Hold a litre of Magic Power up to the light. Strong-tea colour, slightly opaque, a faint earthy smell that anyone who has worked around live compost recognizes.
This post is what a laboratory found in it. Every figure below comes from one of four documents: NviroTek Wynland Laboratories chemistry report S26/3362 and culture microbiology report M26/9725, the DNA sequencing report M26/9726, and the independent interpretation by Elrica de Necker of Eco Veritas, SACNASP registration 139526. All four describe one sample, AF4_1, drawn on 30 June 2026. The full tables and the accreditation status of every result are on the evidence page.
Nothing below is inferred from what a product of this kind is assumed to contain. Every genus named is one the sequencing found, at the share of reads the report gives it.
Where the liquid comes from
Catfish grow in a closed-loop tank on a controlled feed. Their manure settles into a collection layer. That residue is captured and processed aerobically, with the tank’s own microbial community already established in it, until the result is a stable concentrate. Water is biofiltered and recirculated, and only the volume that leaves as product is replaced. Inside a closed-loop fish farm covers the engineering.
The interpretation report describes the output as a fermentation product, and one of the genera it found, Macellibacteroides at 2.3 percent of reads, is a strict anaerobic fermenter that in the report’s words "confirms the product’s fermentation origin" (page 9).
The chemistry, per litre
From report S26/3362, page 1 of 2:
| What | Reported | Inside the lab’s accreditation schedule |
|---|---|---|
| Total nitrogen | 421.5 mg/L | No |
| of which ammonium | 244.0 mg/L | Yes |
| Potassium | 216.8 mg/L | Yes |
| Phosphorus | 16.5 mg/L | No |
| Calcium | 105.6 mg/L | Yes |
| Magnesium | 34.6 mg/L | Yes |
| Sulphur | 29.3 mg/L | No |
| pH | 7.96 | Yes |
| Electrical conductivity | 374 mS/m | Yes |
| Total solids | 0.12 percent | No |
Eleven of the twenty-six analytes on that report sit inside the laboratory’s schedule of accreditation. The rest carry an asterisk that the report’s own legend defines as outside it, and the evidence page marks each row rather than presenting the panel as uniformly accredited.
Arsenic, cadmium, cobalt, chromium, mercury and lead all came back below the detection limit. Molybdenum came back at 1,690 ug/L, which is high, with copper below detection. That pairing matters and it comes up again below.
The arithmetic nobody in this industry likes stating: 421.5 mg/L is 0.42 g of nitrogen in a litre. A kilogram of urea carries about 460 g. This is not a nitrogen product, and any claim that it replaces one is contradicted by the report printed beside it.
The biology, measured two ways
Culture, report M26/9725. Total plate count reported as greater than 30,000,000 cfu/g. Read that as the method’s upper reporting ceiling rather than a measurement: the true count is higher and unknown, and it should never be restated as an exact figure or converted per litre. Yeast at 10 cfu/g, which the interpretation calls negligible. Mould at 3,200 cfu/g, which it calls moderate and normal for a biological ferment. All three results are inside the accredited scope.
Sequencing, report M26/9726. 952 bacterial taxa, 516 of them resolved to species, at a Shannon diversity index of 4.49 on 13,762 reads. The interpretation report states a second read total of 13,327 in its species appendix and we have asked the laboratory to reconcile the two figures rather than pick one.
The genera the sequencing actually found
From the interpretation report, page 9, with the report’s own descriptions:
| Genus | Percent of reads | The report’s reading |
|---|---|---|
| Pseudomonas | 19.7 | Many species solubilise phosphate, produce siderophores, suppress disease. Largely positive. |
| Aeromonas | 9.8 | Water bacteria from organic-rich, low-oxygen environments. An opportunistic pathogen, a hygiene concern, not a plant pathogen. |
| Comamonas | 9.1 | Break down organic compounds, involved in denitrification. Neutral-beneficial. |
| Flavobacterium | 7.9 | Break down organic material, recycle nutrients. Mostly neutral-beneficial. |
| Acinetobacter | 7.8 | Mixed: some solubilise phosphate and promote growth, others are opportunistic. |
| Macellibacteroides | 2.3 | A strict anaerobic fermenter. |
| Citrobacter | 1.8 | Enteric bacteria; can fix nitrogen and solubilise phosphorus, but faecal-associated and opportunistic. |
And the sentence under that table, which is the one a buyer should read twice:
This is a microbe-rich product likely manufactured mainly through fermentation processes. There is no dominance of the classic soil inoculants such as Bacillus, Azospirillum, Rhizobium, Lactobacillus, Trichoderma or mycorrhiza. This means the product can typically be used to build and support general microbial life in the soil, not to apply specific target species.
