I was looking for the working of a two way mirror when I came across this video.
The person says that if you place your finger on the mirror and there's no gap between the image and the finger, then it should be a two-way mirror.
Does this really work? Shouldn't the image always appear to touch one's fingertips because of the principles of a plane mirror?


3 Answers 3


A standard mirror consists of a glass pane with a reflective backing. The reflection that you see comes from the reflective backing, not the glass. The principle of this test is that if you see a gap between your fingertip and its reflection then this gap is due to the thickness of the glass, and you have a standard mirror. However, if you do not see a gap then the reflection may be coming from the front face of the glass itself and you may have a two-way mirror. Or you could have a standard mirror with a very thin piece of glass.

Checking mirrors in my house, I can see a gap in a large wall-mounted mirror, but no gap with two smaller, lighter mirrors. So clearly this test is not conclusive evidence of a two-way mirror. I would have thought that turning the room lights off (to remove any reflection) and looking carefully for anything through the "mirror" would be a more reliable test.

  • 5
    $\begingroup$ Could a two-way mirror also consist of two panes of glass with different density, so the light reflects off the inner boundary? $\endgroup$ Feb 28 at 0:44
  • 4
    $\begingroup$ @PaŭloEbermann You can cause total internal reflection between two pieces of glass. It's how signals are propagated along a fiber optic cable. However, there's a minimum incidence angle required for total internal reflection, the critical angle. If you looked straight at that pair of glass panes, you would see right through it. I see this effect when I bring my waterproof watch into the pool. When looking at it at an angle, I see a perfect mirror reflection. But if I look at it straight on, I can read the time. $\endgroup$
    – Cort Ammon
    Feb 28 at 12:48
  • 9
    $\begingroup$ Killing the lights only works as a test if there is at least some light on the other side of the two-way mirror. You'd need to provide a light source that doesn't ruin your vision, such as pressing a flashlight against the suspect mirror so that most of the light goes through. $\endgroup$
    – user288750
    Feb 28 at 14:47
  • 12
    $\begingroup$ A mirror with the reflective coating, or "silvering" on the front is called a first-surface mirror, whereas a mirror with the reflective surface behind glass is a second-surface mirror. First-surface mirrors are utilized in wavelengths where a glass substrate would cause significant attenuation, in the infrared and UV, and where ghosting from the air-glass interface is an issue. The main disadvantage is the exposed metallization is subject to mechanical damage and oxidation. $\endgroup$
    – user71659
    Feb 28 at 22:02
  • 4
    $\begingroup$ Re, "...the reflection may be coming from the front face of the glass..." Not "may," "is." If it's glass, and if there is no gap between the reflection of your finger and your actual finger, then your finger is touching the metalized surface. But that doesn't help you to know whether there is a dark room on the other side of it from which a concealed person could look out, undetected. A so-called "one-way mirror" could be installed with either surface facing out, and it would work equally well either way. $\endgroup$ Feb 29 at 20:02

The fingernail test simply measures the distance between the surface of the mirror and its reflective layer. It tells you nothing about the reflective layer itself. Neither does it tell you whether the reflective layer allows a portion of the light to pass through, nor does it tell you whether or not there is an opaque layer behind the reflective layer.

In short, it's like assessing the potability of a liquid by observing whether it's transparent. There are plenty of poisons that are transparent, and plenty of good drinks that are not. It's just the wrong test.


The metal layer on any mirror is delicate. It needs protection from scratches and from atmospheric oxygen.

The cheapest mirrors protect it on one side by the glass itself, and on the other side by a layer of paint. Obviously, you can't see through the paint, so that type of mirror cannot be used for "one-way" surveilance applications.

A slightly more expensive type of mirror uses an extremely thin, transparent protective coating over top of the metal layer, and if the metal is thin enough (it almost always is thin enough) then that type can be used as "one way" glass.

The fingernail test will tell you which side of the glass the metal layer is on: If you see a gap, then you're touching glass, and the metal is on the "back" side. If you don't see a gap, then the metal, and an invisibly thin, transparent coating are on on the "front" side.

But the fingernail test does not tell you whether there is a dark space behind the mirror where a person or a camera could be hidden to watch you. Some mirrors (often called, "first surface mirrors") have the metal on the front side even though they are not being used as a "one-way" window. And if you do see a gap under your fingernail, that's no guarantee of privacy because a "one-way" mirror works equally well with either side facing out.

  • $\begingroup$ As an example of a "false positive": a lot of decorative mirror-like shards, as shown for example by the art installation at hackaday.com/2022/02/16/… , are very thin metal-coated mylar (PET). Naively, the fingernail test would suggest that somebody was sitting behind it. $\endgroup$ Mar 1 at 8:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.