R07 Ledger in Plain English — Where Cannabis Roots Live

Companion to Claim Ledger v1.0 (signed off 2026-09-10). Each entry restates the claim for a general reader, says how solid it is in plain words, and points at the evidence. Full citations, grading, boundary lines, the eleven-field root-zone descriptor and the evidence-type tags live in ledger.md and SOURCE_LIBRARY (S-219–S-240, plus reused entries).

A practitioner-record passage is withheld from the public copy.

Four root-zone classes used throughout: water culture (roots in solution: DWC, NFT, aeroponics) · inert substrate (rockwool, perlite, clay pebbles, foam, fed by solution) · fibrous soilless substrate (peat, coir, bark, wood fibre — “organic-matter-based” describes what it is made of, not how it is fed) · soil (native soil, amended beds, mixes with real mineral soil). Most “soil” in the studies is bagged peat mix, which is the third class, not the fourth.

One warning that applies to most of the head-to-head entries (C007–C017): they come from two laboratories (Hohenheim in Germany and Guelph in Canada), often with the same plant varieties. They are several experiments, not several research groups. The only finding two independent labs share is that soilless media carry root pathogens (C024).

What the labels mean, and what coco actually is (C001–C006)

C001 — KEYSTONE: the science measures root-zone systems, not “soil,” “coco” or “hydro.” Solid. What the medium is made of and how water and nutrients are delivered are separate dials, and every result belongs to the whole bundle: substrate and grade, pot size, drainage, plumbing, watering frequency, root temperature, water chemistry and feed. The papers themselves call a peat pot on synthetic feed “hydroponic”; Danny’s archive taught “coco is hydro” for eight years. Here every system is re-filed into its class, and a comparison that can’t be described that way is set aside.
C002 — KEYSTONE: in the only trial that grew cannabis in three root-zone systems side by side, the peat pot on synthetic feed grew 42–116% more flower than the two fish-fed setups. Promising (one study, two varieties, abstract only). It cannot be read as “hydro beats soil”: the winning “hydroponic” arm was an 11-litre peat pot, and the three arms differed at once in feed, pot size and plumbing. The gap is well verified; why cannot be untangled. A second researcher’s description of this paper (60-litre tubs, other varieties, a water temperature) was fabricated; do not repeat it.
C003 — KEYSTONE: “coco” is not one material. Solid, from ornamental-plant substrate science rather than cannabis. Across 13 coir sources, salinity ran from 39 to 597 mS/m and the fibre’s capacity to hold nutrient ions from 31.7 to 95.4 cmol(c)/kg, and both varied within a source too; 17 more sources arrived loaded with potassium, sodium, calcium and magnesium. A number quoted for “coco” means nothing without the source, the grade (pith, chip, fibre or blend), how it was compressed and what pot it is in.
C004 — Fresh coir can grab calcium and magnesium out of your first feeds, and rinsing it is not the same as buffering it. Promising on the mechanism, unknown on the size. Nobody has measured how much Ca/Mg cannabis actually loses to unbuffered coir, and no cannabis trial has compared buffered with unbuffered coir, so “coco always locks out calcium” is rejected as a rule; one researcher’s “40–100” figure and “mandatory buffering” were not in its cited papers. Danny’s sites mostly ran RO water, exactly when this chemistry matters most, and never had it lab-tested.
C005 — Roots respond to pot size, substrate physics, water, oxygen, temperature and feed. Solid as general horticulture, with no cannabis number in it. Background only.
C006 — The “container ladder” (start small, pot up in steps, don’t drown a small plant in a big pot) is convention, not a measured effect. Honestly unknown for the effect; solid that the practice exists. No cannabis trial has isolated pot size with everything else fixed. The reasoning (roots suffocating in a wet oversized pot) rests on what Danny saw at a clone counter and his own overfill, and the oxygen mechanism has never been measured in cannabis.

