Who is writing this page?
I have taken this medical examination around twenty times, and I have practiced most of the tests it contains on the other side. But let's be clear from the first line: I am not a doctor. This page makes no diagnosis and judges no fitness. It does one single thing: show you what the text actually says, and where it says nothing.
In short
Colour vision falls under MED.B.075, not MED.B.070, which deals with the visual system. The first-line test is the Ishihara, and failing it concludes nothing: it triggers further tests. In Class 1, the applicant must be referred to the authority's medical assessor, and the criterion is "normal trichromat or colour safe". In Class 2, even after failing further tests, the intended outcome is not unfitness but a daytime flying limitation, an outcome that does not exist in Class 1. An often ignored point: the regulation contains no numbers, all thresholds live in the acceptable means of compliance (AMC).
First things first: which class are we talking about?
Everything that follows depends on it, so we might as well state it upfront. Class 1 is the certificate for professional licences CPL, MPL, ATPL: it is the one you take if you are aiming for the profession. Class 2 is for private flying (PPL). The LAPL is a lighter leisure licence, with its own medical certificate and its own rules.
On colour vision, these three are not judged in the same way, and that is exactly why the testimonials you read online contradict each other: two pilots can recount two opposite experiences, without either of them being wrong.
"Colour safe". You will come across it everywhere. The regulation uses it without defining it, and I am not going to define it instead. What can be said about it without extrapolating: it is not "having normal colour vision", it is a status established by the second-line tests described below. This is why the text writes "normal trichromat or colour safe": two different ways of meeting the same criterion.
The "medical assessor". This is the doctor of the licensing authority, not your AME (the approved aeromedical examiner who sees you in their practice). When the text mentions "referral to the medical assessor", it means going from the practice doctor to the authority doctor.
D'abord, le bon article : MED.B.075
This is the error most often encountered in the sources, including in content that looks serious: quoting MED.B.070 to talk about colour blindness. MED.B.070 is the visual system, visual acuity, refraction, visual field, eye surgery. It says nothing about colour vision.
The article dealing with colour vision is MED.B.075, in Annex IV (Part-MED) of Regulation (EU) 1178/2011. Its current version comes from Implementing Regulation (EU) 2019/27 of 19 December 2018, which replaced Subparts A and B of this annex.
When an article, a forum or a school talks to you about colour vision quoting MED.B.070, they have not opened the text.
What regulation applies?
It is MED.B.075 "Colour vision" of Annex IV (Part-MED) of Regulation (EU) 1178/2011, in its version resulting from Implementing Regulation (EU) 2019/27. The most common pitfall is stopping at MED.B.070: this article deals with the visual system (acuity, field, surgery) but says nothing about colors.
MED.B.075 is a short text, built in two stages. First, the initial test (Ishihara). Then, if it fails, what happens depending on your medical class.
| Stage | Class 1 | Class 2 |
|---|---|---|
| Initial examination | Ishihara test, for the initial issue of the certificate | |
| If Ishihara fails | Referral to the licensing authority's medical assessor, and further testing | Further testing, without mandatory referral to the medical assessor |
| Criterion to be met | Be a normal trichromat or colour safe | Have a satisfactory color perception |
| If further tests fail | Unfit. The regulation does not provide for any limitation in class 1: it's fit or unfit. | No unfitness, limited to day privileges only |
Reading the text changes two things. Failing the Ishihara decides nothing: it does not end the examination, it triggers the next step. And class 1 and class 2 do not follow the same circuit: class 2 has an intermediate outcome, the limitation, which class 1 does not have.
An honest clarification on class 2: the text uses two wordings, "colour safe" for the further tests, "satisfactory colour perception" for the limitation, without specifying if they exactly overlap. I won't decide for it. And if the word "unfit" on the last line stops you: appeal processes exist, they are covered on the unfitness and appeal page.
Pilot's point: the regulation contains no numbers
This is the point that took me the longest to understand, and it explains half of the misunderstandings on the subject. MED.B.075 gives no numbers: neither number of plates, nor threshold, nor tolerance, nor test name, other than "Ishihara". Everything that is quantified lives one floor down, in the acceptable means of compliance (AMC), published separately by EASA.
Practical consequence: these thresholds can be modified by an Agency decision, without going through the European legislator. A number read in a dated article is therefore not guaranteed to be up to date, including the numbers on this page: they correspond to the AMC wording resulting from the January 2019 decision, and I have not been able to collate the most recent revision of the Easy Access Rules. Check the current version before relying on a value.
Not to be misunderstood either: an AMC is not an optional suggestion. It is the default route, the one your AME follows. The difference lies elsewhere: these figures move more easily than a regulation.
