Ultraviolet (UV) radiation has been a source of controversy among reef hobbyists.   The effects of natural solar radiation have been well studied and are a matter of record. However, few studies of UV produced by artificial lighting are available to aquarists. (Our initial studies of lamps commonly used in reef aquaria applications and their ultraviolet radiation output were published in Aquarium Frontiers in 1996 and were meant to complement the previous work of ( Bingman, 1995.)   This article will present the results of our on-going testing in this important aspect of reefkeeping and supplements works by Bingman as well as Joshi and Morgan (1998).

Before continuing, perhaps a definition of ultraviolet radiation is in order. Ultraviolet energy is classified into three categories;

UV-A is radiation just below the violet portion of the visible spectrum and consists of those wavelengths between 320 and 400 nanometers (nm).  UV-A is invisible to the human eye and is the least destructive.

UV-B wavelengths (280 – 320 nm) are biologically destructive and, through overexposure, cause erythrema (sunburn).   

The third UV category consists of wavelengths between 200 and 280 nm and is referred to as UV-C.   The Earth’s atmosphere absorbs UV-C produced by the sun and therefore is not found naturally.   However, UV-C can be produced by artificial means. Arc welding produces UV-C; so do certain specialty electrical lamps (such as those employed by hobbyists for the sterilization of aquarium water).   

Our discussion will focus on UV-A and UV-B.   (We should note again that UV radiation is not visible to the human eye.  It is not, as many hobbyists believe, related to the “blueness” of the light.)  Fluorescence is the phenomena in which UV light is absorbed by certain pigments and emitted in the visible portion of the spectrum.)  We measured the output of popular aquarium lamps using an instrument called a radiometer. Manufactured by UVP, Inc. (Upland, CA) this radiometer uses separate probes – one to measure UV-A (calibration point of 365 nm); the other senses UV-B (calibration point of 310 nm). This instrument reports UV energy in units of microwatts per square centimeter per second - µW·cm²·sec, or, for brevity, µW.   The readings from this instrument are NIST-traceable. For comparisons, this instrument reports maximum temperate latitude UV-A as about 2,000-2,100 µW; UV-B is about 1,000 µW. Measurements taken in Hawai’i (Makena Landing, Maui) at noon on a sunny November day showed UV-A as 1,260 µW and UV-B as 960 µW.

Why should aquarists concern themselves with ultraviolet radiation?   Don’t most metal halide lamps use a glass envelope to “shield” the UV-producing inner arc tube, thereby reducing UV to “safe” levels?   And, after all, fluorescent lamps don’t produce any UV, correct? Additionally, isn’t UV the reason that corals are colorful at all?   We investigated these issues and report our findings here, much of it for the first time.

Ultraviolet Output of VHO Fluorescent Lamps
It isn’t uncommon to see lamp advertisements touting that fluorescent lamps don’t make any ultraviolet energy. Not so. The amount generated may be low, but it is there. This UV energy is produced by an electrical arc between the electrodes located at either end of the lamp tube. Phosphors within the arc tube absorb the invisible UV and fluoresce –“emit” - most of it in the visible spectrum. However, some UV is transmitted through the glass tube.

Figure One shows the UV-A distribution of 4-110 watt URI lamps (two “Actinic-Day”; two “Super Actinic”).   These lamps were suspended 4.5” above the UV sensors. The internal reflective surface was painted white and there was no lens or “splash guard”. As one can see, the amount of UV energy generated by these lamps is relatively low.

Ultraviolet Output of Compact Fluorescent Lamps
Compact Fluorescent lamps are relatively new to the aquarium trade. Advertisements routinely claim that these lamps produce more “light” than comparable VHO lamps, (which our research confirms). They also produce more ultraviolet radiation. Figure Two shows the UV-A energy distribution pattern of 4-55 watt lamps marketed by Hamilton Technologies.

 Figure Three shows the UV-A energy distribution pattern of 4-96 watt lamps distributed by CustomSeaLife.

We should warn against making direct comparisons of Figures 1-3. These measurements were not made under standardized conditions (in these cases, reflector material and lamp height above sensor).   We wish to demonstrate only that fluorescent lamps generate UV, albeit in relatively small amounts.

