I cut and pasted (sorry about the formatting) because you have to join (free) to read it online, but here's the site link:
http://www.podarcis.nl/cgi-bin/toc.cgi?lang=eng
It's titled "UV-light emitting light sources, for use in vivaria" in the 2000-1 issue. There's pics and graphs in that version.
Pod@rcis 1(1) uk 2000 18
§
UV-light emitting light sources,
for use in vivaria
INTRODUCTION
The importance of ultraviolet (UV) light for reptiles, and to a lesser extent for amphibians, has
been frequently described (BOONMAN, 1987; IN DEN BOSCH, 1994). Discussions on UV-light are
usually on vitamin D production and the related calcium metabolism (MANTEL, 1994). Other
functions attributed to UV-light can be summarised as those influencing the biorhythm of
animals or their general welfare (FLEISCHMAN, et al., 1993). In this article the focus is
principally on the role of UV-light in vitamin D production. It is generally accepted and very
well documented that only the narrow range of UV-light from 290 to 300 nm is of biological
importance. This is part of the UV-B-light, also called D-UV, based on the role in vitamin D
production attributed to it (BARNARD, 1995). Past work primarily dealt with the theory of the
light effects. Occasionally a light source was recommended but concrete experiments with
particular lamps are scarce and usually limited to general considerations like: “My animals do
very well on it” (GORSEMAN, 1998; VANWESTBROEK, 1994). Although these contributions have
a certain value, a more systematic examination of the various possibilities, preferably based on
more or less solid experiments, would have been welcome. This article is an attempt in that
direction. Systematic reviewing of lamps with potentially beneficial effects turned out to be a
relatively simple task while collecting solid experimental data, including a translation into
Dutch vivarium practice, was substantially more complicated.
LIGHT SOURCES IN GENERAL
INCANDESCENT LIGHT PRODUCING BULBS
Transporting an electrical current through a filament produces incandescent light. This filament is usually made of
tungsten and resistance to it produces heat in the wire. The heated wire will start to glow red and upon progressive
heating will turn white. From observations that light produced by a lamp will become whiter as the temperature rises,
it can be concluded that a colour shift occurs from red towards violet, proceeding beyond the visible light spectrum
into the ultraviolet region. If normal bulbs ever became hot enough to produce UV-light, the light would be absorbed
by the glass cover and lose its value for our applications. Therefore, this type of lamp is of no value as a source of UVlight,
though they are still used as a source of heat and visible light. Halogen lamps, in contrast, have been shown to
be high performance incandescent lamps (IN DEN BOSCH, 1994). Because they get considerably warmer than ordinary
lamps, they produce a much whiter light with some UV component in it. Due to the heat, these bulbs are not made of
glass but rather UV-light transmitting quartz. Keep in mind that dimming halogen lamps results in a dramatic loss of
heat and consequently of UV-light.
Halogen lamps can be divided into various types. First is the group of low voltage (12 V) lamps that operate via a
transformer. This type shows the highest filament temperature and therefore produces the most useful UV-light. For
that reason these lamps are now usually provided with an UV-light absorbing coating or are fitted into a glass-covered
bowl. Fortunately, the application of an UV-block is usually indicated on the cover label. Keep in mind that ordinary
John Boonman
Tormentil 17
2631 DD Nootdorp
The Netherlands
jymbo@bart.nl
Small halogen lamp with
protective glass cover.
Contruction lamp without
protective glass.
Osram Ultravitalux lamp. ZooMed-lamp Reptisun
UV light source.
lamps, including halogen ones, are meant to operate in a human environment. As a result, although rather inconvenient
for our hobby, the production of UV-light is not a marketing asset whereas the presence of an UV-filter is. It is still
possible to find unfiltered lamps although growing effort must be applied in looking for them. Otherwise, a handy
men must be able to remove from the ready-to-use lamps, in a subtle way, the protective glass cover fitted in a
reflecting bowl.
A second type of halogen lamp, working on normal voltage, possesses a screw thread and a glass bulb and is therefore
of no use. The same drawback is found with the third group of lamps, although in this case it is easily circumvented.
Quartz tubes with porcelain end pieces that fit in ball shaped holders are supplied by rectangular metal reflectors (as
used in construction areas) and covered with a protective glass shield. This shield can be easily removed, leaving a
useful light source. The high power lamps are only useful in large enclosures. Constant attention must be paid to the
possible future application of UV-filters in this type of lamp, although it is unlikely because of the protective glass.
Whether the filament temperature is high enough to produce substantial amounts of UV-light is not known but seems
likely. Nevertheless, it seems wise to maintain a minimal distance between the light source and the animals to obtain
some benefit but do not forget the intense heat generated by the lamps.
