The Gold Standard of Teaching Archaeological Fish Fin Ray and Spine Analysis | Age Reading, Resorption and Fishing Histories

Fish fin rays and spines can preserve growth zones that sometimes allow biologists to estimate age, and their detached remains may appear in ancient fisheries and settlement deposits. Yet annuality is not automatic: some species yield useful fin-spine ages, while published validation studies find severe underestimation or unvalidated increments in others. This chapter explains the anatomy, microstructure, resorption and archaeological source history needed before a ring or fragment becomes a defensible historical claim.

At a glance: identify the exact fin element; examine growth bands and the missing core; apply only species-validated ageing rules; compare readers against a genuine reference; separate archaeological fragments from fish individuals; preserve source and sampling information. Two transparent worked datasets below are fictional. The featured science figure is genuine modern Goliath Grouper research, not a specimen from the invented teaching contexts.

The essential contrast is illustrated by Carroll and colleagues’ 2023 Goliath Grouper study, the 2024 four-species Texas validation study and the 2024 White Sturgeon known-age evaluation.

1. A fin spine is not a complete fishing history

A fish fin ray or spine may preserve growth marks in hard tissue that can be examined in section. Under validated species-specific methods, those marks sometimes inform fish age, growth and the history of a population. When a fin element is recovered in archaeological sediment, however, additional questions arise: Which fish and body part is it? Was it deposited through fishing, predation, food waste or water transport? Does the preserved section include the earliest growth record? One ring count cannot answer all these questions, and a fossil fragment without provenance cannot automatically be converted into an ancient catch.

Start a lesson with three independent claims. A section shows several rings. The rings represent annual growth and indicate the fish’s age. Ancient residents caught that fish at one particular season. The first can be examined on an image; the second needs validation and structural completeness; the third needs contextual and potentially separate seasonality evidence. Students should identify the additional evidence required at each step, instead of letting one attractive microscope image become an archaeological narrative.

2. Distinguish rays, rigid spines and the otolith

Many bony fishes possess segmented soft fin rays, often built from paired skeletal elements called lepidotrichia, and some also have relatively rigid spines. Their morphology is not uniform across species. An otolith, by contrast, is a mineralised structure within the inner ear. All three may show growth marks, yet their composition, biology and preservation differ. Ray cross sections may have paired elements with a groove, while some spines display a central lumen and vascular change. The type of structure must be identified before a reading technique is selected.

Give students three real or documented reference micrographs and ask them to label the fish structure, where a core or inner growth region occurs, and the outer margin. Use the linked Goliath Grouper images for a genuine side-by-side comparison. The visual comparison is a foundation for method choice, not proof that one structure always gives a more accurate age.

3. Learn the difference between hard-ray anatomy and a loose archaeological fragment

A complete dorsal fin in a modern reference fish provides orientation, position and relationships among neighbouring rays. Archaeology usually offers detached, broken, burnt or modified remains separated from those landmarks. The same taxon can have multiple rays with different shapes and cross-sectional development. A fossil rod lacking a diagnostic base may therefore be hard to distinguish from another skeletal element or species. Morphology should be documented with scale, surface details and reference specimens before age or taxon is assigned.

Present one labelled modern fin with its spine positions and a bag of fictional recovered fragments. Learners should write which anatomical details remain available and which have been lost. A cautious broad identification may be a better scientific result than an unsupported assignment to dorsal spine number two. Every later statement about validated ageing must take this first uncertainty seriously.

4. What annuli actually measure

An annulus is a growth-related mark interpreted as periodic—often annual when validated—but it is not identical to every narrow line visible in a hard structure. Fish growth may slow or change with seasons, food availability, reproduction and environmental stress. These processes can leave distinct optical zones, checks or crowding. The timing of the zone must be established in the species and region before a count becomes years of age. A beautifully regular circle is not a guarantee of annual formation.

Ask students to annotate a real section with clearly observed light and dark features and then use a separate key to identify which kind of zone a study counts. In the 2023 Goliath Grouper study, translucent zones were used for rays and spines while opaque zones formed the count basis in otoliths. Mixing those conventions without reading the methods would create false comparisons even when every image was clear.

5. Section location changes what the analyst sees

The cross-sectional plane determines which growth marks are visible. A cut too far distally or obliquely may miss central early zones or distort ring geometry. A cut extremely close to a complicated base can introduce different problems, including asymmetry and resorption, depending on species. Fisheries laboratories therefore document which ray or spine was selected and the anatomical level at which sections were prepared. Archaeological fragments often lose exactly the base-to-tip information needed to apply those standards confidently.

