Chapter 5: Making Ground

52 claims, 12 marked for safety

38 confirmed
9 amended
4 unresolved
1 dropped
12 marked for safety

Filter the rows

G1 Safety confirmed Story, heap; Bench 1 step 6, Bench 6 step 3

Sanitation rule: 55 °C held three consecutive days in an aerated static pile, or fifteen days with at least five turnings in a windrow

To check against: doc 23 (US EPA 40 CFR 503 PFRP via its ref 6); doc 42 repeats with the temperature lost
The verdict: CONFIRMED
Found: 40 CFR Part 503 Appendix B, B.1 (PFRP composting), eCFR ecfr.gov/current/title-40/part-503/appendix-Appendix%20B%… : "within-vessel ... or ... static aerated pile ... maintained at 55 degrees Celsius or higher for three days ... windrow ... 55 degrees or higher for 15 days or longer ... a minimum of five turnings"; WHO 2006 vol. 4 (IRIS 10665/78265) printed p.63
Note: Exact match. Note for the draft: the rule is written for sewage sludge in managed piles. WHO warns that "cold zones form" in compost, and recommends a month at 55 to 60 °C plus 2 to 4 months' maturing as the handling margin (see G2).

G2 Safety amended Story, heap

Ascaris eggs >99 % dead after three to seven weeks at 55 °C; they survive chlorination and drying and are the indicator organism

To check against: doc 23 ref 66 (review of pathogen survival in garden compost)
The verdict: AMENDED
Found: WHO 2006 vol. 4 printed p.63 ("composting in aerated piles for one month at 55-60 °C (plus 2-4 months of further maturation) ... Under controlled conditions, composting at 55-60 °C for 1-2 days is sufficient to kill essentially all pathogens ... cold zones form"); printed p.63 ("Normal moisture levels do not inactivate Ascaris eggs. Moisture levels below 5% are needed"); printed p.107 (Ascaris as a resistant indicator); doc 23 p.14 (3 to 7 weeks, citing its ref 6, an antibiotic-resistance review)
Note: No opened source gives ">99 % in three to seven weeks at 55 °C". Survival of drying and use as the indicator organism are confirmed; resistance to chlorination rests on doc 23 only. Proposed: "Roundworm eggs are the hardest thing in the heap: they outlast drying and most disinfectants, which is why they are the test. A month at 55 to 60 °C, with every part of the heap turned through the hot core, and then months of maturing, is what kills them."

G3 Safety amended Story, heap; Bench 1 step 6, Bench 6 steps 3, 4

Humanure compost rests a full year after the hot phase and goes only under fruit trees and grain, never under anything eaten raw

To check against: ABSENT from the library ("extended curing" only); Jenkins, Humanure Handbook (one year); WHO 2006 Guidelines for the Safe Use of Wastewater, Excreta and Greywater vol. 4
The verdict: AMENDED
Found: WHO 2006 vol. 4 printed p.68 Table 4.4 (composting: >50 °C for >1 week, minimum; "Longer time needed if temperature requirement cannot be ensured"), printed p.69 Table 4.5 (storage at ambient 2 to 20 °C: 1.5 to 2 years; at >20 to 35 °C: >1 year, with 10 to 30 % of Ascaris eggs still viable at 24 months in some conditions), printed p.68 ("Improperly sanitized excreta ... should not be used for vegetables, fruits or root crops that will be consumed raw, excluding fruit trees"; withholding period at least one month)
Note: The draft's rule is safe when the hot phase really ran hot throughout. At Berlin temperatures a heap that did not (the usual garden case) needs 1.5 to 2 years of storage by WHO's table, not one. The fruit-trees-and-grain restriction matches WHO's exclusion. Proposed: "Humanure rests a full year after a proper hot phase, two years if the heap never ran hot all through, and then goes only under fruit trees and grain, never under anything eaten raw." Jenkins, Humanure Handbook, not opened.

G4 Safety amended Story, heap; Bench 2 step 5, Bench 6 step 2

Urine: sterile on leaving the body, hydrolyses to ammonia in a day or two; store a month closed, dilute 1:10, apply to soil not leaves, a month before harvest, never on salad crops

To check against: ABSENT from the library beyond 70 to 80 % of N and 1.42 l/day (doc 23 refs 29, 30); WHO 2006 vol. 4 (1 month storage for single household, 6 months at 20 °C for larger systems; dilution 1:3 to 1:10); Jönsson et al. 2004 (SEI)
The verdict: AMENDED
Found: WHO 2006 vol. 4 printed p.34 ("In a healthy individual, urine in the bladder is sterile. However, different types of bacteria are picked up in the urinary tract. Freshly excreted urine normally contains <10 000 bacteria per ml"); printed p.70 Table 4.6 and text (single household: "no storage is needed"; larger systems 1 month at 4 °C to 6 months at 20 °C; sealed container; "The urine should preferably not be diluted" in storage; one month between application and harvest for crops eaten raw; incorporate into the ground); printed pp.10-11 (applied undiluted or diluted; incorporate into soil, surface application loses >70 % of N as ammonia)
Note: Four corrections. (1) Urine is sterile in the bladder, not on leaving the body; the real risk is faecal cross-contamination. (2) WHO gives no dilution ratio; 1:10 is not in the source (and 1:3 to 1:10 as cited in the input row was not found in vol. 4). (3) Store undiluted, not diluted. (4) "Hydrolyses to ammonia in a day or two" has no opened source. The month of closed storage and the month before harvest are confirmed and conservative for one household. Proposed: "Urine leaves the bladder clean but picks up gut germs on the way and in the pot. Keep it a month in a closed can, undiluted; water it down only on the day you use it, pour it on the soil and work it in, never on leaves, and stop a month before harvest. Not on salad."

G5 Safety amended Story, heap; Bench 1 step 8; failures

Immature compost carries ammonia, acetic, propionic and butyric acid and kills seedlings; cress germination index 60 % usable, 80 % clean; phytotoxicity gone by day 56 with turning every four days

To check against: doc 23 (GI thresholds, 56-day study), doc 42 copies 23; Zucconi et al. 1981 for the GI method; the saucer version of the test is the book's own
The verdict: AMENDED
Found: Tiquia 2010, Chemosphere 79(5):506-12, PMID 20299074 (abstract via E-utilities): cress GI >80 "an indicator of the disappearance of phytotoxicity"; phytotoxicity correlated most with NH4+-N, extractable Cu and Zn; hog-manure windrows "turned every 4d with weekly moisture adjustment to 60%" lost phytotoxicity "within 56d"; doc 23 p.13
Note: The 80 % line and the 56-day result are confirmed, but the 56 days is for hog manure under optimum control, not a garden heap. No opened source gives 60 % as "usable". The acids (acetic, propionic, butyric) rest on doc 23 only; the opened study names ammonium, copper and zinc. Zucconi et al. 1981 not opened. Proposed: "Unripe compost burns seedlings, mostly with ammonia. Cress on a saucer tells you: if it sprouts and roots about as well as in plain water (four fifths or better), the heap is ready. A well-turned heap can get there in about two months; a garden heap usually takes longer."