ARC-DISCHARGE RADIATORS
In a discharge radiator, type TL (or in its compact, screw thread fitted form PL) light is being generated by means of
an electrical discharge in a tube filled with a mercury salt vapour. Excited mercury falls back into its ground state
producing light in two sharp lines of 185 and 253.7 nm, the so-called mercury lines. Light of these wavelengths would
be extremely dangerous if the glass tube (or even the atmosphere) did not absorb it. In the radiators under discussion,
the light is being used to activate a phosphor coating applied on the inside of the tube. The activated phosphor starts
fluorescing and emitting light. The choice of phosphor powder allows light of any desired wavelength to be created,
including the area of our wishes. For the lighting to be used in an in-house application, the marketing drive is valid
that no (harmful) UV-light is supposed to be emitted which will govern the choice of powder. These types of radiators
will therefore be of no use as a source of UV-light.
For use in special applications other than simple illumination, some interesting lamps can be found in this category.
They include radiators for sun tanning devices, for medical applications and recently phosphor-coated radiators
specially developed for use in vivaria. They are principally made of a type of glass that transmits light of the desired
wavelength (and filters out harmful light). All relevant radiators will be discussed below.
OTHER LAMPS
In addition to the TL-type lamps, other discharge radiators can potentially be used. In general, they are made of a
special type of glass that allows the produced UV-light to be transmitted. Because this type of lamp produces light as
a line spectrum with a few narrow wavelength areas, they are of limited use. The light produced seldom falls in the
optimal range and they are exclusively available in high powers that often require complex technical facilities. Osram
Ultra-Vitalux is the only representative of this group that is used frequently. The properties of this lamp have been
described in an earlier paper (BOONMAN, 1987).
WHICH LAMPS ARE SUITABLE IN THEORY?
We call a lamp suitable in theory when it produces light in the wavelength region between 290 and 300 nm. When it
falls above or below or whether it is useful in practice is another question that will probably be answered later.
In the category of light sources designed for normal illumination purposes only halogen lamps meet the criteria. The
spectrum of 12 V lamps shown in IN DEN BOSCH (1994) indicates that some light is produced in the desired spectral
area. Since minimal radiation is produced, the distance must be kept short. Those very small lamps might therefore be
more suitable for small housings. As was previously mentioned, no UV-block must be present. Whether the 220 V
lamps are still useful remains an open question. The filament temperature is lower and as a consequence they produce
less UV-light than their low voltage counterparts. In addition, because of their power (or better heat), they can only be
used in larger vivaria where the distances are greater and the yields correspondingly lower. Lamps used in construction
areas, with the protective glass shields removed, are of potential use.
The group of special arc-discharge radiators, not specifically designed for vivaria but of potential benefit is comprised
of several interesting types. The Philips 05 radiator was previously described (BOONMAN, 1987) and shows an emission
peak at around 365 nm, with some side radiation around 300 nm. Probably more suitable are the radiators of the Philips
09 family that are recommended elsewhere as well (MADSEN, 1998). They are designed for sun-tanning purposes and,
according to the manufacturer, produce 0.7 % UV-B (PHILIPS, 1995). The power range is 20 to 160 W and they are
available in PL format (36 W). Because commercial sun-tanning studios have an interest in a quick result for their
customers, a professional version can be obtained having 1.4 % UV-B, type 109 (80-160 W). It will probably be more
Pod@rcis 1(1) uk 2000 19
§
difficult to obtain the radiators designed for medical purposes. Type 01, with a very narrow emission peak at 312 nm
(100 W), is meant for the treatment of psoriasis. Radiator type 12 (20-100 W, equivalent to PL-S/2p, 9 W) shows a
relatively broad emission peak at around 306 nm and will therefore produce considerable amounts of light in the
desired area.
Numerous other radiators designed for more or less normal use are available on the market. These products are often
labelled with stimulating one-liners that are far from standardised and therefore of limited benefit. Nevertheless
BARNARD (1995) tried to attribute some generalities to these radiators. Radiators promoted as cool white hardly
produce any light of a wavelength lower than 315 nm. Warm white radiators generally do produce some UV-B light,
sometimes even in the area of 290-300 nm but their yields are very low. Daylight types are comparable with warm
white types, probably with somewhat higher yields. Radiators of the sunlight type usually produce intense light
including some radiation around 315 nm. All these more or less obscure light sources can be indicated as of limited
value unless a reliable emission spectrum is presented simultaneously.