Ultraviolet Output of Metal Halide Lamps
Metal halide lamps have many myths surrounding them. They are variously noted to produce “high” amounts of UV energy, or “none” due to absorption by protective glass envelopes. We address this issue and also demonstrate that seemingly small variables can make profound differences in the amount of UV energy actually reaching a coral within an aquarium (generally, these variables also make a difference in the amount of photosynthetically active radiation – PAR – making it into an aquarium).

Evaluation of Shielding by Metal Halide Glass Envelopes
It is a common belief among hobbyists that the outer borosilicate glass “globe” absorbs most or all UV energy generated by metal halide lamps (and others). This seems reasonable since most lamps used in the hobby carry an “R” warning (see 'Markings' below). We wondered if the glass used in American, Japanese or German metal halide lamps possessed different UV transmission qualities.  To this end, we developed a simple experiment to test metal halide glass envelopes. A mercury vapor “sun lamp” was used as the UV source.  We then placed shards of glass from shattered outer envelopes of Coralife and Radium lamps over our UV-A and UV-B sensors.  An unbroken Iwasaki lamp was also tested and, obviously, since this envelope was intact, these readings represent the UV energy traveling through two layers of glass.  Table One reports the somewhat surprising results of our testing.

As we can see, the glass envelopes of these lamps transmit most UV energy. So, the next logical question asks us to investigate the amount of UV energy produced by commonly used metal halide lamps.

Evaluating Metal Halide Lamps
In order to make meaningful comparisons of different lamps; we must standardize testing parameters.  For our tests, we obtained new lamps and “burned” them continuously for 100 hours using the same ballast. We burned the lamps continuously for 8 hours after changing orientation (vertical to horizontal or vice versa). We tested all lamps using the same ballast of appropriate wattage, lamp (arc tube) height above the sensor was 9.5 inches, no reflectors or shields were used, all testing was conducted in a darkened room, etc.

We found metal halide lamps to produce varying degrees of UV radiation. However, these measurements should not be considered absolute. Quality control, at its best, allows a wide degree of variance from lamp to lamp, even in the same production run. We checked three new (100 hour burn-in period), “identical” Aqualine Buschke 175 watt 10,000K metal halide lamps for maximum UV-A and UV-B output. Testing was conducted under standardized conditions (height, lamp orientation, etc).  Table Two reports this information.

Does UV distribution vary from lamp to lamp? To determine this, we tested four new (100 hr. burn-in) Coralife 400 watt 10,000K lamps for UV distribution. We measured UV radiation across a grid system (24”X48”) for a total of 128 readings each. Distribution patterns were similar between the four lamps and Figure Four is typical of these patterns. (UV output was also fairly consistent, with a difference of 7.8% between the lamps producing the most and least radiation. Compare this to the single maximum readings in Table Two).

Influences of Lamp Burn Position
Generally, fluorescent lamps are mounted horizontally (the electrodes are on a horizontal plane) and the operation position (called the burn position) is not much of an issue.   The same is true for power compacts (although it is possible to mount these lamps vertically).  With metal halides, burn position is critical for proper lamp performance (PAR output, color temperature) and lamp life. Metal halide lamps can be mounted vertically, horizontally or at an angle.   

Markings Found on Metal Halide Lamps
Most metal halide lamps have markings somewhere on the glass envelope.   It is important to know how to read these. They provide information for proper operation and lamp life.   Disregard markings on the metal base; these generally denote quality control information, such as date of manufacture.

Example: Coralife 10000K MH175/U/ED18 USA R

Coralife – Maker or Distributor

10000K – Kelvin rating (K): 10000, a very blue spectrum, higher numbers (14000, 20000, etc.) indicate that the light produced by the lamp is “bluer” or “cooler”.   A lesser number (5000, 6000, etc.) indicated that the light is more yellow or “warmer”.   Sunlight is generally rated at 6500K.

MH - Metal halide

175 - Lamp Wattage

U - Universal burn position.

ED -  Lamp Shape (Elliptical Dimpled) and Maximum Diameter (in this case, lamp diameter is 2.25 inches). 

18 - 18-8ths, which equals 2.25.

USA – Country of Origin.

