The ReefTox Methodology: How Live-Coral Ecotoxicity Tests Are Run
ReefTox tests substances and products on live coral grown in its own closed-loop laboratory in Brittany, France. Fresh fragments are cut from cultivated mother colonies of up to 50 species from 30 genera, exposed for 96 hours with no recovery period in the standard acute test, or for 3 months plus 1 month of recovery in the optional chronic test, and scored for bleaching, tissue loss, mortality and behavioural anomalies. For cosmetics, a free predictive pre-evaluation, included with a paid test, screens the formulation before any coral is exposed. Protocols follow OECD guidance principles.
Published 26 October 2026.
The ReefTox process in six steps
- 1
Cultivate the corals
Mother colonies of up to 50 species from 30 genera are grown in the closed-loop laboratory, and test fragments are cut in-house from them.
- 2
Screen before exposure (cosmetics only)
A free predictive pre-evaluation, included with a paid test, scores the formulation on six pillars plus water analysis and screens it against Hawaii, US Virgin Islands and Palau ingredient laws, then routes it to a compatible coral panel.
- 3
Expose fresh fragments
Fresh fragments are exposed for 96 hours with no recovery period, with the medium renewed at 48 hours. Formulations are tested at 50 mg/l; ingredients and marine test articles across a concentration range.
- 4
Score the endpoints
Bleaching (computer-assisted), tissue loss, mortality and morphological or behavioural anomalies are recorded with before and after photography.
- 5
Add chronic exposure where needed
An optional chronic test runs 3 months of exposure plus 1 month of recovery across 5 concentrations to derive EC10 and EC50 values.
- 6
Report
Results are reported per species and endpoint. For cosmetics, a passing formulation can carry the ReefTox Tested seal of approval, scoped to the panel tested.
Why cultivated coral and fresh fragments?
Testing on adult fragments, sometimes called nubbins, lets a laboratory run many replicates from the same colonies at any time of year, an approach described in the coral literature since at least Shafir et al. (2003). Growing the mother colonies in-house removes the need to collect from reefs for each test and keeps the history of the test animals known.
The ReefTox collection covers up to 50 species from 30 genera, hard small-polyp, hard large-polyp and soft corals, representative of Indo-Pacific, Caribbean and Mediterranean reefs. Each test uses fresh fragments, never reused. The facility itself is described on the lab page.
The free predictive pre-evaluation (cosmetics only)
Before any coral is exposed, a cosmetic formulation is scored on six formulation pillars (oil phase, emollients, nitrogen and phosphorus compounds, preservatives, humectants and emulsifiers, UV filters), analysed for trace-element release into water, and screened against the ingredient lists of Hawaii Act 104, US Virgin Islands Act 8185 and Palau RPPL 10-30.
The output is a risk classification and the coral panel the formulation is compatible with. It is a predictive step, not a live-coral test, and it is complimentary and included with a paid test. Details are on the free pre-evaluation page.
How is the acute coral test designed?
All acute tests share the same exposure: 96 hours of contact, no recovery period, test medium renewed at 48 hours. What changes is the depth (number of concentrations) and the breadth (number of species).
| Test | Tests | Design | Output |
|---|---|---|---|
| ReefFilter | SPF ingredients | 1 to 3 corals, 6 concentrations, 6 replicates and 6 blanks per concentration | EC50, LOEC, NOEC |
| ReefTox 1 and ReefTox 3 | Finished formulations | 1 or 3 species, 1 concentration (50 mg/l), 6 replicates, 6 blanks | Observed acute effects per species |
| ReefTox 12 and ReefTox 50 | Finished formulations | 12 or up to 50 species, 1 concentration, 2 replicates, 1 blank | Cross-species screen |
The 50 mg/l formulation concentration is based on the SCCS reference of 36 g of sunscreen per application, diluted in 750 litres of water, as a realistic maximum for crowded reef tourism sites. Single-concentration tests report observed effects and do not produce EC50, LOEC or NOEC values; that requires a concentration range, as in ReefFilter. Why panel breadth matters is explained in multi-species coral testing.