The black light lamps, which cause white shirts to shine brightly in the disco, produce an large amount of UV-A light
(peaking around 370 nm) and some radiation in the valuable area. However, the appearance of the emitted light is
generally not acceptable and it will have to be compensated for with a lot of other light.
SPECIAL LAMPS
In addition to the Ultra-Vitalux lamps that are rather popular in the German linguistic areas (c.f. SCHOOP, 1989), only
phosphor-coated arc-discharge radiators have been designed to produce extra beneficial UV-light. BARNARD (1995)
and GEHRMANN (1997) have presented overviews of available lamps, which probably differ in accessibility between
countries. A detailed summary of the amounts of various radiations that are produced is given in table I. The term DUV
refers to the amount of light ranging from 290-300 nm that is involved in vitamin D production.
Table I: Radiation (µW/cm2) determined at 61 cm distance (after BARNARD (1995))
Type lamp UV-C D-UV UV-B UV-A Visible % D-UV
(of UV-B)
Duro-Test Vita Lite 0.0018 0.165 2.3712 27.37 51.3 7
ESU Reptile Daylight 0.0024 0.3162 3.756 24.85 51.63 8.4
ZooMed Reptisun 5.0 0.0011 0.1428 2.8287 75.43 42.01 5
General Electric Daylight 0.0011 0.1281 1.9518 11.05 48.74 6.6
General Electric Cool White 0.0001 0.0001 0.0863 9.3 49.45 0.1
General Electric Warm White 0.0003 0.1292 1.9822 6.17 45.69 6.5
General Electric Blacklight 0.0012 0.0354 0.6403 360.31 9.29 5.5
Sylvania 350 Blacklight 0.0074 0.417 6.4956 358.75 13.98 6.4
Sylvania Experimental Reptile 0.0013 1.0898 7.061 20.03 51.24 15.4
All the radiators mentioned, except the cool white type lamps that are mentioned as a comparison, are theoretically
useful light sources because they produce an amount of UV-light that we consider appropriate (D-UV). The
experimental Sylvania lamp is the most prominent type of radiator. It produces much of the desired light without the
interference of other light but in contrast to the blacklights, with a substantial amount of visible light. However, this
must be put into some perspective. Some values that have been reproduced from BARNARD (1995) show a significant
discrepancy from the other observations. For example, GEHRMANN (1997) puts the ZooMed Reptisun 5.0 at a
considerably higher level in absolute terms (UV-B production of 10 mW/cm2) in comparison to other light sources.
Discrepancies like these are inevitable, as observations are strongly dependent on the methods used for determination.
Comparing the emission spectra of several specially designed lamps with standard light sources, it is remarkable that
the ZooMed radiator produces a giant peak around 350 nm which is comparable to a blacklight. In practice the
ZooMed lamp exhibits some kind of a disco effect so that a combination with other light sources will be necessary at
any time. In the UV-B area, light emission starts below 300 nm and a small peak can be observed around 310 nm.
Therefore that radiator meets the criteria for usefulness in theory. The spectrum of the experimental Sylvania reptile
lamp is more or less comparable with emissions starting as low as 290 nm and peaking at 310 nm. As a giant peak in
the UV-A region is absent a more pleasant light performance can be expected. Recently Sylvania marketed a radiator
called Reptistar. The spectrum depicted in the product flyer shows some agreements with the experimental lamp so it
can be speculated that the Reptistar line will be the commercialised version of the experiment. The radiators are
available in 15, 18 and 30 W versions corresponding to lengths of 45, 60 and 90 cm, respectively.
Pod@rcis 1(1) uk 2000 20
§
LAMPS IN PRACTICE
Literature data on the benefit of UV-lamps in practice are scarce and limited to generalities like: “My animals behave
well” or “Since I used that lamp more hatchlings are produced”. Only Joni Barnard’s thesis (1995) gives solid
observational data. The first question answered was: do the UV-lamps we are focusing on convert pro-vitamin D into
vitamin D? In a quartz reaction tube a solution of pro-vitamin D was radiated a number of times with several lamps
from varying distances. The amount of pro-vitamin D that was converted was determined and compared to a natural
situation. Table II clearly shows that the experimental Sylvania lamp and the ZooMed radiator realised some
conversion, the Sylvania lamp even at considerable distances. Unfortunately more data on the ZooMed lamp are
missing. The conclusion that can be reached is that some lamps at least do something but whether it is enough remains
an unanswered question.
The ultimate experiment for answering the question of whether radiation with UV-lamps would have some effect
should look like the following. The animal under observation will be exposed to the light source of choice, in a
situation resembling life as closely as possible (so not at a distance of 5 cm for 24 hours). Subsequently, the blood is
analysed for compounds that indicate a conversion of pro-vitamin D into vitamin D or one of the relevant derivatives.