R - Indicates lamp will not extinguish if the outer envelope is broken.   Interestingly, these lamps carry a warning about potential UV exposure should the outer envelope break, even though this envelope does little to attenuate UV when it’s intact!   A “T” indicates that the lamp will extinguish if the outer envelope fails.

Some lamps will not have all the above information; others will have different codes.   Here are some examples:

Common Markings Designating Lamp as a Metal Halide:

MH, MT, “Halogen-Metalldampfl”.

Common Kelvin Ratings:
4,500, 5,500, 6,000, 6,500, 10,000, 12,000K, 14,000, 20,000 K.

DX = Daylight Deluxe (about 6,000 - 6,500K)
DL= Daylight (about 6,000 – 6,500K)
BDX = Blue Deluxe

Lamp Shapes:
T = Tubular.   
T7=23 mm diameter; 
T8= 25 mm diameter.   
T lamps have no outer envelope; they are a naked arc tube.

BD17/BD54 = Elliptical Dimpled, 2.125 inch diameter.
E17N/E54N = Narrow Neck, 2.125 inch diameter.
E28 or ED28/E90= Elliptical Dimpled, 3.5 inch diameter.
ED18= Elliptical Dimpled, 2.25 inch diameter.
ED37/E120 = Elliptical Dimpled, 4.625 inch diameter
BT28= Blown Tubular, 3.5 inch diameter
BT37/BT120= Blown Tubular, 4.625 inch diameter
BT56/BT17= Blown Tubular, 7 inch diameter (usually 1,000 watt lamps)

Common Wattages:
50, 75, 100, 150, 175, 250, 400, 1000

Burn Positions (failure to burn a lamp in the proper position can reduce lamp life and affect “light” and UV output, or the lamp may not start at all):

U: Universal, lamp base vertical (up, down), horizontal, angled – (e.g., no specific burn position necessary).

BU: Base Up
VBD, VBU, BU/BD or BUD: Base Up or Down (within 15° of vertical)
BD: Base Down
V: Vertical
HBU: Horizontal to Base Up
HBD: Horizontal to Base Down
HOR or BH: Horizontal

Influences of Luminaire Shapes and Reflective Surfaces
Reflector materials commonly include painted surfaces, cast aluminum, polished aluminum and mirrored surfaces.   Each of these reflects UV radiation differently. Table Three lists the relative reflective qualities of three materials.

*Results based on percentage increase of UV.
**Polished Aluminum Surface painted with Flat White Enamel. 
***”Best” does not necessarily mean “good!” 

The shape of the reflective surface can also profoundly affect the amount UV radiation falling upon a given point.   Generally, pendant (vertical) mounts focus PAR and UV more effectively than horizontal mounts.   Some pendants are much more effective than others are.  Figures Five and Six illustrate how differently shaped (and constructed) pendants can focus UV radiation.   These tests used the same lamp (Coralife 400w 10K), ballast (CWA), height (9.5”), etc.; the only difference was the luminaire.   

As we can see, the luminaire can make a dramatic difference - maximum UV-A output is almost 5,000 µW – 2.5 times the amount the amount of maximum temperate latitude UV. (If this luminaire were lowered to 4.5” above the water, there would be a smaller area of higher UV, something around 14,000 µW  - 7X the amount of temperate sunlight.)   We are basing this projection on data we gathered using jigs and our UV sensors.   

Lamp Warm Up and UV Production
We were curious about the production of UV energy during a metal halide’s warm up stage (called lamp strike).   We reasoned that invisible UV energy would be produced during the strike and gradually subside as the halides within the arc tube vaporized and began to emit visible light.   Our test results showed the maximum UV-A produced during lamp strike to be 17% higher than a fully warmed lamp; UV-B was 100% higher.   Figure Seven shows the results. 

