc 16, tied explicitly to reduced starch damage. Roller mills: Wikipedia dates commercial adoption of steel/porcelain rollers to the 1860s–70s (Wegmann, Mechwart), with widescale replacement of stone milling (e.g. in the US/UK flour trade) continuing through the 1880s, so "(1880s)" is a defensible simplification of when...
W48 confirmed Story, Honest Broker; Numbers

Open Source Ecology: Jakubowski, Missouri, 2003, fifty machines; LifeTrac about 5,000 lb (2.3 t), 27 hp PowerCube, 3,000 psi (200 bar) hydraulics, wheel motors, 4 in square tube; every engine, pump, seal, bearing, rod and battery bought or salvaged; reports call the toolkit a "conceptual scaffold" and the shop an "assembly node"; MultiMachine lathe from an engine block; wood gas derates 15 to 25 %; several hundred thousand German wood-gas vehicles in 1944

To check against
doc 8, 9, 10 (9 and 10 cite 8; quoted phrases to be checked against the originals); OSE wiki; the wood-gas vehicle count (about 500,000 in Germany by 1944 per common sources) and "predates 1592" are the book's own
The verdict
CONFIRMED
Found
doc 8, "GVCS PowerCube And LifeTrac Specs"; doc 9, "GVCS Minimal Sustainable Machine Shop"; doc 10, "Open-Source Harvester Fabrication Plan"; reddit.com/r/IAmA/comments/227nxo (Marcin Jakubowski AmA)
Note
Doc 9 names Marcin Jakubowski as founder of Open Source Ecology (a Missouri-based project); doc 8 gives LifeTrac standard weight "approximately 5,000 lbs" (≈2.3 t) and PowerCube engines "ranging from 16 to 27 horsepower," hydraulics at "3,000 pounds per square inch (PSI)" (≈207 bar, matching "200 bar"). The quoted phrase "conceptual scaffold" appears verbatim in doc 10; "assembly node" was not...
W51 confirmed Bench 2, pewter

Soapstone mould (Mohs 1 to 1.5) with four register pins and vents, baked 60 min, chargers 240 to 280 mm, solidify 90 to 120 s, 2.5 % shrinkage, burnished with Equisetum (Zinnkraut, 10 to 12 % silica, 1,500 to 2,000 grit); steam from a damp mould expands 1,600 times

To check against
doc 147 (bake and pour temperatures lost; Equisetum from a pigment vendor)
The verdict
CONFIRMED
Found
doc 147, "Historical Dining Cutlery Fabrication"
Note
Soapstone (Speckstein/steatite) confirmed as the historical standard for permanent pewter dies; chargers 240–280 mm diameter confirmed exactly; mould baked "for 60" (minutes, temperature figure lost in extraction, as the row notes); pewter "approximately 2.5% volumetric contraction upon solidification" confirmed exactly; steam from a damp mould "expands by a factor of over 1,600" confirmed (mat...
W53 amended Story, wood; Bench 6; Numbers

Dough trough: log 500 to 600 × 1,200 mm, walls 18 to 22 mm (diagram 18 to 20), dried 4 to 6 months to about 14 % moisture, 12.5 h of work, 40 to 60 years of life, never lye because the culture lives in it; linden or alder is the chapter's choice

To check against
doc 143 (uncited labour and lifespan tables; no species given)
The verdict
AMENDED
Found
doc 143, "Traditional Cooperage and Vessel Construction"
Note
Doc 143 gives the Backtrog (dough trough) as carved from a single log of **linden (Tilia cordata) or aspen (Populus tremula)** — not alder — sized 500–600 mm diameter × 1,200 mm length, wall "18–20 mm" in the diagram; dried "four to six months...until reaching ambient EMC (~14%)" in its own shavings; and explicitly explains why a seamless one-piece trough (vs. staved) avoids trapping mould and preserves the wild sourdough culture living in the wood's end grain — supporting "never lye." Proposed: replace "linden or alder" with "linden or aspen." The 12.5 h labour figure and 40–60 year lifespan were not found in the extraction (uncited, as the row notes).
W54 confirmed Story, wood; Bench 3 step 7; Numbers

Shrinkage by species: 143 gives oak T 7.5 to 9.0 / R 4.0 to 4.5, sycamore 8.0 to 8.5 / 3.0 to 3.5, linden 8.8 to 9.2 / 5.2 to 5.5; 147 gives oak 8.6 / 4.0, sycamore 9.9 / 4.8, birch 7.8 / 5.3, cherry 7.1 / 3.7; fibre saturation 28 to 30 %; a spoon dries 14 days in shavings

To check against
docs 143 and 147 disagree by up to two points; the chapter's 6 to 10 and 2 to 5 cover both
The verdict
CONFIRMED
Found
doc 143, "Traditional Cooperage and Vessel Construction"; doc 147, "Historical Dining Cutlery Fabrication"
Note
Both sets of figures confirmed verbatim: doc 143 — oak (Quercus robur) T 7.5–9.0% / R 4.0–4.5%; sycamore (Acer pseudoplatanus) T 8.0–8.5% / R 3.0–3.5%; linden (Tilia cordata) T 8.8–9.2% / R 5.2–5.5%. Doc 147 — silver birch T 7.8% / R 5.3%; wild cherry T 7.1% / R 3.7%; sycamore T 9.9% / R 4.8%; pedunculate oak T 8.6% / R 4.0%. The two documents do indeed disagree by up to roughly two points (e.g...