Some clues are given in the aforementioned thesis.
After treatment with the experimental Sylvania radiator, green iguanas lacking vitamin D showed elevated levels in
the blood of one of the vitamin D derivatives (25-hydroxy vitamin D [25-OH-D]). Concentrations of 200-1200 ng/ml
were measured as compared to vitamin D injections (without radiation) where the concentrations only reached 42-230
ng/ml. According to generally accepted criteria the condition of the bones improved considerably. In a second
experiment (and, in my opinion, a quite convincing one), young green iguanas with a demonstrated vitamin D
deficiency (concentration 25-OH-D in blood < 4.8 ng/ml) were fed vitamin D at enormous levels as high as 8.5 IU/g
body weight. After seven days the 25-OH-D concentrations reached a level of about 180 ng/ml. After the same period
of time, the group treated with the Sylvania experimental lamp reached a 25-OH-D concentration of more than 420
ng/ml. Once again the condition of the bone system improved and no side effects could be observed.
CONCLUSIONS
From the experiments described in the thesis, it can be concluded that the experimental Sylvania radiator works
both in a test tube with the conversion of provitamin D into vitamin D and in vivo in green iguanas. I see no reason
why the commercial version of the lamp would not have the same effects as its experimental counterpart nor why
radiators from the ZooMed-line would not be able to do the same. Comparable other light sources may also be
useful, although the described theoretical considerations indicate that only a few are serious candidates. Compared
to the proven suitable light sources, the majority of the other ones produce marginal amounts of suitable UV-light
so that positive effects are doubtful.
As a side conclusion, the frequently expressed impression that green iguanas absorb vitamin D poorly from their
feed seems to be confirmed. In addition, it is clear that what holds for a green iguana does not necessarily hold for
other animals. The frequently mentioned thesis describes experiments with the tortoises Gopherus agassizii and
Geochelone sulcata. Under none of the experimental conditions was any 25-OH-D demonstrated in the blood of
these animals or in the control group. However the bone system of these animals improved after treatment with the
experimental radiator. It is, therefore, extremely difficult to draw conclusions. Tortoises may have a different mode
of vitamin D metabolism and consequently the wrong metabolites may have been examined. As well, it is likely
that ground-dwelling species that sometimes prefer shady forests are adapted to UV-light in a different way than
basking tree-dwelling animals.
EPILOGUE
If the individual habitat requirements of animals and the associated adaptations are not considered, it is interesting to
develop a sense of the duration and intensity of radiation required with lamps to produce the same amount of UV-B
light as natural sunlight in tropical regions. BALL (1996) made that comparison, albeit with rather ordinary lamps, but
still with interesting conclusions. Using the amount of UV-B light that one obtains at the equator during five minutes
in the sun at 12 noon as a reference, a session of 5.2 minutes would be necessary in Florida at the same time of day
on June 21 to obtain the same dose. With sunny weather, on July 13 in Nürnberg, Germany, it would be 7.8 minutes
and on July 14 in Longyearbyen, Norway (78.2° NB) it would be 24 minutes. On a cloudy day in March in Texas, we
would have to stay 1.47 hours outdoors to get our dose. And what about the lamps? A Duro-Test Vita-Lite lamp of 20
W at a distance of 15 cm produces the reference amount of UV-B light only after 7.1 hours and at a distance of 76 cm
we have to be patient for 110 hours. A Westinghouse FS20 Sunlamp 20 performs better: 8.3 minutes at 15 cm and 1.6
hours at 76 cm.
Pod@rcis 1(1) uk 2000 21
§
Table II: Formation of vitamin D from pro-vitamin D under influence of a UV-lamp
Lamp Radiation time (min) Distance (cm) Conversion (%)
Control (wrapped tube) 60 61 0
Control (wrapped tube) 120 213 0
Type Daylight 15 61 0
Type Daylight 30 61 0.03
Type Daylight 60 61 0.06
Type Daylight 15 213 0
Type Daylight 120 213 0
Experimental Sylvania 15 61 0.13
Experimental Sylvania 60 61 0.58
Experimental Sylvania 15 213 0.01
Experimental Sylvania 120 213 0.14
ZooMed Reptisun 5.0 60 30 0.5
Sun light 60 Summer day, 6
noon, N42°
Spectral power distribution of
Philips phosphor coated UV radiators.
(Lamptype: upper righthand
corner.)