Focusing of UV Radiation by Waves
Any visitor to a pool or shallow body of water has seen chain link-like patterns of light dance across the bottom.   These lines are called the “caustic network” (hobbyists often, and incorrectly call these “glitter lines”.   Glitter is a reflection of light such as one sees on the water surface at sunrise or sunset). The caustic network is caused by waves focusing and defocusing of light  (these waves act like concave and convex lenses). Lynch and Livingstone (1995) give a simple formula to determine the depth at which light is at its focal point – Multiply the wavelength (crest-to-crest or trough-to-trough) of the water wave by 5. Example: If an aquarium has surface waves with a wavelength of about 0.5 inches (which is typically for many aquaria), then we could expect the focal point to be about 2.5 inches below the surface. We checked the focusing of UV in an aquarium and found measurements of up to 5% higher at the predicted focal point than above the water’s surface. The same held true for PAR. We believe our instruments’ reports of peak readings are very conservative. Stramski and Legendre (1992) report the flashes are of only 5-20 milliseconds in duration and their sensor detected light pulses five times greater than surface readings. 

Discussion
Ultraviolet radiation was generated by all light sources that we investigated.   The actual amount produced depended upon many factors, including lamp type, lamp orientation and even small manufacturing differences between lamps.   When we consider the significant differences that the reflector shape and construction add to the mix, it is apparent that only generalities can be made as to what an aquarist can expect from his or her particular light system.   In some applications, we contend that there be reason for concern, especially when we consider that UV-A and UV-B can be significantly higher during lamp strike and then focused by waves on the water surface.

Biological Effects of Ultraviolet Radiation
It has long been recognized that UV radiation penetrates natural waters (Jerlov, 1950; 1976).  Lewis (1995) noted that reef organisms employ several mechanisms to protect themselves from UV: avoidance, protection and repair.  Obviously, many corals colonize areas with relatively high UV energy, so they have measures to protect themselves.  Shibata (1969) found UV-absorbing pigments in certain Staghorn corals (Acropora) and Cauliflower corals (Pocillopora) as well as a blue-green alga.  He speculated that a compound he called S-320 (meaning it absorbed UV energy at and around 320 nm) might protect marine organisms from ultraviolet energy penetrating shallow water. An even earlier researcher, (Kawaguti, 1944) speculated that some coral pigments offered protection against “strong sunlight.”   Jokiel (1980) was perhaps the first to report UV as an important ecological factor on shallow water tropical ecosystems. Together, these papers along with others suggested that UV radiation could have deleterious effects on coral reef communities. 

Perhaps the best single source of information on this subject was published in 1995.   Entitled “Ultraviolet Radiation and Coral Reefs;” this work contains many papers on the effects of UV on corals, zooxanthellae, plankton, etc. Clearly, these present a convincing argument that ultraviolet radiation can have a major impact on coral reefs in many different ways. 

What effects can UV have on coral reef organisms?   A review of the literature suggests that the effects can be numerous and far-reaching. 

Ultraviolet radiation can play a role in determining the survival of corals from their very beginnings.   Baker (1995) reported that UV-A inhibited larvae settlements of the Cauliflower coral (Pocillopora damicornis); Gulko (1995) demonstrated that UV-B damaged the gametes of the Plate coral (Fungia scutaria) - possibly explaining why corals often spawn in darkness. 

UV (or UV in synergy with high PAR, temperature, etc.) is known to inhibit photosynthesis.    Glynn et al found a combination of high temperature (31°C) and high UV (A & B, 25-30% of direct exposure) caused a high mortality rate among Acropora vallida specimens (significantly more Pocillopora specimens survived this treatment).  Lewis (1995) suggests that the zooxanthellae in the stony coral Montipora verrucosa may be damaged by exposure to ultraviolet radiation. Shick (1991) reported that the octocoral Clavularia (commonly called Star Polyps by hobbyists) exhibited a 50% decrease in photosynthesis when exposed to “high” levels of UV-A and UV-B.   Hohlbauch (1995) reported the same effect on the Plate coral (Fungia scutaria). Gleason and Wellington (1992) used an underwater spectroradiometer to determine that increased dosages of UV could induce bleaching (loss of zooxanthellae) in the stony coral Montastrea annularis.  UV was found to reduce the amounts of photosynthetic pigments and cause photoinhibiton (a reduction in the rate of photosynthesis) in the red alga Porphyra leucosticta  (Figueroa et al, 1997). Donkor and Hader (1997) found that UV-B radiation caused bleaching of photopigments in the cyanobacterium Phormidium. The list of references could go on and on.   It should be sufficient to say that ultraviolet radiation is potentially harmful to natural and captive ecosystems.