The Ishihara test: the exact criterion
The criterion is found in the AMCs (AMC1 and AMC2 MED.B.075), resulting from an EASA decision in January 2019. It fits in one sentence, and each of its three elements is regularly distorted:
- The 24-plate edition. This is the version of the test used. It is not the number of plates to pass.
- The first 15 plates, presented in a random order. The AMC specifies the random order; it does not give the reason.
- Identified without error. The test is considered passed at zero errors. There is no "we tolerate two or three plates".
You will see "you have to pass all 24 plates" circulating: it's false, the 24 is the edition. You will also see "you are allowed a few mistakes": it's false for classes 1 and 2. The only tolerance threshold that exists in the system concerns the LAPL, and only for night flying; I'll get to that below.
If Ishihara is not passed: secondary tests
The regulation simply says "further colour perception testing". It does not name them. It is the AMCs that list the three accepted tests, each with its own criterion:
| Test | What the AMC says |
|---|---|
| Anomaloscopy (Nagel type) | Passed if the color match is trichromatic and if the matching range is 4 scale units or less. |
| Lantern test | Passed if the applicant passes without error a test with an accepted lantern. The AMCs cite the Spectrolux, Beynes and Holmes-Wright lanterns. |
| CAD | Passed if the threshold is less than 6 SN units for a deutan deficiency, or less than 12 SN units for a protan deficiency (the two forms of red-green color blindness; "SN" for standard normal, the scale specific to the CAD test). A threshold greater than 2 SN units in tritan suggests an acquired cause, to be investigated. |
A word about the last line, because it's not there by chance: the AMC does not just set a tritan threshold, it writes that exceeding it "indicates an acquired cause which should be investigated". This is the only place in the system where the text explicitly refers to a medical work-up. What that means in a given case, only a doctor can say, I do not interpret it here.
Beware of the British source. The UK CAA does not accept the lantern as proof of being colour safe. The EU AMCs, however, explicitly accept it. This is a real post-Brexit divergence, and it's a frequent trap: the UK CAA is often the most readable source found when searching online, but since Brexit it no longer describes European law. The same warning applies to American content: the FAA is a separate system, which I have not examined here.
What about the LAPL? The "9 out of 15" and the misunderstanding it creates
The article that governs the LAPL medical examination, MED.B.095, does not mention color vision. That is a fact, and it often surprises people.
The requirement does exist, however, but one level down, in AMC14 MED.B.095, and only for the addition of a night rating: the applicant should correctly identify 9 of the first 15 plates of the 24-plate edition, or be colour safe.
This is where the most widespread misconception among applicants comes from: the "9 out of 15" circulates as if it were the general threshold for the Ishihara test. It is not. It is the threshold for the LAPL night rating, and nothing else. For both class 1 and class 2, the criterion remains "without error".
The "day flying" limitation: the VCL code
In class 2, the outcome provided for by the text is not unfitness but limitation to day privileges. On the certificate, this limitation carries a code: VCL, for "Valid by day only".
Two clarifications to avoid mistakes. The VCL code does not appear in the regulation: the article that lists the limitations, MED.B.001, only names OML, OSL, OPL, ORL and SSL. VCL comes from the AMC level (AMC2 MED.B.001). Writing "the regulation provides for the VCL code" is therefore inaccurate. And do not confuse VCL with VDL, which concerns distant visual correction: similar acronyms, different subjects. Visual correction codes are covered on the page dedicated to glasses for pilots.
I already hold a certificate: will I have to take it again?
The regulation only requires the Ishihara test for the initial issue of the medical certificate. After that, the answer is in the AMCs: at revalidation and renewal examinations, colour vision should be tested on clinical indication. In other words: not systematically.
This formulation does not distinguish between situations, whereas medicine distinguishes them clearly, and this explains the logic of the clause. A congenital colour blindness is stable: it does not degrade over the years. Once it has been assessed at the initial examination, there is nothing new to detect. An acquired disorder, however, can appear and evolve, and is sometimes the first sign of an ocular or general impairment. It is on this side that the interest of a check during a career lies: to detect a change, not to re-establish what is already known. How your examiner applies this in your specific case, on the other hand, is up to them.
However, I will stop there. Many pilots want to know if there is a form of acquired right (grandfathering) that would protect a holder whose colour vision degrades over time. I have found no such provision, nor can I state that none exists. Not finding is not proof of absence. If this is your situation, this is exactly the kind of question to ask your AME, who has access to the guidance from their authority.