Head to head: substrates, water and plumbing (C007–C018)

C007 — The same substrate can halve one variety’s yield and leave another untouched. Promising (one lab, two varieties, abstract figures). ‘KANADA’ made 8.6 g of flower per plant in peat but only 3.8 g in coir and 4.9 g in green fibre; ‘0.2x’ held 7.9–9.2 g in all three. A substrate ranking from one variety does not carry over to another.
C008 — Rockwool, peat-perlite and coir, each watered on its own schedule, grew differently in leaf and stem but made flower with the same CBD percentage. Promising (one study, one strain, abstract only). The authors put the differences down to water-holding and irrigation management, not the material. The watering was deliberately not matched, which is the point: it was a comparison of bundles. No figure is quoted until the full paper is read.
C009 — Across five coir/peat/perlite mixes on the same feed, how much water the mix made available to the plant was the main thing that predicted growth. Promising (one study, a CBD hemp variety). The two coir-fibre-plus-perlite mixes grew least. Measured physics, not the name on the bag, correlated with performance.
C010 — Within coir, the drier of two formulations gave 29% more dry flower in the flowering stage. Promising (one lab, abstract only). The authors suggest drier coir may suit flowering. Read with C009, which favoured more available water in the growth stage: the direction changes with stage, so “drier is better” is not a rule.
C011 — Organic liquid fertilizer works in coir, and its best rate depends on which coir. Solid (full text; inherits R06-C008). The flowering optimum was 212–261 mg N/L and differed between two coir mixes; the vegetative trial found one optimum (389 mg N/L) for both.
C012 — In a Thai trial, a peat:perlite:vermiculite mix grew taller, thicker plants with more flower than the coir-based mix, which had greener leaves. Promising at best (abstract only; plant count, pot size and durations unverified). No number is drawn from it yet.
C013 — KEYSTONE: with the same clay pebbles and the same feed, recirculating the nutrient solution instead of draining it away raised total THC harvested, but not flower weight. Promising (one study, one clone, clay pebbles, and the two setups ran one after the other, not side by side). Extra phosphorus, potassium and iron did nothing conclusive. The advantage showed mainly after week nine, so the authors suggest harvesting recirculated plants at week eleven. A total THC result, not a yield result; the “+182%” that circulates is recirculating-versus-drain-to-waste under the enriched feed only.
C014 — Salt-stress “crop steering” on rockwool made both varieties 15% shorter, cut flower yield about 20% in a variety-dependent pattern, and did not raise potency. Promising (one study, full text). No compensating gain in cannabinoid percentage.
C015 — Dry the plants back and you get three different answers from three studies. Solid that responses differ; nothing poolable. One study saw 12–13% higher THCA/CBDA percentages and 43% more THCA harvested under drought; one saw yield loss with no potency gain; one saw water savings with no loss. Three varieties, three systems, three definitions of stress.
C016 — Moderate deficit watering cut water use without costing biomass or CBD; severe drought hurt and gained nothing. Promising (one lab, two CBD varieties, numbers pending the full text). A water saving from how you water, at a fixed medium, not from the medium.
C017 — Burying the drip line cut irrigation water 18.6% outdoors, with 5% more flower and 9% more CBD. Promising (one study, full text, three CBD varieties in field soil under a tunnel). A plumbing effect on water use; it does not transfer to indoor water culture.

A practitioner-record passage is withheld from the public copy.

Mechanisms in the root zone (C019–C021)

C019 — Under salt stress, plants in a mineral nutrient solution suffered at 40 mM sodium chloride while plants in a fish-based (aquaponic) solution did not; cannabinoid percentages fell with salt in both. Promising (one study, abstract only, full text pending). Solution chemistry changed how salt stress landed.
C020 — Nobody has measured how fast things go wrong, or how fast they can be fixed, in one system versus another. Honestly unknown. “Hydro is unforgiving, soil buffers mistakes” is a plausible idea from general substrate physics; in cannabis it is grower convention. Danny’s record supports the direction (an uncontrolled reservoir and one overfill killed a water-culture crop) and nothing more.
C021 — The report ends with a property map, not a winner. Structural, not evidence. Systems are rated on what they let you control (buffering, root oxygen, feed precision, water use, recirculation, biology, media reuse, monitoring, failure speed, automation, cost and labor), and in cannabis almost every cell reads “unmeasured.” One distinction kept from a researcher whose numbers were withdrawn: “how soon you see a problem” and “how fast you can fix it” are different things, and a forgiving medium may be slow on both.

Taste and chemistry (C022–C023)

C022 — KEYSTONE: no blind taste test of soil versus soilless versus water-culture cannabis exists. Honestly unknown, and the report’s strongest absence. “Soil tastes better,” “hydro is bland” and “coco gives cleaner terps” are all anecdote. Danny’s archive never actually claimed soil tastes better.
C023 — Substrate changes flower chemistry; nobody has shown it changes what people taste. Solid on the distinction. Every paper that reports cannabinoids or terpenes by substrate stops at the lab result.