Pod@rcis 1(1) uk 2000 22
§
These observations are of limited value, first of all because specifications of the mentioned light sources are missing
and because no known lamps have been tested. In addition, the individual vitamin D requirement of various species
are not known, neither is the rate or efficiency with which it is formed nor the duration of time that it lasts. It is feasible
that lipid-soluble vitamin D can be stored for a time in the fatty tissues and mobilised when it is needed. It is therefore
possible that an animal only requires a relatively short radiation time to maintain its vitamin D supplies at the necessary
level, a time that will differ among species.
The light sources that were tested in the thesis all lost about one quarter of the intensity of produced UV-B light within
the first week of use. During the next four months the level stayed constant.
SUMMARY: ULTRA-VIOLET LAMPS
An overview of lamps available on the Dutch market that emit light in the part of the spectrum that is generally seen
as necessary for the production of vitamin D (290-300 nm), is given in this article. The lamps “Zoo-Med Reptisun”
and Sylvania´s “Reptistar” were developed especially for terraria and are suitable in smaller enclosures where the
animals can approach the lamps closely. Both lamps have been shown, in a test tube as well as in Iguana iguana, to
facilitate change of provitamin D into vitamin D. These lamps must be used in combination with other lamps to provide
adequate warmth radiation and light in the visible spectrum. If animals can approach the lamps quite closely, the
Philips “09” lamps - developed for use in solaria - may work. Some special lamps are suited to specific situations. It
is apparently also possible to use 12V halogen lamps in small enclosures, as long as an UV-block in the form of a
protective glass or a coating has not been applied. Advantages of the halogen lamps are that they provide enough
warmth and visible light, thus it is not necessary to combine usage with that of other lamps. There is no direct evidence
that they facilitate production of vitamin D however. The same is true for the halogen lamps suitable to larger
enclosures. In such cases the Osram “Ultra-Vitalux” is considered the most suitable UV-B source, although there is no
hard evidence of its effectiveness either.
ACKNOWLEGDMENTS
The contributions to this article from Herman in den Bosch, Paul Gorseman, Gerrit Hofstra and Gerard Stoer are
kindly acknowledged.
Pod@rcis 1(1) uk 2000 23
§
Spectral irradiance of the Sylvania Blacklight Blue, from
250 to 700 nm, expressed in µW/cm2.
Spectral irradiance of the Sylvania 100% 2096 Phosphor,
from 250 to 700 nm, expressed in µW/cm2.
Spectral irradiance of the Sylvania 35% 2096 Phosphor,
from 250 to 700 nm, expressed in µW/cm2.
Wavelength (nm) Wavelength (nm)
Wavelength (nm) Wavelength (nm)
Spectral irradiance of the Sylvania 350 Blacklight, from
250 to 700 nm, expressed in µW/cm2.
Irradiance (µW/cm2
Irradiance (µW/cm2
Irradiance (µW/cm2
Irradiance (µW/cm2
LITERATURE
BALL, J.C., 1996. A comparison of the UV-B irradiance of low-intensity, full-spectrum lamps with
natural sunlight. [On-line] Source: http://www.sonic.net/melissk/j_ball.html
BARNARD, J.B., 1995. Spectral irradiance of fluorescent lamps and their efficacy for promoting vitamin
D synthesis in herbivorous reptiles. Thesis Michigan State University.
BOONMAN, J.C.P., 1987. De zin en onzin van UV-licht. Lacerta 46: 22-27.
BOSCH, H.A.J. IN DEN, 1994. Ultraviolet en halogeenlampjes. Lacerta 52: 136-143.
FLEISHMAN, L.J., E.R. LOEW Ultraviolet vision in lizards. Nature 365: 397.
& M. LEAL, 1993.
GEHRMANN, W.H., 1997. Reptile Lighting: A current perspective. Vivarium 8: 44-45, 62.
GORSEMAN, P.D., 1998. Tiliqua gigas, de Nieuw-Guinea-blauwtongskink. Lacerta 57: 54-63.
MADSEN, M., 1998. Noget om Lys. Nordisk Herpetologisk Forening 5: 100-101.
MANTEL, P., 1994. De werking van vitamine D. Lacerta 52: 131-135.
PHILIPS LIGHTING B.V., 1995. UV Radiators and applications. Correspondence course lighting 12. Philips,
Eindhoven.
SCHOOP, E., 1989. Ultraviolett-Bestrahlung bei Reptilien. DATZ 42: 470-471.
WESTBROEK, R. VAN, 1994. Halogeen-verlichting: het ei van Columbus? Lacerta 52: 144-146.
Pod@rcis 1(1) uk 2000 24
§