Students should inspect a diagram of one ray cut at three different positions. They mark which section contains the best chance of a complete history, which may omit early information and which is difficult to interpret because of orientation. The response is not a universal instruction to cut at one distance, but a justification tied to a species-specific validated method. Specialist equipment and protected specimen handling require trained laboratory oversight.

6. Distinguish occlusion from central resorption

Fin spines and rays can change internally during growth. Vascular encroachment can obscure early marks, while resorption can remove hard tissue containing part of the original record. The two phenomena may both make inner annuli unreadable but do not necessarily produce the same microscopic appearance. In some spines earlier bands can remain partly visible despite encroachment; in some rays resorption removes them. If early annuli are lost, even perfectly counted outer zones will underestimate age.

Make two schematic sections: one with a vascularised area covering visible early rings and another with a true central void where those rings are gone. Students should determine what is observed and what cannot be reconstructed. They should not simply add a guessed number of missing years unless a correction has been validated for that species, age range and sampling location.

7. Why an older fish can look implausibly young

Resorption, central occlusion and crowding of the newest marginal bands can create age underestimation that grows with fish age. It may be tempting to give every adult one or two extra years, but the error is not necessarily constant across individuals. Structure thickness, fish growth, reproductive history and section level can all vary. A correction derived for one species or age range should not be transferred automatically to a different fish, even one from the same family.

Use a fictional fish whose well-supported otolith age is twenty-four years while the fin-ray reading shows twenty-two. Ask whether that proves two missing annuli for every archaeological fish. It does not. The observed discrepancy belongs to the particular fish and protocol. Learners should look for independent known-age data or statistically justified validation before proposing any general adjustment.

8. The 2023 Goliath Grouper finding: a positive case

Carroll and colleagues evaluated dorsal fin spines, rays and otoliths of Atlantic Goliath Grouper in a modern fisheries research programme. Their paper documented growth-zone timing, core issues and age comparisons, and found dorsal spines particularly useful for a nonlethal-ageing strategy after appropriate corrections. The article’s featured research figure comes from this study, showing genuine comparative cross-sections rather than fictional archaeology. It demonstrates that well-designed, species-specific work can make fin structures informative, not that every fin structure in every fish works.

Students should outline the evidence behind the conclusion: matching structure type, reference otolith ages, repeated readings, seasonal margin assessment, observation of occlusion and evaluation of bias with age. Then they should explain why a disconnected ancient spine fragment lacks some of those controls. A successful modern result is a guide to what archaeological validation would require, not a licence for automatic conversion.

9. The 2024 Texas study: an important negative result

A 2024 North American Journal of Fisheries Management study investigated otoliths, spines and rays in Guadalupe Bass, Channel Catfish, Gray Redhorse and Redbreast Sunfish. Its validation work did not confirm annual increments in the tested fin-ray or spine structures, even while some otoliths yielded validated annual growth. This is a valuable counterexample to the claim that visible fin bands are always yearly marks. A methodology can be excellent for one study organism and unsupported for another.

Invite students to compare the positive Goliath results with the Texas validation study. The point is not to decide which scientists are correct in all settings; their research questions and species differ. The teaching conclusion is conditional: a band is an age indicator only when its periodic meaning is established for the actual taxon and method. Unvalidated archaeological fin rings should be reported as observed zones rather than assumed birthdays.

10. The 2024 White Sturgeon known-age test

The US Geological Survey reported a study of known-age White Sturgeon sampled through a long-running mark-recapture programme. Among 162 fish, between-reader agreement on fin-ray ages was 58.7 percent, while consensus agreement with known ages was only 30.7 percent and declined with age. The same study found that fin-ray measurements could support back-calculation of lengths under an appropriate model even though ageing was unreliable. This distinction is crucial: a structure can be useful for one variable and unsuitable for another.

Ask learners to explain why agreement between readers and agreement with reality are different. Two analysts may reproduce the same incorrect missing-annulus count. Conversely, a geometric measurement can be valuable even when the chronological interpretation is weak. A report that says the method is globally useless would overlook the size-related information the study retained.

11. Age validation is not the same as reader agreement

Reader agreement measures how consistently observers assign the same count, while validation asks whether those counts correspond to an independently known age or deposition period. A training group may agree perfectly because all members learned the same flawed counting rule. Known-age fish, suitable mark-recapture experiments, chemically marked specimens and appropriate reference structures can support validation under specialist research protocols. None is recreated by a classroom discussion alone.