Disease and safety (C024–C028)

C024 — KEYSTONE: soilless media are not sterile in practice. Solid, and the one finding two independent labs share. Disease-causing Pythium, Globisporangium and Fusarium were recovered from coir and rockwool under sick plants in a Connecticut facility, and Pythium root rot was confirmed in British Columbia greenhouses. “Sterile media, fewer root diseases” is rejected. But “soil has fewer diseases” is equally unsupported: nobody has ranked disease rates by medium on a fair basis.
C025 — Recirculating solution is a documented route for disease to spread. Solid (abstract-verified). Fusarium was detected in the recirculating nutrient solution of a hydroponic facility sampled over three years. How reservoir temperature, oxygen and pathogens interact has never been measured in cannabis; chiller thresholds are convention.
C026 — Rockwool fibre has workplace exposure limits, and nobody has measured what cannabis workers actually breathe. Solid on the limits (U.S. NIOSH and OSHA values plus a non-governmental ACGIH guideline, as of September 2026); unknown for cannabis exposure to rockwool, peat dust or reservoir biofilm.
C027 — There is no USDA-organic cannabis; California’s OCal is the “comparable-to-organic” program. Solid, as of September 2026 (inherits R06-C004).
C028 — “You can’t grow organic in hydro or coco” is false as an absolute. Solid on feasibility. Organic liquid feed in coir has a measured dose curve, and aquaponic cannabis (fish waste turned to nitrate by bacteria) has been grown to harvest. The catch is real: organic nutrients need biology and time, which water culture and inert media only supply through a biological compartment. Whether it competes is C002, where the synthetic-fed peat pot grew 42–116% more flower than the two fish-fed setups. Certification is separate.

Water, waste and footprint (C029–C031)

C029 — KEYSTONE: “hydroponics uses 90% less water” is a leafy-greens number, not a cannabis one. Solid that the figure is borrowed; honestly unknown for cannabis. No study meters water use by root-zone system; the real cannabis savings (deficit watering, buried drip) come from irrigation management at a fixed medium. Indoors, about 70% of applied water came back as air-conditioner condensate in one metered study — recoverable, not reused. Water applied, drained, recirculated, recovered and discharged are different numbers and are never pooled.
C030 — “Rockwool goes to landfill” is a direction, not a measurement. Honestly unknown for the amount. No study weighs spent media per unit of flower or tracks where it goes; reviews call rockwool single-use and hard to recycle. Danny’s record: rockwool left the rooms “by the bag, weekly,” destination unknown; greenhouse soil was reused, count unknown.
C031 — Coir versus peat footprint rankings flip depending on what you measure and where. Promising, from horticultural life-cycle studies rather than cannabis. One Spanish study put coir lowest in five of eight impact categories; a 2026 European dataset put peat lower on a single score. These are per cubic metre of medium, not per kilogram of flower, and rockwool was not in them. No soil/coir/rockwool ranking is claimed.

Skill, labor and money (C032–C034)

C032 — No study compares the cost, labor or failure risk of soil, coir, rockwool and water-culture systems with output held to the same yardstick. Honestly unknown. “Hydro costs more, soil is cheaper” has no denominator. Danny’s 2019 “organic beds drastically cut costs” was an informal 65-gallon-pot experiment with no cost or yield numbers recorded.
C033 — The one cannabis economics study varies flowering strategy, not root zone. Solid as reported; adjacent. Five two-month autoflower cycles versus four three-month photoperiod cycles a year moved the returns, and it is the only cycles-per-year figure in the corpus.
C034 — KEYSTONE: nobody has compared a living-soil bed with a soilless or hydro system across several cycles under matched conditions. Honestly unknown, confirmed by two independent searches. Every “soil vs hydro” result is a single-cycle pot trial, and almost every “soil” in it is bagged peat mix.

A practitioner-record passage is withheld from the public copy.

Practitioner-record section withheld from the public copy.

Source file in the research repository: research/reports/R07-soil-vs-coco-hydroponic/ledger-plain-language.md. Private links, some names and practitioner-record passages are removed from the public copy, and broken character encoding from file conversion is repaired; nothing else is changed.