White Light for Reptiles
(Amphibians have a colour vision similar to humans, so this text is not valid for amphibians)
Human colour vision is based on the three basic colours red, green and blue. Reptile colour vision
however is based on four basic colours: red, green, blue and UVA.
It is often elaborated that the correct correlated colour temperature (CCT, Kelvin) and the correct
colour rendering index (CRI) are very important in reptile husbandry. Presumably this shows the big
desire to use a lamp that is most similar to sunlight for reptiles and enables unaltered best vision.
Unfortunately these two values are only valid for human colour vision and meaningless for reptiles.
Anyone who wants sunlike light colours should use metal halide lamps or fluorescent lamps with UVA
phosphor and keep away from LEDs, no matter what CRI and CCT are.
One might ask, whether this is needed at all: I do not know of any scientific study that investigated
whether reptiles are harmed in any way if kept under non-white light. But this is no proof that they
are not affected. Colour seems to play an important role in the live of reptiles. Some have a strikingly
colourful and gender-specific coloration. Some lizards change their colouration within a few minutes
for communication or during the year to indicate that they are ready to mate. Flowers and fruits
eaten by reptiles often have a striking UVA colour pattern that could make them attractive to
reptiles. In the past 20 years some reptile keepers have reported that their animals thrive when
replacing standard fluorescent lamps with Narva Biovital lamps. Perhaps this is due to the white
colour of these lamps.
Method
The next figures each show:
(1) The spectrum of a lamp, measured with a high resolution spectrometer. The spectrum is
shown with the typical rainbow colours in the background.
(2) As reference, the spectrum of sunlight, shown as a green line.
These high-resolution spectra show all physical details, e.g. which molecules are present in the
earth's atmosphere and filter the light or which molecules are present in a discharge lamp. However,
this high resolution is often misleading for colour vision. You need a spectrometer with many
hundreds or thousands of camera pixels to make the fine lines visible in the spectrum. The reptile
eye does not have thousands but only four receptors for four different wavelengths. All the details of
the fine lines are invisible to the eye. What is important is how the energy is distributed over a much
coarser resolution.
Therefore, the figures also show:
(3) The effective irradiance of the lamp’s light towards the four cones of a reptile. These are
displayed as coloured bars.
(4) The effective irradiance of sunlight towards the four cones of a reptile. These are displayed
as thick dark green markers.
If the four coloured bars are close to the four green markers, then the lamp has a similar colour than
sunlight.
These graphs are also found in the reptile lamp database (www.lamps.licht-im-terrarium.de) for
many lamps and also for a reptile with 3 cones and the human eye.
Sunlight
The left-hand graph shows my reference spectrum, so the dark-green markers lie exactly at the top
of the coloured bars. The right hand graph shows a different solar spectrum with a bit less red in the
spectrum. The UVA cone and the blue cone get more light and the red (or better yellow) cone gets
less light. These variations are still ok and within the range of natural sunlight.
LED
LEDs offer a lot of control over the composition of the spectrum, so there is a big variety in the
spectrums of LEDs. But all white LEDs have in common that they do not emit infrared or UV. The
spectrum usually starts around 420 nm (blue) and ends at around 700 nm (orange/red). For the
human eye this light is white and the colour temperature can range from 2700 K to 6500 K with CRIs
well above 90 when the spectrum has no gaps.
From a reptile’s perspective, the picture is different! The effective irradiances for the blue, green and
red cones are like hat for sunlight. But the UVA cone does not see light because LEDs do not emit
UVA.
If the LED also had UVA it would be white for reptiles. But it ONLY lacks UVA, so it has the
complementary colour to UVA. Complementary colours to primary colours usually produce a very
strong colour impression. Example for human vision: Light is white when the red, green and blue
cone get almost the same amount of light. If the green cone does not see light, the colour is the
complementary colour to green: pink. This is the case for plant-LEDs that only emit red and blue but
not green. These lamps have a – in my view – very ugly and strong pink colour. So LEDs will not seem
something like “slightly different colour temperature” or “slight colour cast” towards reptile but
strongly coloured.
If you want to use LEDs, even if in addition to other, white, lamps, you should always ask yourself if
you would do that if the LED was pink. It is so misleading that these lamps look wonderfully white to
us humans.
Fluorescent Lamps
“Office tubes”
The typical office fluorescent lamps with colour codes 830 – 860 and 930 – 930 have no UV
phosphor. However, the mercury produces radiation at 365 nm and 405 nm. This ensures that the
UVA cones sees at least some light. The lamps are probably at least whitish for reptiles.
Narva BioVital and other fluorescent lamps with UVA phosphor
Better than office tubes are special fluorescent lamps with an UVA phosphor. For reptiles their colour
is very close to sunlight. In Germany lamps are available from https://www.natur-
nah.de/shop/beleuchtung/leuchtstofflampen/, also good seem to be https://www.true-light.eu/
UVB fluorescent lamps
UVB fluorescent lamps can also have a very balanced spectrum. Depending on the wavelength up to
which a reptile can see into the UV, it might be that the UV cone sees too much light and the colour
of the lamp has a UV colour cast.
Metal Halides
Non-UVB metal halides
These lamps developed for retail lighting have a very sunlike spectrum. The Osram Powerball HCI 943
in the left is very sunlike. The Osram HQI on the right shows that the quartz burner has a slightly less
balanced spectrum. From human vision perspective and lifetime, ceramic burners are superior. But
from reptile colour vision, I do not see large differences.
The Iwasaki Eye Colour has a particularly sun-like spectrum with an impressive colour rendering index
of 96. From a reptile’s perspective, I do not see a clear difference to the other metal halides.
UVB metal halides
UVB metal halides (Lucky Reptile Bright Sun and others) do not emit the same sunlike spectrum. The
spectrum consists of more individual lines, which is not necessarily negative for colour vision,
because the eye can not see the individual lines. Their disadvantage is the too strong UVA emission
which gives the light a strong UVA colour cast.
Dr. Sarina Wunderlich
www.licht-im-terrarium.de