Have one pair of students count an image twice and another pair compare to a documented known-age reference. Require them to report both exact agreement and error against reference. Then ask which statistic is relevant when someone claims ancient fish lived for twenty years. A reliable age claim needs support for chronological meaning, not only a low disagreement percentage.

12. Learn from the 2024 fisheries-ageing handbook

The Atlantic States Marine Fisheries Commission and Gulf of Mexico fisheries ageing handbook explains that rays and spines can lose early growth information through resorption and that the appropriate cross-section level depends on species. It also distinguishes apparent checks from actual annuli and recommends validation of deposition periodicity. A handbook helps standardise professional work, but it should be read as guidance for selecting and testing a protocol rather than a ready-made age key for every archaeological object.

Ask students to create a methods checklist from the handbook: taxon, exact ray or spine, sample location, cross-sectional orientation, visibility of early growth, marginal checks, reader comparison and validation reference. If the ancient object lacks three of these facts, the report must explicitly lower the strength of its age interpretation.

13. A shark fin spine is not the same ageing system as a teleost ray

Some cartilaginous fishes, including spiny dogfish, are aged using dorsal fin spines, but their skeletal anatomy and validation history differ from teleost fin-ray systems. A 2024 NOAA workshop on Atlantic Spiny Dogfish documented decades of debate, new validation evidence and the need for standardised readings. Students must avoid treating a sturgeon fin ray, a goliath grouper spine and a dogfish dorsal spine as biologically interchangeable merely because each can show bands.

Give three scientific reference captions without their species names and ask learners what contextual information they require before comparison. The correct response emphasises anatomy and validated methodology. It is valuable for young researchers to learn that a measurement label such as ‘annual spine ring’ may hide different organisms and operational definitions.

14. Nonlethal sampling is a modern research concern

Modern fish scientists may sample a ray or spine without killing the animal, particularly for protected species. This benefit is important, but neither removal method nor welfare impact is universal; it depends on species and procedure. Archaeologists typically work with material already detached, discarded or preserved in collections. They should not copy fisheries extraction instructions as a classroom experiment or assume that nonlethal modern sampling makes destructive handling of a protected fossil acceptable.

The teaching task is to compare three ethical questions: how to minimise harm when collecting data from living animals, how to limit destruction of historical specimens, and how to preserve analytical reproducibility. A suitable school or university lesson can use genuine published images and fictional data to teach the scientific reasoning without handling live protected fish or cutting museum objects.

15. Taxonomy comes before age or fishery history

Archaeological fin elements may be identified to a broad fish group, yet a species-level assignment can be difficult when they are fragmentary. A reliable reference collection should include regional fish fauna and skeletal positions. If a species cannot be identified, applying a species-specific age-growth curve would be unjustified. Even broad length estimates may not transfer between fishes that differ in fin size and body proportions.

Ask students to evaluate a fragment labelled ‘probably a large fish’. Can it support age, catch size, habitat or trade origin? The answer differs for each question. They should first record morphology, preservation, stratigraphic association and plausible taxonomic alternatives, then propose measurements capable of increasing diagnostic confidence.

16. Field recovery, mesh size and sorting bias

Fine fin rays and small spines may be overlooked in hand-collected excavations, while robust elements survive and are recovered more readily. Sample mass, sieve aperture, sorting time and surface condition determine what enters the documented assemblage. Two archaeological deposits processed differently cannot be compared simply by their raw bone-fragment totals. Larger fish may also produce pieces more readily recognised by a sorter, creating a false impression of size selection.

Compare a waterlogged occupation layer recovered through measured fine sieving with a dry layer collected by eye. Students must request the processed sediment mass, recovery fraction and recording decisions before concluding the ancient fishing strategy changed. Sampling methodology is part of the historical evidence.

17. Recognise burning, digestion and water transport

Fin bones can be modified by cooking, burning, predator digestion, trampling, water movement and later burial chemistry. Charred surfaces do not necessarily prove a fish was cooked in a human meal; burning can occur in secondary rubbish or natural fire contexts. Abrasion can round diagnostic ends, and resorption within a living fish should be distinguished from damage after death. All these processes affect identifiability and the visibility of growth bands.