“Filtering” of Ultraviolet Radiation by Aquarium Water
Many aquarists believe that yellowing substances normally found in aquarium water will quickly absorb UV. We tested for UV transmission in aquarium water of several “colors” and report the results of two tests here.   We used submersible UV probes, a sliding depth gauge, a Hach Color test kit and a “sunlamp” for the UV source. Aquarium water was “colored” with skimmate from a foam fractionator (protein skimmer) filtered through Whatman #4 paper.   Figures 8 and 9 illustrate the water transmission properties. 

As can be seen, UV energy falls off rather rapidly.   However, under certain conditions (high wattage, reflector material with high UV reflectance, etc.), it is possible for UV-A radiation levels at the bottom of an aquarium to exceed UV found at the ocean’s surface. 

Corals’ Natural Sunscreens
Many shallow-water organisms contain natural sunscreens to protect them from UV radiation.   These include holothurians (sea cucumbers, Shick et al, 1992), algae (Shibata, 1969), corals (Shibata, 1969; Dunlap and Chalker, 1986; WuWon et al, 1995, 1997) and others (Dunlap, Chalker and Bandaranayake, 1988). These sunscreens are amino acids, more specifically, mycosporine amino acids (MAA’s). There are many MAA’s; each of these absorbs different UV wavelengths. For example, an MAA called S-320 absorbs UV at and about 320 nm; others have absorption maxima of 310 nm through 360 nm. These substances are colorless and block only UV – they allow visible radiation (PAR) to pass.

Factors Affecting MAA Concentrations in Coral Tissues
Increased UV levels (due to the thinning of the earth’s ozone layer) have lent an urgency to the study of UV-absorbing compounds.   Kuffner et al (1995) analyzed the MAA content of several coral genera (Pocillopora, Porites and Montipora spp.).   They found that MAA concentrations were inversely proportional to depth with its associated lower dosage of UV energy. (We must note that other researchers have suggested that water motion and other environmental factors may trigger the production of MAA’s.)   Gleason (1993) transplanted brown Porites specimens from areas of relatively low UV energy to areas of higher UV and found these corals to exhibit reduced growth and possible photoinhibiton (he felt that UV energy in the range of 310-350 nm was responsible).   He also found that corals could proportionally alter the amounts of certain MAA’s in response to UV levels.   MAA’s may be obtained through the diet (as in the case of the non-hermatypic coral Tubastrea); it is also possible that zooxanthellae translocate MAA’s to the coral animal.   In any case, the coral’s response to increased UV may be an increase in MAA’s within its tissues.   This response is not instantaneous.   (The analogy of a pasty-white sunbather trying to get a golden tan in one day is appropriate!)

In nature, corals exist in a relatively stable UV environment.   If we were to graph diurnal UV doses, we would see a bell curve, that is, UV increases as the sun climbs toward its apogee and then decreases towards sunset.   UV may peak during a given month (depending upon latitude) but these changes are incremental and predictable.   Such is not the case in aquaria.

UV Depreciation of Aquarium Lamps
During the course of our investigations, we’ve checked the time-course UV output of many lamps.   Figures 10 and 11 show this UV-A output of two metal halide lamps.   As can be seen, one lamp’s UV output dropped over a period of time.   Interestingly, UV output increased rather dramatically in the other lamp.   Our point is that neither of these lamps offers a stable UV environment.   Increases will occur either during normal operation of the lamp or when the hobbyist “suddenly” replaces the lamp. (We’ve found fluorescent lamps lose their UV output over time as well.) 

This leads us to an important question – Is there a risk of eye damage from an aquarium using a light source with high UV output?   We do not claim to be ophthalmologists and are not qualified to give medical advice.   We can report the results of our testing.   Using a glass aquarium, we found UV-A levels of up to 70% being transmitted through the glass above the water surface.   Under certain conditions, we believe this level of radiation could exceed that of natural sunlight.   Most acrylics attenuate UV quite well; however, comprehensive testing is required before a definitive statement can be made. 