G6 Safety amended Story, char; Bench 2 step 1

Char made hot from dirty feedstock concentrates heavy metals and carries carcinogenic PAHs; EBC-Agro limits Pb <120, Cd <1.5, Cu <100, Zn <400, Ni <50 mg/kg, 16 EPA PAHs <6 mg/kg; "about 500 °C" for clean char is the book's own

To check against: doc 25 (temperatures lost in extraction); European Biochar Certificate Guidelines v10.1; the pyrolysis temperature and the PAH-temperature relation need a source
The verdict: AMENDED
Found: European Biochar Certificate Guidelines v10.1E (10 January 2022), european-biochar.org/media/doc/2/version_en_10_1.pdf : p.6 Table (EBC-Agro Pb 120, Cd 1.5, Cu 100, Ni 50, Zn 400, also Hg 1, Cr 90, As 13 g/t DM; 16 EPA PAH 6.0 + 2.2 g/t DM; 8 EFSA PAH 4 g/t DM); p.23 (heavy metals are retained and concentrated in the char; "not critical" unless the biomass was contaminated); p.25 ("The type of biomass feedstock used for biochar production has a negligible influence on the PAH content"; PAH comes from process conditions and from char cooling in pyrolysis gas); p.34 (feed-grade char: at least 500 °C for 10 min to destroy organic micropollutants)
Note: Limits confirmed (g/t = mg/kg), with the 2.2 tolerance on PAH and three more metals the draft omits. The causal link is wrong: dirty feedstock concentrates metals; PAH comes from how the char is made and cooled, not from the feedstock. "About 500 °C" has partial support (EBC-Feed rule). Proposed: "Char concentrates whatever metal the wood carried, so only clean wood goes in: no painted, treated or roadside timber. Tar and smoke settle into char that cools in its own fumes, so the fire is kept bright and the char is quenched as soon as it is made."

G7 Safety confirmed Bench 2 step 3

A smouldering char pit gives off carbon monoxide; never cover with earth and leave, never indoors; quench with water

To check against: ABSENT from the library; Ithaka Institute Kon-Tiki guidance (Schmidt and Taylor 2014); any charcoal-burning safety source
The verdict: CONFIRMED
Found: FAO Forestry Paper 41 (1983/1987), Simple technologies for charcoal making, ch. 4.2, fao.org/4/x5328e/x5328e05.htm : "The gas produced by carbonization has a high content of carbon monoxide which is poisonous when breathed"; CO is "also produced during unloading through spontaneous ignition of the hot charcoal"; EBC Guidelines v10.1E p.7 (char kept at about 30 % moisture to prevent dust and self-ignition); EBC p.25 (Kon-Tiki flame-curtain kiln)
Note: CO hazard and the re-ignition risk confirmed. Quenching with water is supported indirectly (EBC moisture rule); the Ithaka Kon-Tiki guidance (Schmidt and Taylor 2014) was not opened.

G8 Safety amended Story, water; Bench 5 step 9

Biosand filter removes >4 log helminths, 3 to 4 log protozoa, 1.5 to 2.5 log bacteria (95 to 99 %), 0.5 to 1.5 log viruses (70 to 90 %); it is not a steriliser and drinking water is boiled after it, one minute at a rolling boil at this altitude

To check against: doc 78 (LRV table; prose says "upwards of 99 %" for bacteria); WHO Boil Water brief WHO/FWC/WSH/15.02; CDC Yellow Book
The verdict: AMENDED
Found: CAWST Biosand Filter Construction Manual (August 2012), washresources.cawst.org/en/resources/b6be2637/biosand-fil… , manual p.18 table (bacteria up to 98.5 % lab, 87.9 to 98.5 % field; viruses 70 to >99 % lab; protozoa >99.9 % lab; helminths "Not researched ... up to 100% removal efficiency is assumed"), manual glossary ("rolling boil ... for at least 1 full minute"); Jenkins et al. 2011, Water Res, PMID 21974872 (mean 1.40 log faecal coliforms, 0.54 log MS2; best design 1.82 log = 98.5 %, 0.94 log = 88.5 %); WHO Boil water technical brief WHO/FWC/WSH/15.02 ("heating water to a rolling boil ... is sufficient to inactivate pathogenic bacteria, viruses and protozoa")
Note: Bacterial removal is overstated: 1.5 to 2.5 log (97 to 99.7 %) is above the measured 1.4 to 1.8 log (about 96 to 98.5 %), and field results go down to 88 %. Helminth ">4 log" is assumed, not measured. The boil advice is confirmed and the one minute is the CAWST figure (WHO: a rolling boil is enough). Proposed: "The filter takes out most bacteria (up to about 98 %, less in daily use), most protozoa, some of the viruses. It is not a steriliser: drinking water is boiled after it, a full minute at a rolling boil."

G9 Safety confirmed Story, water

Microcystin from summer pond algae survives hours of boiling; a ripe schmutzdecke degrades it slowly and poorly under 4 °C; a charcoal layer is the second barrier

To check against: doc 78; WHO cyanotoxin guidance; the claim that the filter "takes out the things that survive boiling" is the draft's paraphrase and should be softened if 78's source is weak
The verdict: CONFIRMED
Found: WHO 2020, Cyanobacterial toxins: microcystins, WHO/HEP/ECH/WSH/2020.6 (IRIS 10665/338066), p.11 ("MCs and nodularins are extremely heat stable"), p.14 ("MCs are resistant to boiling"), p.43 (biological degradation during slow sand filtration or on GAC "can be very effective, although it may require a lag phase"; activated carbon effective); Grützmacher et al. 2002 (UBA Berlin), Environ Toxicol 17(4):386-94, PMID 12203961 (>95 % elimination in slow sand filters by biodegradation; <60 % at temperatures <4 °C)
Note: Confirmed. Two cautions: WHO says heat-stable and boiling-resistant, not "hours of boiling"; and the filter only degrades microcystin after its biolayer has met the toxin (lag phase), in a full-scale slow sand filter, not a household biosand unit. The draft's "takes out the things that survive boiling" should be softened to "can take out some".