Set out four hypothetical finds: a blackened ray fragment near a hearth, an abraded spine in a channel, a relatively intact ray in a sealed refuse lens and a specimen with central resorption but little exterior damage. Have students propose alternative histories and define which microscopic or stratigraphic features could distinguish them. One dramatic surface character cannot carry the entire interpretation.

18. Worked dataset A: independent-age agreement

In a wholly fictional demonstration, twelve modern comparison fish have both a documented reference age and a fin-structure age estimate. The imagined age readers agree exactly with the reference for seven fish, differ by one year for three fish and differ by more than one year for two. Exact agreement is 7 / 12, about 58.3 percent. Agreement within one year is (7 + 3) / 12, or 83.3 percent. These figures show how two different metrics can make the same dataset appear more or less reliable.

The figures are invented, not a reproduction of the USGS sturgeon sample or the 2023 Goliath data. Students should compute both, then ask whether the two larger errors occur in older fish. Without inspecting error by age, a single average may conceal exactly the historical ages most at risk of underestimation. A strong paper reports the pattern of disagreement and its relevance to the claim being made.

19. Worked dataset B: average signed error

Expand the fictional comparison to five clearly illustrative reference ages: 3, 5, 9, 16 and 24 years. Corresponding fin-age estimates are 3, 5, 8, 14 and 21. The signed errors (estimate minus reference) are 0, 0, −1, −2 and −3 years. Their sum is −6, giving a mean signed error of −1.2 years. The fact that errors become more negative in this invented age-ordered subset is consistent with an underageing problem, but the tiny selection is not a validated predictive model.

Students should write a second table with absolute errors, 0, 0, 1, 2 and 3, whose mean is also 1.2 years in this particular case. In other datasets mean signed error and mean absolute error can differ dramatically if overestimates and underestimates cancel. The exercise teaches why one number alone cannot describe accuracy. No universal ‘+1.2 years’ correction follows from this toy exercise.

20. A fragile cross-section may preserve partial information

A damaged central region may prevent reliable total age estimation while some outer increments or dimensional features remain observable. Whether those observations support a partial growth history or size-related inference depends on species, reference calibration and source location. A report should not describe the specimen as either entirely worthless or conclusively dated; it should separate which measurements remain defensible from those invalidated by damage.

Ask learners to annotate what the sample still shows: section outline, visible outer bands, degree of occlusion and any preserved baseline landmark. They then write a claim limited to the visible features. An argument for further analysis should specify what uncertainty the new measurement could realistically resolve.

Fictional reference-agreement calculation

21. Age-at-catch is not the date of archaeological disposal

Even a correctly aged ancient fish, if one could validate that estimate, yields a biological age-at-death or growth-history parameter rather than the calendar date of capture, processing or deposition. Fish can be dried, traded, stored and discarded later; bones can be moved again after burial. Archaeological dating must be supplied by independent stratigraphy and appropriate chronometric or documentary evidence. A twenty-year-old fish buried in a medieval layer did not necessarily spend its whole life near that settlement.

Draw a timeline for fish growth, death, possible preservation, trade, consumption, discard and secondary redeposition. Have students identify which observation constrains each event. The distinction is essential before interpreting ancient fishery effort, market organisation or seasonal occupation.

Fictional five-fish signed-error example

22. Size selection and the body-region problem

Some ageing structures can also be measured in relation to fish body length, but those relationships must be established for the species, section location and population. The 2024 White Sturgeon analysis, for example, found useful back-calculated length results despite poor fin-ray age accuracy. That is a reminder not to discard all quantitative questions when age reading fails. Archaeological fragments, however, may lack the full geometry needed to apply modern regressions accurately.

Give students a published species-specific measurement model and then an unknown archaeological ray fragment. They must assess whether the same anatomical landmarks are preserved and whether the taxon is genuinely comparable. A sensible conclusion may remain ‘body-size category uncertain’ rather than applying a convenient equation outside its validation domain.

23. From bone counts to fishing intensity: two distinct denominators

An excavation may produce many fin rays or spines, but a fish has multiple fin elements and one element can break into pieces. A raw NISP-style count—number of identified specimens—does not directly equal number of caught animals. In addition, sediment mass and excavation recovery conditions can affect how many fragments are recovered per sample. Assemblage comparisons should report both archaeological count units and processed context, with minimum-animal estimates only when skeletal sidedness and duplication support them.