France, Belgium, Switzerland, Luxembourg: what changes
A recurring question is: "Is it harder in France than in Belgium?" Based on the official sources I have been able to consult, none of these four countries adds a numerical national threshold. What differs is procedural: who decides, and through which channel one appeals.
- France. In the mandatory equipment list for approved AMEs, the only required chromatic test is the Ishihara test neither anomaloscope nor lantern. The logical consequence: second-line tests are done somewhere other than the AME's office. Exactly where, and through what referral circuit, I have found no official French source that describes it. This is a question to ask your AME directly, as they know their circuit. Good to know as well: the DGAC does have a written Ishihara procedure, but its content is not public (sections marked "Reserved" in its practice guide). One can therefore assert neither that it is identical to the AMC, nor the opposite. As for appeals, the authority is the CMAC (Conseil Médical de l'Aéronautique Civile): the two-month deadline that appears in the DGAC documentation applies to the general appeal procedure, and not specifically to a chromatic failure.
- Belgium. No national requirement on colour vision. The specificity is procedural: according to the Royal Decree of 12 July 2013, a medical review committee is provided for the further checks of "borderline cases", then an appeals commission (text consulted on an unofficial mirror, to be confirmed in the Belgian Official Gazette). The detailed circuit is covered on the unfitness and appeals page.
- Switzerland. No national requirement found on the subject. The solid point: the FOCA Medical Section (AMS) constitutes the first instance of appeal in the event of a conflict between an aviation professional and a doctor.
- Luxembourg. No additional Luxembourg requirement was identified in the official DAC sources, which refer to the European regulations.
A reservation I prefer to put in writing: rules can exist without being public. Several authorities have guidance intended only for their AMEs. "I found nothing" does not mean "there is nothing".
The most widespread misconceptions
| What's circulating | What the text says |
|---|---|
| "Colour blind = unfit, full stop" | In class 2, failing the further tests gives a day-flying-only limitation, not unfitness. In class 1, unfitness likewise only occurs after failing the further tests. |
| "Failing the Ishihara means it's over" | Failure triggers further testing. It concludes nothing. |
| "You must pass all 24 plates" | 24 is the test edition. The first 15 are shown in random order. |
| "You're allowed 2 or 3 errors" | False for class 1 and 2: the criterion is "no error". The "9 out of 15" is the LAPL, night rating threshold. |
| "You must be a normal trichromat for class 1" | Incomplete: the text says normal trichromat OR colour safe. Being colour safe is enough. |
| "MED.B.070 governs this" | No: MED.B.075. MED.B.070 = visual system. |
| "The thresholds are in the regulation" | The regulation contains no figures. All the figures are in the AMCs. |
| "The lantern is not accepted in Europe" | EU AMCs accept it. It is the UK CAA that refuses it. |
| "The test is redone at every examination" | Systematic at the initial examination; then on clinical indication. |
| "VDL is the colour blindness limitation" | Acronym trap: VDL = distance vision. The day-flying limitation is VCL. |
What you can do, practically
- Ask the question before committing financially. If you know you have a deficiency, or if you suspect it, the time to talk to an AME is before paying for training, not after.
- Gather what you already have. An ophthalmological assessment, a colour test taken elsewhere: it does not bind the decision, but it saves time.
- Do not try to learn the plates. The presentation order is random: memorising a sequence makes no sense. And a candidate who has memorised a test has resolved nothing, neither for himself nor for those who fly with him.
- Tell the two questions apart. "Will I pass the test?" and "can I fly?" are not the same question: between them lie the further tests, and sometimes a limitation.
Frequently Asked Questions
Can you be an airline pilot if you are color blind?
Which article governs pilot color vision?
How many Ishihara plates must be passed?
Are you allowed a few mistakes on the Ishihara test?
What happens if I fail the Ishihara?
Does color blindness make you unfit for class 2?
Is the color test repeated at every medical examination?
Do France, Belgium, Switzerland or Luxembourg have their own thresholds?
Does what the UK CAA says apply in Europe?
Going further
This page covers a specific part of the medical. For the big picture: the overall page on the medical examination (classes, validities, age limits), glasses and visual correction codes, preparing for your medical, and if an issue arises, unfitness and appeal. The general hub: pilot health.
Le Labo's pages remain open access, including this one. Alongside it, there is now a paid guide: the retro-planning before the medical, the specific questions to ask the AME, the country-by-country appendices. Nothing that is free today will go behind a paywall.
Failure on the first test does not tell you what to do next
You now know that it triggers second-line testing and not unfitness. There remains the part that no one publishes: how this file is put together, who processes it, and how long it takes.
The guide covers the eight stations of the examination on the same model, with the referral circuit and the appendix for your country.
See the guide, €59