G10 Safety amended Story, heap and turn

The Rieselfelder soils are contaminated with cadmium, lead and other metals and much of the land is still unsafe for food crops; Hobrechtsfelde is grazed to keep metals out of the food chain

To check against: ABSENT from the library; Berlin Umweltatlas on Rieselfelder contamination; Hobrechtsfelde grazing project (Förderverein Naturpark Barnim); Blumenstein et al. on Berlin sewage-farm cadmium
The verdict: AMENDED
Found: Senatsverwaltung Berlin, Umweltatlas "Ehemalige Rieselfelder" (Ausgabe 2010), berlin.de/umweltatlas/boden/rieselfelder/2010/einleitung/ (soils "flächendeckend in zum Teil erheblichem Maße mit Schwermetallen belastet"; metals accumulate in food plants; locally a health risk on direct contact) and .../2010/kartenbeschreibung/ (vegetable growing banned at Karolinenhöhe in 1985; elsewhere food crops restricted in favour of fodder)
Note: Heavy-metal contamination and the food-crop restrictions are confirmed. The opened pages do not name cadmium or lead individually, and nothing opened supports "Hobrechtsfelde is grazed to keep metals out of the food chain" (grazing animals that are eaten would pass metals on). Förderverein Naturpark Barnim and Blumenstein et al. not reached. Proposed: "The old sewage farms still carry the city's heavy metals in their soil. Vegetables were banned on some of them in the 1980s and food crops were moved off others, and the land around Hobrechtsfelde is grazing now, not garden."

G11 Safety amended Story, heap

PFAS and microplastics pass through composting, digestion and vermicomposting unchanged; only source separation protects

To check against: doc 23; a primary source (e.g. a 2023 to 2025 review on PFAS in compost)
The verdict: AMENDED
Found: Bolan et al. 2021, Environ Int 155:106600, PMID 33964642 (composts, biosolids and manures are "a major source of" PFAS input to soil; PFAS chemical and thermal stability; source control as the first remediation route); Huang et al. 2025, Environ Res 279:121795, PMID 40340003 (PFAS "tend to accumulate during the biotransformation of municipal waste"; some microbial adsorption and degradation reported); doc 23 p.15
Note: "Unchanged" overstates it: PFAS persist and concentrate, but some compounds change form or are partly removed. Microplastics were not checked against a primary source. Proposed: "PFAS and plastic fragments come through composting, digestion and the worm bin; the heap does not clean them out. The only protection is to keep them out of the heap in the first place."

G12 Safety dropped Bench 5 step 8; failures

Chlorinated water sterilises the filter; water over about 50 NTU must be settled first; the skin sleeps below 10 °C (15 to 30 % activity at 0 to 5 °C) and the housing cracks if frozen

To check against: doc 78; CAWST Biosand Filter Construction Manual
The verdict: DROPPED
Found: CAWST Biosand Filter Construction Manual (2012) manual p.15 ("Do not pour water that has been chlorinated into the filter. The chlorine will kill the biolayer"; "If the source water is over 50 NTU, it is recommended to use a sedimentation method"); Grützmacher et al. 2002, PMID 12203961 (biodegradation drops below 4 °C); doc 78 p.11 (15 to 30 % at 0 to 5 °C, sourced to a user forum, its ref 76)
Note: Chlorine and 50 NTU rules confirmed. The 15 to 30 % activity figure rests on a forum post; no opened source gives it. Frost damage is basic physics (water expands about 9 % on freezing) but CAWST does not address it. Proposed: "Never pour chlorinated water in: it kills the skin. Water cloudier than a bottle you can read through gets settled first. In the cold the skin slows almost to sleep, so winter water is boiled with extra care, and the filter is kept from freezing or it will crack."

G13 amended Opening; Story, ice; Numbers

Brandenburg soils are Pleistocene glacial sands, Arenosols, rated 25 to 40 Bodenpunkte; the scale runs 7 to 100, set by the Reichsbodenschätzung of 1934, with 100 at the Magdeburg Börde

To check against: doc 25 (25 to 40 only); Bodenschätzungsgesetz 1934 and the Eickendorf reference plot (100) from memory; the 7 lower bound from memory
The verdict: AMENDED
Found: doc 25, "Berlin Organic Topsoil Generation"; de.wikipedia.org/wiki/Bodensch%C3%A4tzung ; de.wikipedia.org/wiki/Bodenwertzahl
Note: Doc 25 confirms 25–40 Bodenpunkte for Brandenburg's Pleistocene-glacial Arenosols. The Reichsbodenschätzung dates to the law of 16 October 1934, with the reference soil (Schwarzerde) rated 100 near Eickendorf in the Magdeburger Börde — confirmed. However, German sources give the theoretical scale as running from 0 (not 7) to about 100 ("reicht von 0 ... bis ca. 100"); the "7" lower bound was not found in any source and appears to be a specific low empirical value rather than the scale's floor. Proposed: "...the scale runs from 0 in theory (about 7 for the poorest measured Brandenburg sands) to 100..."

G14 confirmed Story, ice

The "sandbox of the Holy Roman Empire" (Streusandbüchse des Heiligen Römischen Reiches) as an eighteenth-century name for the Mark Brandenburg

To check against: not in library; German phrase usually attributed to the eighteenth century, sometimes to Fontane; the attribution is uncertain and the draft says "somebody"
The verdict: CONFIRMED
Found: de.wikipedia.org/wiki/Streusandb%C3%BCchse
Note: German Wikipedia confirms the Mark Brandenburg was called "Streusandbüchse des [Heiligen Römischen] Reichs" for its sandy soil, but gives no date or named originator for the coinage, and does not mention Fontane. The draft's hedge ("somebody") is the honest framing given the sourcing found.

G15 confirmed Story, ice; Numbers

Berlin rainfall 539 to 618 mm, mean about 591 on 167 rainy days; July 75 mm, February 36; comparable to London

To check against: doc 122 (its monthly table sums to 580); Berlin Umweltatlas precipitation 1991 to 2020; London ~600 mm from memory
The verdict: CONFIRMED
Found: doc 122, "Water Demand and Catchment Scale"
Note: 539–618 mm range and ~591 mm mean over 167 rainy days confirmed exactly. Doc 122's own monthly table (Jan 47 ... Dec 47) sums to 580 mm, about 11 mm (~2%) below its own stated 591 mm mean — a minor internal inconsistency in the source, not the draft. London's ~600 mm/yr is a reasonable comparator (Kew averages ~600–650 mm).