Coral Coloration and UV Radiation
It is quite popular to believe that increased coral coloration is a response, at least in part, to UV radiation.   Our experiences indicate that some corals will turn green as a response to increased UV.   However, we have observed many corals (especially Acroporids, Pocilloporids, etc) exhibiting vivid coloration when maintained for years under conditions of practically no UV (~1 µW UV-A; <1 µW UV-B).   Figures 12 and 13 show two Acropora specimens maintained under such low UV levels (however, visible light – PAR – levels were quite high).

Review
These short articles have shown that all the lamps we tested produced UV radiation.   Metal halide lamps’ outer glass envelopes only weaken UV; they do not eliminate it.   The actual amount of UV produced by a metal halide lamp depends upon many factors such as lamp wattage, lamp orientation and arc tube construction (universal burn position versus bottom up or bottom down).   Profound differences can exist between “identical” lamps.   More importantly, the shape (along with the type of reflective surface) can focus UV energy into “hot spots” where the UV energy exceeds that found in nature.   Even higher UV energy is produced during lamp strike and water surface waves can further focus this radiation.   Seemingly small differences (such as lamp height) can also have a major impact upon the amount of UV entering an aquarium.   With so many variables involved, it is difficult, if not impossible, to know how much UV is being produced without actually making measurements.

We have seen that UV radiation has the potential to damage corals and other coral reef inhabitants.   Many reef animals can produce natural sunscreens (MAA’s) to protect themselves against UV but MAA concentrations are possibly a response to the amount of UV to which they are subjected.   This is an important point since UV production among all tested lamps was not consistent and can change dramatically during normal operation or when the lamp is changed.

Our experiences suggest that coral coloration is a response to PAR levels, not UV.   In short, we find no reason to subject reef aquaria to high UV levels. 

Shielding Aquarium Lamps
Hobbyists may think their aquaria are doing fine under unshielded lamps.   We suggest that they can do even better by removing UV radiation.   We’ve checked many “plastics” and Table Five lists the transmission properties.

*Window glass purchased from a hardware store.
**We’ve found UF-3 to retain its UV-screening properties even after a decade of use.   Coralife and Hamilton reportedly use this material in their luminaires.
***From a Giesemann aquarium luminaire.   Material was yellowed and may have lost its UV screening properties as a result of aging.

We strongly recommend that an appropriate material be used to attenuate (weaken) UV energy produced by all aquarium lamps and especially metal halide and other lamps (such as mercury vapor).  Since UV acting in synergy with temperature damages some corals, removing UV may allow an aquarium to run at slightly higher temperatures (thus lessening the use of expensive water chilling).  There are tradeoffs.  Some photosynthetically active radiation (PAR) will be lost as a result and the shielding material will require periodic cleaning.    Based on our observations (and through discussions with literally hundreds of hobbyists across the country), we believe that these tradeoffs are well worth the benefits. 

We wish to thank Omer Dersom, John Lipsey and Ron Feigen for supplying some of the lamps and/or equipment used in our testing.
 

References
Baker, A., 1995.   The effect of UV on the settlement of the planula larvae of  Pocillopora damicornis.   In: Ultraviolet Radiation and Coral Reefs.  D. Gulko and P.L. Jokiel (eds.), HIMB Technical Report #41.

Bingman, C., 1995.   The effect of activated carbon treatment on the transmission of visible and UV light through aquarium water. I: Time-course of activated carbon treatment and biological effects.   Aquarium Frontiers 2(3): 4,5,16-19.

Donkor, V. and D. Hader, 1997.   Ultraviolet radiation effects on pigmentation in the cyanobacterium Phormidium unicinatum.   Acta Protozoologica 36:49-55.

Dunlap, W.C. and B.E. Chalker, 1986.   Identification and quantification of near-UV absorbing compounds (S-320) in a hermatypic scleractinian.   Coral Reefs, 5:155-159.

Dunlap, W.C., B.E. Chalker and W.M. Bandaranayake, 1988.   Ultraviolet light absorbing agents derived from tropical marine organisms of the Great Barrier Reef, Australia.   Proc. 6th Int. Coral Reef Symp., Australia.   3:89-93.