The format, a 96-hour semi-static exposure of adult fragments, belongs to the same family as the adult-fragment assay proposed for standardisation by Brefeld et al. (2024). How these design choices compare across the field is covered in coral toxicity testing methods.
When is the chronic add-on relevant?
Reef Chronic 1 runs 3 months of exposure followed by 1 month of recovery, across 5 concentrations for EC10 and EC50 determination plus 1 confirmatory concentration. Three months covers three lunar cycles, and the recovery month shows whether effects reverse once exposure stops.
It is optional. It answers a different question from the acute test, the effect of continuous exposure, and is most relevant for persistent ingredients or for continuous discharges in marine programmes. Regression-based EC10 values are the type of result statistical guidance such as OECD (2006) favours over NOEC values.
Endpoints, quality controls and reporting
Endpoints
Bleaching (computer-assisted observation), tissue loss, mortality, and morphological or behavioural anomalies such as polyp debilitation and mesenteric filament ejection, with before and after photography.
Quality controls
Blanks in every test, fresh fragments with no reuse, and annual positive controls with both mineral and organic UV filters.
Reporting
Results per species on a three-level scale: no acute ecotoxicity observed; temporary stress indicators without necrosis; or acute ecotoxicity with necrosis.
One method, two applications
Cosmetics and SPF
Formulations and SPF ingredients, starting with the free pre-evaluation. A formulation that passes its chosen panel can carry the ReefTox Tested seal of approval, a single seal whose scope is the panel tested. See coral testing for sunscreens.
Marine and industrial
Coral as a complementary species alongside the algae, invertebrates and fish already in a marine ecotox programme, for antifouling, ballast water, scrubber wash water and other discharges. No pre-evaluation and no seal apply. See the marine methodology.
What the method does not claim
- •A result applies to the species, concentration and conditions tested. It is not a statement that a product is harmless to every reef.
- •Adult fragments in controlled conditions do not capture every life stage or every field stressor, such as heat waves. Published work shows that early life stages and warmer water can change sensitivity.
- •No coral test guideline exists yet. Systematic reviews such as Ouédraogo et al. (2023) show how varied coral test designs remain across the literature.
Frequently asked questions
Where does ReefTox get its corals?
From its own mother colonies, cultivated in a closed-loop laboratory in Brittany, France. Test fragments are cut in-house and used once; no fragment is reused between tests.
How long does a ReefTox coral test take?
Coral exposure in the acute test lasts 96 hours, and acute results are typically delivered in 2 to 3 weeks. The optional chronic test runs 3 months of exposure plus 1 month of recovery.
Does every ReefTox engagement include a chronic test?
No. The 96-hour acute test is the standard path. The 3-month chronic test is an optional add-on for cases where continuous exposure needs to be characterised.
Is the free pre-evaluation available for marine projects?
No. The pre-evaluation is built for cosmetic formulations. Marine programmes are scoped directly with the ReefTox team around the test article, the concentration range and the endpoints the dossier needs.
Is there an official coral test guideline that ReefTox follows?
No OECD Test Guideline for corals exists yet. ReefTox protocols follow OECD guidance principles: replicated exposures with blanks, defined concentrations, renewal of the test medium and recorded, reviewable endpoints.
Sources
Background literature, checked 26 October 2026. ReefTox protocol parameters are ReefTox's own.
- Shafir S, Van Rijn J, Rinkevich B (2003). The use of coral nubbins in coral reef ecotoxicology testing. Biomolecular Engineering 20:401-406
- Brefeld D et al. (2024). Acute toxicity assays with adult coral fragments: a method for standardization. Toxics 12(1):1
- Downs CA et al. (2016). Toxicopathological effects of the sunscreen UV filter, oxybenzone (benzophenone-3), on coral planulae and cultured primary cells. Archives of Environmental Contamination and Toxicology 70:265-288
- OECD (2006). Current Approaches in the Statistical Analysis of Ecotoxicity Data: A Guidance to Application. OECD Series on Testing and Assessment No. 54
- Ouédraogo D-Y et al. (2023). What are the toxicity thresholds of chemical pollutants for tropical reef-building corals? A systematic review. Environmental Evidence 12:4
Discuss your test design
Formulation, ingredient or marine discharge: tell us what you need to show, and we will propose the design.