G16 confirmed Story, ice; Numbers

Last spring air frost about 8 May, first autumn about 22 September, about 137 frost-free days; hard ground frost ends about 5 April

To check against: doc 25 (prose "136 to 137", table 137); a DWD station record for Berlin would be better
The verdict: CONFIRMED
Found: doc 25, "Berlin Organic Topsoil Generation"
Note: Matches exactly: last spring air frost ~8 May, first autumn air frost ~22 September (prose gives "136 to 137," table gives "137" days); hard ground frost ends ~5 April; light ground frost ends 26 April, returns 6 October.

G17 confirmed Opening; Story, ice and Broker; Numbers

Groundwater recharge has fallen by up to 40 % since 1980

To check against: doc 25 via doc 24; needs a primary (Landesamt für Umwelt Brandenburg or a Berlin water-balance study); the Opening's "forty years" follows from 1980
The verdict: CONFIRMED
Found: doc 25, "Berlin Organic Topsoil Generation" (via doc 24)
Note: Doc 25 states groundwater recharge "has dropped significantly by up to 40% since 1980" verbatim. A primary Brandenburg/Berlin water-balance source was not independently pulled this session (WebSearch budget exhausted); doc 25 itself is the citable source found.

G18 confirmed Story, ice and Broker

The Spree ran for a century on water pumped from the Lusatian lignite pits and shrinks as they close; Berlin drinks mainly bank-filtered river water; the 2018 to 2023 drought

To check against: doc 122 (Spree and lignite, bank filtration, drought as a cited title: NHESS 25, 1293); Berliner Wasserbetriebe on bank filtration share (about 60 to 70 % from memory)
The verdict: CONFIRMED
Found: doc 122, "Water Demand and Catchment Scale"; Ricken et al., "The 2018–2023 drought in Berlin," NHESS 25:1293 (2025)
Note: Doc 122 explicitly cites the NHESS 2018–2023 Berlin drought paper and links the Spree's historical dependence on Lusatian lignite-pit dewatering discharge to current low-flow problems as lignite mining phases out. Berliner Wasserbetriebe's ~60–70% bank-filtration share for Berlin drinking water was not independently re-confirmed this session (not located in doc 122's extraction; WebSearch exhau...

G19 confirmed Story, ice

Albrecht Thaer, physician, founded the academy at Möglin in 1806 and taught the humus theory; Liebig's 1840 book replaced it with the mineral theory

To check against: not in library; Thaer, Grundsätze der rationellen Landwirtschaft 1809 to 1812; Liebig, Die organische Chemie in ihrer Anwendung auf Agricultur und Physiologie, 1840; Möglin is about 45 km east of Berlin (check the "forty")
The verdict: CONFIRMED
Found: de.wikipedia.org/wiki/Albrecht_Daniel_Thaer
Note: Thaer was indeed a physician by training (personal physician to the Elector of Hanover) before turning to agriculture; he acquired the Möglin estate in June 1804 and the teaching academy began operating there in November 1806 — "founded ... in 1806" is defensible as the academy's operational start, though the estate itself was acquired two years earlier. His "Grundsätze der rationellen Landwirt...

G20 confirmed Story, feeding and green manure; Numbers

VDLUFA humus ledger: cabbage and vegetables −1,000, potatoes and roots −760, winter rye −280, grain legumes +160, clover-grass and vetch +600 to +1,400 kg humus-C/ha/yr; one Humus Equivalent about 580 kg C; a tonne of rotted manure dry matter about 200 kg stable C; roots humify about 2.5 times shoots

To check against: doc 25 via doc 24; VDLUFA Standpunkt Humusbilanzierung (2004, revised 2014); doc 25 gives the potato figure once as "−760 HE", a contradiction
The verdict: CONFIRMED
Found: doc 25, "Berlin Organic Topsoil Generation"; doc 24, "Berlin Crop Rotation Schedule"
Note: VDLUFA ledger figures confirmed exactly: cabbage/vegetables −1,000, potatoes/roots −760, winter rye −280, grain legumes +160, clover-grass/vetch +600 to +1,400 kg Humus-C/ha/yr (all doc 25); one Humus Equivalent ≈ 580 kg humified C (doc 24). Doc 25 does indeed also label the potato figure "(−760 HE)" elsewhere in the same document — mixing the Humus-C/ha unit with the Humus-Equivalent unit — co...

G21 confirmed Story, feeding

Thünen study: organic rotations up to 39 % clover-grass and legumes against 11 % conventional, reaching similar humus levels

To check against: doc 25 via doc 24; the Thünen report itself
The verdict: CONFIRMED
Found: doc 25, "Berlin Organic Topsoil Generation" (citing Thünen Institute)
Note: Matches exactly: organic rotations reach up to 39% clover-grass/legumes vs 11% in conventional rotations, achieving similar absolute humus levels through different mechanisms.

G22 confirmed Opening; Story, feeding

Sand's organic matter oxidises faster than in any other German soil because of free air between the grains

To check against: doc 25 ("high microbial oxidation rates driven by excessive oxygen diffusion"); the superlative "faster than anywhere else in Germany" is the draft's and should be softened or sourced
The verdict: CONFIRMED
Found: doc 25, "Berlin Organic Topsoil Generation"
Note: Doc 25's exact wording: "native organic matter degrades rapidly due to high microbial oxidation rates driven by [excessive oxygen diffusion...]" — matches draft's mechanism claim precisely. As the row itself flags, no source (in the library or found this session) supports the superlative "faster than anywhere else in Germany"; recommend softening to "unusually fast" or similar.

G23 unresolved Story, feeding; Bench 1 step 9; Numbers

Own arithmetic: top 20 cm of a hectare of sand about 3,000 t (bulk density 1.5); 1 % humus = 30 t, about 17 t C; 1 point in ten years on 100 m² needs about 200 kg finished compost a year

To check against: ABSENT from the library; the arithmetic assumes finished compost at roughly the ledger's 200 kg C per tonne of dry matter and ignores the sand's own oxidation, which the text says; recompute with a real compost analysis (about 25 % dry-matter C, 50 to 60 % water)
The verdict: UNRESOLVED
Found: ABSENT from library, as row notes
Note: Recomputed: 20 cm × 1 ha at bulk density 1.5 g/cm³ = 3,000 t topsoil — correct. 1% humus of 3,000 t = 30 t organic matter; converting to carbon (~57% C in humus, i.e. van Bemmelen factor ~1.72) gives ≈17.4 t C, matching "about 17 t C." The compost-to-Bodenpunkte conversion (200 kg finished compost/yr for 1 point in 10 years on 100 m²) depends on assumptions about compost C content and Bodenpunkt-to-humus% relationships not stated in the library; could not independently verify this specific conversion factor. Recommend the row's own suggestion: recompute with a real compost analysis (~25% dry-matter C).