Bacillus, Lactobacillus and Trichoderma are the three a buyer usually asks after. The sequencing found none of them dominant.
On the fungal side, 99.2 percent of the fungal DNA was unclassifiable against the reference database. Among the fraction identified were water fungi consistent with a fermentation origin, and two Aspergillus species capable of producing aflatoxins, detected at 9 and 8 reads out of roughly 13,900 fungal reads. That is DNA, not a viability test and not a toxin measurement. It is still a food-safety flag, and it sets the pre-harvest rule below.
What is not measured
Enzymes and metabolites. Proteases, cellulases, phosphatases, organic acids, B-vitamins and hormone-like compounds appear on no report we hold, and none of them has been quantified in this product. The mechanisms are real in the wider literature for microbial communities of this kind. Until an assay exists, they are a hypothesis, not a specification.
Batch consistency is also unmeasured. What exists is one panel on one sample. A finished-product panel on a retail batch is commissioned, together with a pathogen panel, an aflatoxin B1 assay and a viable-count series at 3, 6 and 12 months for shelf life.
Chain of custody is incomplete. No document ties AF4_1 to a specific container, site or batch, so the panel describes a sample rather than a product specification, and the evidence page prints "Not recorded" against those fields.
How to apply it
Two regimes, and they are not the same thing.
The season programme, which is how the product is sold: 15 to 25 litres per hectare banded in the starter fertilizer line at planting, capped at 30 where fertigation events follow (FishIt integration guide, section 1). Dilute at least 1:10 with water and place the band about 5 cm to the side of or below the seed, never in seed contact. The reason is the ammonium at 244 mg/L and the salt load at EC 374 mS/m, either of which will scorch germinating seed (interpretation report, page 11).
Continuous fertigation, where a grower runs one: an injection concentration of up to 1 part product per 1,000 parts irrigation water. That is a concentration at the injector and nothing else. The annual volume per hectare is set by how much water the crop takes. Read "1 litre per 1,000 litres" as a concentration at the injector, never as a dose per hectare.
No foliar rate is published. The safety data sheet lists foliar under uses advised against, and phytotoxicity, pathogen and residue testing is commissioned but not complete.
The cautions that travel with it
The interpretation report lists six points of attention and all six publish, in full, on the evidence page. Three of them change how you handle the drum.
Wear gloves and eye protection. Aeromonas and Citrobacter are handling-hygiene concerns rather than plant-pathogen risks, and the report says so plainly.
Do not apply to edible parts close to harvest. On grain and feed, apply early in the season or to the soil. The opportunistic bacteria and the aflatoxin-forming Aspergillus DNA are the reason.
Watch molybdenum on pasture. Molybdenum is high and copper is absent, and a wide molybdenum to copper ratio can produce copper deficiency in ruminants. The report recommends adding copper, not more molybdenum.
The sixth point is the one that governs the whole programme: the product is nutritionally incomplete, low in phosphorus, with no measurable copper, zinc or boron, and low carbon and solids. In the report’s words, "on its own, the product is unbalanced and low in plant nutrients, and will therefore not replace a complete fertility program".
Storage and format
Two formats: 30 litre containers for smaller plots and 220 litre drums for cooperatives and large operators. Store cool, dark, ventilated, not frozen, not in direct sun, closed between uses. Shelf life is not yet measured. The safety data sheet claims 24 months with no stability data behind it, and the viable-count series that would settle it is commissioned. Record the production date of every drum.
Pricing is on application.
What a litre is for
One litre carries a large, diverse, living bacterial community and a small quantity of plant nutrition. Used inside the programme, the argument for it is nutrient-use efficiency and soil biological function. The argument against reading it as a fertilizer is the nitrogen figure at the top of this post.
That is the claim the documents support, and it is the only one this page makes. The documents themselves, with report numbers, dates and the laboratory’s own contact details so you can check directly, are on the evidence page. The paired-plot protocol that would test the programme on your land is on the trials page.