Use fictional trench A with sixty diagnostic ray fragments from 2 kg of processed sediment and trench B with ninety from 6 kg. The simple rates are thirty and fifteen fragments per kilogram. B has more raw fragments but a lower observed concentration under the imaginary shared protocol. Neither rate is a direct estimate of fish harvested per year; chronology and skeletal recovery remain necessary.

24. When seasonal growth zones are insufficient

A section’s terminal growth pattern may sometimes inform timing of formation in well-studied modern fishes, but archaeology adds uncertainty about death, capture, trade and disposal. Even a modern period-of-year classification requires validated species-specific deposition timing and an intact outer margin. Broken or eroded fin structures cannot simply be assigned a capture month by comparison with a generic calendar diagram. The chapter on archaeological otoliths develops chemical seasonality questions under their own separate controls.

Ask students to explain what a damaged ray could still reveal if the outer margin is absent. A cautious result might record interpretable interior zones without claiming season. If an older paper assigns a precise month from such material, the student should request the validation reference and marginal preservation evidence before accepting the claim.

25. Build a reproducible archaeological specimen record

Each specimen needs a unique identifier, site and stratigraphic context, taxonomic basis, fin element and anatomical position if known, raw surface photographs, section-plane information, preparation history, tissue condition, visible bands, central occlusion or resorption, reader identity and confidence. Counts and calculated ages should be stored in separate fields so an independent reviewer can reconstruct the inference. A revised interpretation should preserve the original record and reason for change.

Set an archival handover test: one group provides a result saying ‘fish age 14’ but no image, specimen ID or protocol. Another group must identify what data are missing. The better report includes a scale-labelled section, band annotations, taxon and known limitations. This is not paperwork added after the science; without it, the chronological statement cannot be independently audited.

26. Preserve protected specimens and distinguish teaching from cutting

Archaeological fish bones may be rare, friable or culturally and legally protected. Creating a transverse section can consume unique material or damage relationships needed for future methods. Destructive sampling should be justified by an answerable research question, a specialist protocol and the appropriate permission. Modern fisheries techniques intended for live fish require their own animal-welfare authorisations. Teaching can build excellent method awareness through licensed scientific figures, original diagrams and fabricated data rather than improvising specimen cutting.

Students should draft a sampling decision memo explaining the information expected from sectioning, the chance of a reliable result, specimen significance, less destructive alternatives and archival retention. A scientifically sound answer may decide that a fragment is too incomplete to sacrifice for a low-probability age estimate.

Fictional archaeological recovery denominators

27. Publish a conclusion that matches the structure

An archaeological fin-ray or spine study should report its observed anatomical structures, sampling and preservation limitations, any age validation, raw counts and chronology alongside the proposed historical interpretation. If the taxon is uncertain, central annuli are missing and no known-age reference exists, an exact age distribution is not supportable. The finding may still document particular fish groups or deposit formation processes, especially when combined with otoliths, vertebrae and other fish remains.

The fictional lessons demonstrate two transferable principles: a numerical comparison depends on the denominator, and a visible growth band gains a time meaning only through biological validation. The gold standard is not the most dramatic catch history. It is a research report that another reader can check, challenge and build upon without first undoing hidden assumptions.

Practical case audits: independent questions for students

Every case below is an invented decision exercise. These are not records from a specific archaeological excavation, and each case should be evaluated using the relevant biological or depositional evidence.

1. A convincing ring with no validation

An intact fin spine has eight visible zones and the reader announces an eight-year-old fish. Students ask which species, anatomical position and known-age reference confirm annual formation. They record eight apparent zones rather than eight proven years until validation is available.

2. The two-photo anatomy test

One image shows paired ray elements with a groove, while another depicts a spine with an internal lumen. Learners correctly label the structures and explain why their resorption and sectioning challenges can differ. A generic ‘fish bone cross-section’ label is inadequate for age-method selection.

3. A resorbed centre

The outer bands are crisp but inner tissue is missing. Students annotate what has genuinely been lost and determine why the visible count may be a minimum, not the total age. Their next step is to seek a validated correction, not invent a missing-ring number.

4. The older sturgeon case

A team cites 58.7 percent agreement between White Sturgeon readers as proof of high age accuracy. Students find the separate 30.7 percent agreement with known ages and explain what each number means. The exercise trains interpretation of actual validation statistics.

5. The successful Goliath method

Students use the 2023 Goliath Grouper comparison to claim every fin spine can be aged reliably. Their peers identify the role of species, anatomical position, marginal checks and correction factors and restate the narrower result.