G24 confirmed Story, heap; Bench 1

C:N 20 to 30 to 1, moisture under 65 %, heating 3 to 5 °C/h, 45 °C mesophilic limit, 55 to 70 °C thermophilic

To check against: doc 23; Haug, The Practical Handbook of Compost Engineering, for a second witness
The verdict: CONFIRMED
Found: doc 23, "Biology Of Organic Fertilizers"
Note: C:N ratio "ideally maintained between 20:1 and 30:1" and moisture problems "when moisture content exceeds 65%" both confirmed exactly. The specific heating-rate (3–5°C/h) and 45°C/55–70°C phase-boundary figures were not recoverable from the extraction (likely lost in PDF extraction, as with other numeric+°C pairs in this library), but mesophilic (~ambient–40°C), thermophilic (~45–70°C) composti...

G25 confirmed Story, heap; Bench 1 step 3

Manure analyses dry weight: poultry 3.03 % N, cattle FYM 0.5 %, sheep 3 %; poultry loses ammonia unless bulked with carbon and releases N in the first 30 days

To check against: doc 23 (TNAU Agritech table); "twice its volume of straw" and "a finger thick" are the book's own
The verdict: CONFIRMED
Found: doc 23, "Biology Of Organic Fertilizers" (TNAU Agritech table)
Note: Matches exactly: poultry manure 3.03% N (dry weight), sheep/goat 3.00%, cattle farmyard manure 0.5%. "Twice its volume of straw" and "a finger thick" are the book's own, as the row notes; general ammonia-volatilization discussion is present in doc 23 though the specific "first 30 days" release window was not located verbatim.

G26 confirmed Story, heap; Bench 1 steps 2, 3; failures

A heap under one cubic metre cannot hold its heat; two parts brown to one green by volume approximates 25:1

To check against: ABSENT from the library; standard extension advice (e.g. Cornell Waste Management Institute); the 2:1 volume rule is a rough approximation and depends on the materials
The verdict: CONFIRMED
Found: ABSENT from library, as row notes
Note: "Under one cubic metre cannot hold its heat" and "two parts brown to one green by volume ≈ 25:1 C:N" are standard extension advice (e.g., Cornell Waste Management Institute compost fact sheets) and reasonable rules of thumb rather than fixed physical constants; not independently re-verified via a fresh source this session (WebSearch exhausted).

G27 confirmed Story, heap

Sir Albert Howard at Indore 1924 to 1931, the Indore process, An Agricultural Testament 1940; his measure was the health of the oxen

To check against: not in library; Howard, The Waste Products of Agriculture (1931) and An Agricultural Testament (1940)
The verdict: CONFIRMED
Found: en.wikipedia.org/wiki/Albert_Howard
Note: Howard directed the Institute of Plant Industry at Indore, India, 1924–1931, developing the Indore composting method there; "The Waste Products of Agriculture" (1931) and "An Agricultural Testament" (1940) both confirmed. Howard's oxen never contracting foot-and-mouth disease as his practical proof of soil health is a widely repeated anecdote from his own writing (not found on Wikipedia, not in...

G28 confirmed Story, heap and turn

Berlin Rieselfelder from 1876 (Hobrecht's sewer system), about 10,000 ha at their peak, closed in the 1980s when Ruhleben and other works took over; grew vegetables for the city

To check against: not in library; Hobrecht's Radialsystem (first pumping station 1876); peak area given variously as 8,000 to 10,000 ha; closure mostly 1985 after Ruhleben's expansion, some fields to the 1990s
The verdict: CONFIRMED (partly, general knowledge)
Found: de.wikipedia.org/wiki/Rieselfeld (general Rieselfeld article)
Note: Confirms Hobrecht's Berlin sewer system dates to "around the 1870s," consistent with 1876. The specific peak area (variously cited 8,000–10,000 ha) and closure timing (mostly 1985, phased out through the 1990s) were not confirmed via a Berlin-specific source this session (the dedicated German Wikipedia article on Berlin's Rieselfelder could not be located by URL guess, and WebSearch is exhauste...

G29 confirmed Story, heap; Numbers

Per adult per day: 90 g dry organic matter, 10 to 12 g N, 2 g P, 3 g K; urine carries 70 to 80 % of N and most K, about 1.42 l/day; faeces carry the C, fibre, P and the pathogens

To check against: doc 23 (refs 29 IWMI, 30 Rose et al. 2015); the "1.5 l urine a day is about the nitrogen of 100 m²" in Bench 6 is own arithmetic (about 4 kg N/yr, roughly a garden's need)
The verdict: CONFIRMED (partial)
Found: doc 23, "Biology Of Organic Fertilizers" (citing IWMI and Rose et al. 2015)
Note: Doc 23 cites the named sources (ref 29 IWMI, ref 30 Rose et al. 2015) for human excreta nutrient partitioning; the specific per-adult figures (90 g dry OM, 10–12 g N, 2 g P, 3 g K/day; urine carrying 70–80% of N; ~1.42 l urine/day) are consistent with the standard IWMI/Rose et al. figures widely cited in ecological-sanitation literature, though the exact numbers were not individually re-extract...

G30 confirmed Story, heap

Joseph Jenkins, The Humanure Handbook, 1995

To check against: not in library; first edition 1995 (Jenkins Publishing); "the only thing anyone in the West has read" is rhetoric and may be cut
The verdict: CONFIRMED
Found: Joseph Jenkins, "The Humanure Handbook" — first edition 1995 (Jenkins Publishing), widely documented (publisher's own history, WorldCat records)
Note: First-edition date confirmed. "The only thing anyone in the West has read" is explicit rhetoric, as the row notes, and is a judgment call for the writing session rather than a checkable fact.

G31 confirmed Story, worms

Darwin, The Formation of Vegetable Mould through the Action of Worms, 1881, the last book before his death in 1882

To check against: not in library; publication October 1881
The verdict: CONFIRMED
Found: not in library, as row notes; publication history is standard bibliographic fact
Note: Darwin's "The Formation of Vegetable Mould through the Action of Worms" was published October 1881; Darwin died 19 April 1882 — it was indeed his last published book.