Figueroa, F.L., S. Salles, J. Aguilera, C. Jiménez, J. Mercado, B. Vinegla, A. Flores-Moya and M. Altamirano, 1997.   Effects of solar radiation on photoinhibition and pigmentation in the red alga Porphyra leucosticta.  Marine Ecology Progress Series, 151:81-90.

Gleason, D.F., 1993.   Different effects of ultraviolet radiation on green and brown morphs of the Caribbean coral Porites astreoides.   Limnol. Oceanogr., 38(7): 1452-1463.

Gleason, D.F. and G.M. Wellington, 1993.   The intensities of ultraviolet radiation that induce bleaching of a Caribbean coral.   Proc. 7th Int. Coral Reef Symp., Guam. 1: 71. (Abstract).

Gulko, D., 1995.   Effects of ultraviolet radiation on fertilization in the Hawaiian coral Fungia scutaria. In: Ultraviolet Radiation and Coral Reefs.   D. Gulko and P.L. Jokiel, eds.   HIMB Tech. Report #41. 

Hohlbauch, S.V., 1995.   The metabolic response of Fungia scutaria to elevated temperatures under various UV light regimes. In: Ultraviolet Radiation and Coral Reefs.   D. Gulko and P.L. Jokiel, eds.   HIMB Tech. Report #41.

IES Lighting Handbook, 1984.   John Kaufman and Jack Christensen, Editors.   Illuminating Engineering Society of America, New York.

Jerlov, N., 1950.   Ultraviolet radiation in the sea.   Nature (London) 166: 111-112.

Jerlov, N., 1976.   Marine Optics.   Elsevier Oceanography Series, Elsevier Sci. Publ. Co., New York. 231 pp.

Jokiel, P.L., 1980.   Solar ultraviolet radiation and coral reef epifauna.   Science, 207:1069-1071.

Joshi, S. and D. Morgan, 1998.  Spectral analysis of metal halide lamps used in the reef aquarium hobby. Part 1: New 400-watt lamps.   Aquarium Frontiers On-line.  November. 8 pp.

Kawaguti, S., 1944.   On the physiology of reef corals VI. Study on the pigments.   Palao Trop. Biol. Sta. Study, 2:617-674.

Kuffner, I.B., M.E. Ondrusek and M.P. Lesser, 1995.   Distribution of mycosporine-like amino acids in the tissues of Hawaiian scleractinia: a depth profile. In: Ultraviolet Radiation and Coral Reefs.   D. Gulko and P.L. Jokiel, eds.   HIMB Tech. Report #41.

Lewis, S., 1995.   Response of a stony coral to short-term exposure to ultraviolet and visible light.   In: Ultraviolet Radiation and Coral Reefs.   D. Gulko and P.L. Jokiel, eds.   HIMB Tech. Report #41.

Lynch, D. and W. Livingstone, 1995.   Color and Light in Nature.  Press Syndicate of the University of Cambridge.   New York. 254 pp.

Shibata, K., 1969.   Pigments and a UV-absorbing substance in corals and a blue-green alga living in the Great Barrier Reef.   Plant and Cell Physiol., 10:325-335.

Shick, J.M., W.C. Dunlap, B.E. Chalker, A.T. Banaszak and T.K. Rosenzweig, 1992.   Survey of ultraviolet radiation-absorbing mycosporine-like amino acids in organs of coral reef holothuroids.   Mar. Ecol. Prog. Ser., 90:139-148.

Shick, J.M., Lesser, M.P. and W.R. Stockaj, 1991.  Ultraviolet radiation and photooxidative stress in zooxanthellate anthozoans: the sea anemone semoni and the octocoral Clavularia sp.   Symbiosis 10: 145-173.

Stramski, D. and L. Legendre, 1992.   Laboratory simulation of light-focusing by water surface waves.   Mar. Biol., 114:341-348.

Wu Won, J., B. Chalker and J. Rideout, 1997.   Two new UV-absorbing compounds from Stylophora pistillata: sulfate esters of mycosporine-like amino acids.   Tetrahedron Letters, 38(14):2525-2526.

Wu Won, J., J. Rideout and B. Chalker, 1995.   Isolation and structure of a novel mycosporine-like amino acid from the reef-building corals Pocillopora damicornis and Stylophora pistillata.   Tetrahedron Letters, 36(29):5255-5256.