6. The Texas negative result

A fin-ray study on a new fish species produces regular-looking bands. Students compare this with the 2024 central Texas species validation study and explain why annual deposition must be demonstrated rather than assumed.

7. A section cut off axis

A prepared ray cross-section misses the expected central landmark. Learners show why the ring path can be distorted and what information would help determine whether reliable age reading remains possible.

8. The stress check counted as a birthday

A narrow extra mark appears after an unusual environmental episode. Students distinguish possible checks from independently validated annuli and describe how repeat readers and modern references assess the classification.

9. The fragment from unknown body position

A detached archaeological ray lacks its original fin location. The researcher applies a protocol for one precisely sampled modern spine. Students identify the anatomical mismatch and adjust the conclusion to what remains supported.

10. The raw NISP increase

Trench A yields sixty fragments from two kilograms, and trench B ninety from six. Students calculate thirty and fifteen fragments per kilogram and explain why the raw increase reverses under a sediment denominator without becoming a fish population estimate.

11. The two competing depositional histories

A charred fin spine near a hearth is called dinner waste. Students propose secondary burning and transported remains as alternatives and list cut marks, heating patterns, association and context tests that might discriminate them.

12. An imported fish bone

Fin fragments from a marine fish appear at an inland site. Students consider dried-fish trade, transport and site formation before reconstructing a prehistoric shoreline. Their report separates taxon identification from geographic origin.

13. Two age readers, one wrong method

Readers agree perfectly on a set of unvalidated growth checks. Students explain why agreement does not prove those marks are annual and propose independent known-age or marked reference data.

14. The absent last annulus

A specimen’s outer edge is broken. Learners decide why the last growth increment and possible seasonal status remain uncertain, despite a clear interior count. A precise capture month would be unjustified.

15. A corrected-age overreach

One species-specific paper adds years to ageing-structure estimates at certain age thresholds. A student exports the same arithmetic to a different archaeological fish. Reviewers ask for an independent calibration before permitting correction.

16. The size-back-calculation puzzle

A validated study estimates fish lengths from fin-ray dimensions even when ages are unreliable. Students describe how the two outcomes can coexist and why a fragment missing measurement landmarks may still be unusable for the length model.

17. A low-recovery layer

Few small spines are recovered from a coarse-sieved context. Learners assess screening and sorting biases before claiming the disappearance of small fish. They develop a comparable recovery plan.

18. The last peer-review pass

A fictional manuscript gives fish age, fishing season and annual catch from a few loose spines. Reviewers retain identifiable skeletal observations and request age validation, depositional chronology and skeletal count rules before the larger claims can proceed.

A four-part teaching and assessment sequence

• Module 1: recognise a fin ray, rigid spine and otolith; sketch authentic cross-sections with labels for the outer margin, core, structural groove and obscured annuli.
• Module 2: distinguish observed growth checks from validated annual zones; study contrasting 2023–2024 research results and the role of true known-age references.
• Module 3: work the two fictional age datasets, compare accuracy and agreement, and diagnose resorption, reader and section-level errors.
• Module 4: audit recovery, fragmentation, chronologies, environmental transport and ethical specimen handling; produce a reproducible archaeological conclusion.

Research-grade reporting checklist

Frequently asked questions

Verified scientific sources and connected eduKate Gold Standard chapters

• Carroll et al. (2023), Dorsal Fin Spines and Rays for Nonlethal Ageing of Goliath Grouper, Fishes 8:239
• Pfennig et al. (2024), Age Estimation and Validation in Otoliths, Spines and Fin Rays from Four Central Texas Fishes
• Ghere et al. (2024), Known-age White Sturgeon and fin-ray ageing accuracy, USGS
• Atlantic States Marine Fisheries Commission, Practical Handbook for Determining Fish Ages (2020 edition)
• NOAA Fisheries, Age and Growth
• Indian Ocean Tuna Commission (2024), first dorsal-fin spine ageing protocol for Skipjack Tuna
• 2024 Spiny Dogfish Ageing Workshop proceedings, NOAA
• Teaching Archaeological Fish Scale Analysis
• Teaching Archaeological Otolith Analysis
• Teaching Zooarchaeology
• The Gold Standard of Palaeoecology
• Gold Standard master hub

A reliable fin-ray or spine analysis is one where the structure, counting rule, validation, uncertainty and archaeological source history are all explicit. An incomplete growth record can still be useful evidence, but it should never be promoted into a precise historical fish age, catch season or annual harvest size merely because a neat set of rings is visible.