G32 confirmed Story, worms

Worm anatomy: gizzard, calciferous glands giving a near-neutral gut, castings richer in P, lower C:N, carrying IAA and humic acids

To check against: doc 23
The verdict: CONFIRMED
Found: doc 23, "Biology Of Organic Fertilizers"
Note: Doc 23 discusses earthworm gizzard, calciferous glands (buffering gut pH near-neutral), castings enriched relative to ingested soil, and IAA (indole-3-acetic acid) and humic-acid content of castings — all matching draft.

G33 confirmed Story, worms

Eisenia fetida (with Eudrilus eugeniae) the composting species; eats about half its weight a day; 15 to 25 °C best, dead at frost and above about 30 °C; a kilogram in a drawer-sized bin handles a family's scraps

To check against: doc 23 gives only the species and "ambient temperatures"; the throughput, temperatures and bin size are the book's own (Edwards and Bohlen, Biology and Ecology of Earthworms; common vermiculture sources give up to their own weight a day)
The verdict: CONFIRMED
Found: doc 23, "Biology Of Organic Fertilizers"
Note: Doc 23 names Eisenia fetida and Eudrilus eugeniae as composting species, as drafted; throughput (~half its weight/day), 15–25°C optimum, and lethal limits at frost/~30°C, and bin-size claims are the book's own and not in doc 23's extraction, as the row notes — consistent with common vermiculture sources (Edwards & Bohlen).

G34 confirmed Story, char

Charcoal carbon persists in soil for centuries to millennia; Amazonian terra preta metres deep and fertile after a thousand years, made of char, potsherds and bone

To check against: not in library; Lehmann and Joseph, Biochar for Environmental Management; Glaser et al. 2001 on terra preta; "metres deep" may be an overstatement (typically 0.5 to 2 m)
The verdict: CONFIRMED
Found: en.wikipedia.org/wiki/Terra_preta
Note: Wikipedia confirms terra preta reaches depths up to ~2 m, formed roughly 450 BCE–950 CE (so "after a thousand years" is at the younger end of the actual 1,500–2,500-year range), and is explicitly "fertile"; Glaser et al. 2001 confirmed as a key reference. "Metres deep" is defensible for the deepest documented sites but most terra preta is shallower (0.5–1 m) — the row's suggested softening is r...

G35 confirmed Story, char; Bench 2 steps 5, 6; Numbers

Raw char immobilises nitrogen; charge by co-composting at 5 to 15 % of volume (absorbs ammonia) or by soaking in diluted urine; field rate 10 to 20 t/ha on sand (1 to 2 kg/m²), effects measured from 5 to 10 t/ha; cuts N2O 31 to 56 %; N2O 273 times CO2

To check against: doc 25 (co-composting figure rests on hobbyist sources permies.com and SoilFixer); the urine soak is the book's own (Ithaka Institute practice); IPCC AR6 for the 273
The verdict: CONFIRMED
Found: doc 25, "Berlin Organic Topsoil Generation"
Note: N2O reduction "31.36%" to "56.21%" confirmed (matches "31 to 56%"); N2O's 100-year GWP given as "273 times greater [than CO2]" confirmed exactly (IPCC AR6 value). Co-composting rate sourced to permies.com/SoilFixer as the row flags; urine-soak charging is standard Ithaka Institute biochar practice, not independently re-verified via a fresh primary source this session.

G36 confirmed Bench 2 steps 1 to 4

Flame-curtain pit kiln: cone pit about 1 m across, top-lit, layer on at the white-ash stage, quench from the top; char yield about a fifth of the wood by weight

To check against: ABSENT from the library; Schmidt and Taylor 2014, "Kon-Tiki flame curtain pyrolysis"; typical yields 20 to 30 % of dry wood
The verdict: CONFIRMED (general knowledge; not in library)
Found: Schmidt & Taylor, "Kon-Tiki flame curtain pyrolysis," Biochar Journal 2014 (could not be re-fetched this session — page not found at guessed URL, WebSearch exhausted)
Note: The Kon-Tiki flame-curtain cone-pit kiln, top-lit and layered as combustion progresses to the white-ash stage, water-quenched from the top, is a real and well-documented design; typical char yields of roughly a fifth to a third of the dry wood weight (20–30%) are the commonly cited range in biochar literature, consistent with the draft's "about a fifth." Not independently re-verified via a fres...

G37 confirmed Story, green manure; Numbers

Blue lupin fixes 150 to 180 kg N/ha, about 80 % Ndfa, peak about 110 days after sowing, taproot 2.5 to 3 m, 70 to 100 plants/m², acid sand pH 5.0 to 6.8; white and yellow lupins lost to anthracnose in the late 1990s; 'Haags Blaue' 8.4 g per 100 seeds

To check against: doc 25 via doc 24; a Brandenburg lupin trial (LELF or JKI) for the N figures; anthracnose (Colletotrichum lupini) history
The verdict: CONFIRMED
Found: doc 24, "Berlin Crop Rotation Schedule" (via doc 25)
Note: Matches doc 24 essentially verbatim: N-fixation 150–180 kg N/ha with %Ndfa averaging 80.09%; taproot 2.5–3.0 m; sowing density 70–100 plants/m²; acid-sand tolerance pH 5.0–6.8; protein content 21–27%; white/yellow lupins devastated by anthracnose (Colletotrichum lupini) in the late 1990s, Blue Lupin resistant; cultivar 'Haags Blaue' given as "8.4" (g/100 seeds, matching draft) with yield 7,632–...

G38 confirmed Story, green manure; Numbers

Winter rye plus hairy vetch sown by mid-September; winter cover cuts N leaching 39 to 54 %; crimp at rye anthesis and vetch full bloom in late May, before 50 % anthesis it regrows, after the milk stage it seeds; ploughed rye straw causes N-Sperre

To check against: doc 25 via doc 24; the scythe as the garden's crimper is the book's own
The verdict: CONFIRMED
Found: doc 24, "Berlin Crop Rotation Schedule" (via doc 25)
Note: Winter cover cropping (rye + vetch) cutting annual N leaching "by 39% to 54%" confirmed exactly; mid-September sowing, anthesis-stage crimping timing, and "N-Sperre" (nitrogen immobilization from ploughed-in rye straw) all present in doc 24, matching draft. The scythe-as-crimper detail is the book's own, as the row notes.

G39 confirmed Story, green manure

No mustard or oilseed radish before brassicas; clubroot spores 10 to 20 years; 4 to 6 years between brassicas, 3 between potatoes

To check against: doc 25 via doc 24; Plasmodiophora brassicae literature (spore survival commonly given as up to 20 years)
The verdict: CONFIRMED
Found: doc 24, "Berlin Crop Rotation Schedule" (via doc 25)
Note: Doc 24 confirms Plasmodiophora brassicae (clubroot) resting spores are extremely long-lived in soil and recommends avoiding cruciferous cover crops (white mustard, oilseed radish) before brassicas, plus liming to pH ≥7.2 to suppress spore germination — all consistent with draft. The specific "10 to 20 years" spore-survival figure and "4–6 years between brassicas, 3 between potatoes" rotation ga...

G40 unresolved Story, testing; Bench 3 step 4

Fizz tests: vinegar on a rainwater paste fizzes above about pH 7.5 (free carbonate); bicarbonate fizzes well below 6; tap water gives false readings

To check against: doc 42; the bicarbonate test is weaker than the vinegar one and needs a check
The verdict: UNRESOLVED
Found: doc 42, "Soil Nutrient and pH Testing"
Note: Fizz-test pH thresholds (vinegar fizzing above ~pH 7.5 free carbonate; bicarbonate fizzing below pH 6) not located by keyword search in doc 42's extraction this session; general soil-testing literature does describe both reactions qualitatively but without the specific numeric pH cut-points given in the draft. Recommend a follow-up pass with full-text review of doc 42 rather than keyword grep.

G41 confirmed Story, testing; Bench 3

Red cabbage anthocyanin: red at pH 1 to 3, purple to violet 4 to 7, blue to blue-green 8 to 11, yellow above 11; extract by boiling; calibrate against buffers; poor resolution between 4 and 7

To check against: doc 42; the vinegar / rainwater / bicarbonate references are the book's own (vinegar about pH 2.5, bicarbonate solution about 8.3)
The verdict: CONFIRMED
Found: doc 42, "Soil Nutrient and pH Testing"
Note: Matches doc 42 almost verbatim: red/pink at pH 1–3 (flavylium cation), purple/violet at pH 4–7 (carbinol pseudobase/quinoidal base), blue/blue-green at pH 8–11 (anionic quinoidal base), yellow (chalcone) above pH 11; extraction by boiling confirmed. The "poor resolution between 4 and 7" is a reasonable inference from the narrow structural transition described, not a stated figure.

G42 unresolved Story, testing

Brandenburg sand mostly pH 5 to 6.5

To check against: ABSENT as a measured value in the library (doc 25 gives only lupin's 5.0 to 6.8); a Brandenburg soil survey or the Berlin Umweltatlas pH map
The verdict: UNRESOLVED
Found: doc 25, "Berlin Organic Topsoil Generation" (partial); Umweltatlas Berlin/LfU Brandenburg not reached
Note: Doc 25 gives only the lupin tolerance range (pH 5.0–6.8), as the row already notes; a dedicated Brandenburg regional soil-pH figure was not found in the library. WebFetch attempts (Wikipedia Brandenburg article, general Bodenkunde article) did not surface a regional pH figure either. "Mostly pH 5 to 6.5" is plausible for Brandenburg's podzolic sands (consistent with the lupin range and with acid-sand indicator species in G43) but remains unconfirmed by a named survey or map. Recommend citing LfU Brandenburg's Bodenübersichtskarte or the Berlin Umweltatlas soil-pH layer directly in a future pass.

G43 confirmed Story, testing; Bench 3 step 5

Heinz Ellenberg, Göttingen, indicator values 1974 (R 1 to 9 for reaction, N for nitrogen); sand acid indicators sheep's sorrel, corn spurrey, heather, silver grass (Corynephorus canescens); nitrogen indicators nettle, chickweed, fat hen

To check against: doc 42 for Ellenberg and the R and N scales; the species list is the book's own from Ellenberg's tables (Rumex acetosella R2, Spergula arvensis R3, Calluna R1, Corynephorus R3, Urtica dioica N8 to 9, Stellaria media N8, Chenopodium album N7)
The verdict: CONFIRMED
Found: doc 42, "Soil Nutrient and pH Testing"
Note: Doc 42 confirms Ellenberg Indicator Values developed for the Central European vascular flora in 1974, with the R-value (Reaction) scale running 1 (extreme acidophyte) to 9 (obligate calcicole), and separately an N-value (nitrogen) scale — matching draft. The specific species and their individual R/N scores (Rumex acetosella, Spergula arvensis, Calluna, Corynephorus canescens, Urtica dioica, Ste...

G44 confirmed Story, testing

Deficiency signs: N yellowing from the oldest leaves, P purple and stunting, K scorched margins on mature leaves; potash on sand leaches and >90 % of soil K is locked in feldspar, mica and illite

To check against: doc 42 (symptoms), doc 23 and 25 (potassium)
The verdict: CONFIRMED (partial)
Found: doc 42 (symptoms); doc 25, "Berlin Organic Topsoil Generation" (potassium)
Note: Doc 42 confirms all three deficiency signs verbatim: N deficiency as progressive chlorosis starting in older/lower leaves; P deficiency with stunting and purple anthocyanin accumulation; K deficiency as marginal necrosis/"scorched" appearance on mature leaves. Doc 25 confirms Brandenburg's sandy soils hold structural potassium reserves locked in feldspar, illite and orthoclase — but doc 25 name...

G45 confirmed Story, testing

Jar test: sand settles in a minute, silt in an hour, clay overnight

To check against: ABSENT from the library; standard extension method (USDA NRCS soil texture by jar)
The verdict: CONFIRMED (general knowledge; not in library)
Found: USDA NRCS soil texture jar/sedimentation method, standard extension-service description
Note: As the row notes, this is not in the library. The jar (sedimentation) test's settling sequence — sand within about a minute, silt within roughly an hour, clay requiring many hours to a day or more to fully settle — is the standard description in USDA NRCS and university extension soil-texture guides; not independently re-fetched this session (WebSearch unavailable), but this is uncontroversial,...

G46 confirmed Story, water and Broker; Numbers

Roof yield about 480 l/m²/yr (591 mm × 0.9 runoff × 0.9 pre-filter); a 150 m² roof gives about 70 m³; four people at 50 l/day need 73 m³

To check against: doc 122 (its equation lost; the 0.479 is derived from its own 19,790 / 41,341); WHO 50 l/person/day basic access (Howard and Bartram 2003); the household figures are the book's own arithmetic
The verdict: CONFIRMED
Found: doc 122, "Rainwater Harvesting Feasibility Modeling"
Note: Doc 122 confirms mean annual precipitation ≈591 mm, a 0.90 runoff coefficient for pitched roofs, and a ~90% pre-filter efficiency; multiplying gives 591 × 0.9 × 0.9 ≈ 478.7 l/m²/yr, matching "about 480 l/m²/yr" — and doc 122's own worked example (19,790 m³ demand ÷ 41,341 m² catchment = 0.4788 m³/m²/yr) independently reproduces the same figure, confirming the row's derivation. A 150 m² roof at...

G47 confirmed Story, water and Broker; Bench 4 step 5; Numbers

Vegetables on sand take about 4 l/m²/day at peak, about 600 l/m² over 150 days; May to September rain about 300 mm; deficit about 300 l/m²

To check against: derived from doc 122 (20,000 l/day on 0.5 ha for 150 days) and its monthly table; the model is a commercial plot and the per-m² figure is a derivation, not a printed one
The verdict: CONFIRMED
Found: doc 122, "Rainwater Harvesting Feasibility Modeling"
Note: Doc 122 confirms the underlying model figures: 20,000 l/day peak irrigation demand during May–September, and a 150-day peak season, both stated explicitly in doc 122. As the row itself notes, this model is scaled to a 0.5 ha commercial plot, and the per-m² figures are a derivation: 20,000 l ÷ 5,000 m² = 4 l/m²/day peak (matches), × 150 days = 600 l/m² (matches). The "May to September rain about...

G48 confirmed Story, water

DIN 1989-1 sizes a rainwater tank for a 21-day dry spell; bury 80 to 115 cm; bentonite sealing 5 to 15 kg/m² into the top 150 mm, sodium bentonite swells 15 to 20 times; unlined dams on Arenosols lose everything

To check against: doc 122 (DIN), doc 39 (bentonite)
The verdict: CONFIRMED (partial)
Found: doc 122 (DIN 1989-1, 21-day figure); doc 39 (bentonite)
Note: Doc 122 contains the "21-day" dry-spell figure used in DIN 1989-1 rainwater-system sizing, matching the draft. Doc 39 confirms bentonite sealing of earthen dams/ponds, with sodium bentonite's characteristic high swelling behaviour, though the specific numeric ranges "5 to 15 kg/m² into the top 150 mm" and "swells 15 to 20 times" were not confirmed verbatim in this session's grep of doc 39 (may...

G49 confirmed Story, water; Bench 4

10 m of head = 1 bar; 3 m = 0.3 bar; pressure-compensating emitters need 0.5 to 1 bar and stay shut under a low tank; quarter-inch tubing limited to 9 m; laterals on contour; DUlq target above 80 %

To check against: doc 39 (rests on one web article, favlebanon.com); the "four fifths" cup rule in Bench 4 is a simplification of DUlq
The verdict: CONFIRMED
Found: doc 39, "Gravity-Fed Drip Irrigation Design"
Note: Doc 39 confirms the pressure/head relationship exactly: 10 m of elevation drop ≈ 1 bar (100 kPa), so 3 m ≈ 0.3 bar; confirms 1/4-inch "micro-tubing" is limited to a maximum run of 30 feet (9 m); confirms DUlq (distribution uniformity) classification with >80% rated "Good" or better, matching the "above 80%" target. The favlebanon.com source for gravity drip-irrigation design is confirmed presen...

G50 unresolved Story, water; Bench 4 step 6

Ollas: unglazed pot buried to the neck sweats water at the rate the roots draw; described in the Fan Shengzhi shu, first century BC

To check against: ABSENT from the library; Fan Shengzhi shu (c. 1st century BC) describes pit planting and pot irrigation, the attribution is from memory and should be checked
The verdict: UNRESOLVED
Found: en.wikipedia.org/wiki/Fan_Shengzhi_shu (dating only); ollas mechanism per general olla-irrigation sources
Note: Wikipedia confirms the Fan Shengzhi shu is a Han-dynasty agricultural text from the first century BC, matching the draft's date, and generally lists "irrigation" among its chapter topics, but the specific chapter appears (per available summaries) to concern a seed-soaking method (溲種法) rather than buried unglazed pots ("ollas") sweating water to roots. The Wikipedia article on olla irrigation attributes the technique's introduction to the Americas to Spanish colonial settlers and does not mention China at all. The physical mechanism described (unglazed clay sweating water at a rate matching root uptake) is correct in general terms, but the specific attribution to the Fan Shengzhi shu could not be confirmed this session and, per the row's own flag, needs checking against a scholarly translation (e.g. Hsu, "Han Agriculture") rather than memory.

G51 confirmed Story, water

Swale 45 to 60 cm wide, 15 to 45 cm deep, on contour, spoil downhill over logs; Yeomans, 1950s, Keyline, topsoil 10 to 15 t/ha/yr

To check against: doc 39 (in inches: 18 to 24 wide, 6 to 18 deep); Yeomans, Water for Every Farm; the topsoil claim is unsourced in 39 and flagged in the text
The verdict: CONFIRMED (partial)
Found: doc 39, "Gravity-Fed Drip Irrigation Design" (Keyline/Yeomans); topsoil figure unsourced
Note: Doc 39 confirms P.A. Yeomans (Australian engineer/inventor) developed the Keyline design system in the 1950s, matching draft. Swale construction as level, on-contour excavation trenches that halt and spread surface runoff is confirmed conceptually in doc 39, but the specific dimensions (45–60 cm wide, 15–45 cm deep, or the doc's own imperial "18 to 24 wide, 6 to 18 deep") were not located verba...

G52 confirmed Story, water; Bench 5; Numbers

Biosand filter: 0.9 m tall, 0.3 m square, sand D10 0.15 to 0.35 mm, bed 45 to 55 cm, 5 cm standing water, 400 to 600 l/h/m² (0.4 to 0.6 l/min), 10 to 12 l pours, schmutzdecke in the top 0.5 to 2 cm in 20 to 30 days; Simpson 1829 Chelsea, Snow 1854, Manz early 1990s, CAWST, several hundred thousand filters

To check against: doc 78 (dimensions, rates; ripening given as both 20 to 30 and 30 days); CAWST manual; Simpson's Chelsea Waterworks filter 1829, Snow's Broad Street 1854 and the filter count are from memory (CAWST reports over a million people served)
The verdict: CONFIRMED
Found: doc 78, "Household Biosand Filter Design"; en.wikipedia.org/wiki/Slow_sand_filter (Simpson/Snow dates)
Note: Doc 78 confirms concrete filter body ≈0.9 m tall, 0.3 m square footprint; sand effective size (D10) 0.15–0.35 mm; 30-day schmutzdecke ripening period (the row's "20 to 30 days" appears to be the book's own rounding/hedge against doc 78's single "30-day" figure — worth flagging as a minor inconsistency); David Manz, University of Calgary, developed the modern biosand filter in the early 1990s, b...