Methods
In August 1999, a LiCor LI-1000 data logger was used in combination with a calibrated LiCor LI-193SA spherical quantum sensor to measure PAR (photosynthetically active radiation) levels over a three-day period, at a depth of 24”. The LiCor unit took continuous PAR measurements from 07:15 until 18:30. Every fifteen minutes the maximum and mean PAR values from this time period were calculated and stored in memory. Measurements were cut short on the third day due to exhibit maintenance. The data was then transferred via an ASCII dump and serial cable from the data logger unit into a computer and graphed using MS Excel.

On August 1^st, 2^nd and 3^rd of 2002 it was decided to take PAR measurements again. The original tank had been removed but another, identical tank was still outside so this was moved into the exact same position as the anemone tank. The tank was filled with water and gravel placed inside, and the LiCor sensor was placed 24” below the surface.  The same acrylic cover as originally used was then placed over the tank and PAR measurements were taken. Shade cloth was then added and PAR measurements were again taken on August 16^th, 17^th and 18^th (the delay was due to several weeks of high cloud and overcast conditions). No values were collected prior to 08:30 or after 18:00 since the threshold PAR level of 50 uM/m²/s was not reached and the data logger was not triggered to record. The mean and maximum values for both shaded and normal conditions were then graphed and compared to those taken in 1999 (figures 1 to 8).

Results
Figures 1 and 2 (mean and maximum PAR) show that light levels in the indoor tank began to gradually increase at 07:00 as the sun rose into the sky. The greatest increase began at 10:30 corresponding to the sun rising above the concrete wall of the building and shining directly onto the tank. There was then a slight drop in PAR around 11:15 and again at 12:45. These dips in light intensity between 11:15 and 12:45 each day were due to the shadow of one of the skylight support beams falling over the sensor (see photo 5). However, since the shadow was narrow, other areas of the tank would still have received full sunlight. The greatest PAR levels occurred between 11:00 and 13:15 reaching a maximum 1448 uM/m²/s at 12:00 when the sun was shining directly into the tank and no shadows from the skylight hardware were over the sensor. The sharp drop after 13:15 was due to the sun passing behind the concrete wall on the ocean side of the building and thus, direct sunlight no longer fell on the water surface of the tank, though indirect light continued to impact the exhibit until sunset. Drops in PAR seen on one day but not others were most likely due to passing clouds. From 07:15 to 10:45, and from 13:15 to 18:15 PAR levels did not exceed 300 uM/m²/s. Therefore, for 8.5 hours PAR levels were less than 300 uM/m²/s and maximum levels were attained for only 2.5 hours.

Figures 3 and 4 show the mean and maximum PAR values in August of 2002 without (8/1-8/3) and with (8/16-8/18) shade cloth installed on the outdoor exhibit. At the outdoor location, buildings and nearby trees impacted the light field at certain portions of the day. The most noticeable ones can be seen in figures 3 and 4 at 12:15 and 14:45 when there was a dramatic drop in mean and maximum PAR values due to the shadow of a palm tree passing over the sensor. When no shade cloth was used maximum PAR values exceeded 500 uM/m²/s between 09:00 and 17:15 and between 09:45 and 15:45 maximum PAR values exceeded 1500 uM/m²/s. The maximum PAR value reached was 2634 uM/m²/s at 13:45. When shade cloth was added to the cover, PAR values (both mean and maximum) were dramatically lower and did not rise above 500 uM/m²/s over the duration of any of the three days.

Figures 5 and 6 compare the mean and maximum PAR values of the indoor exhibit versus the values of the outdoor tank without shade cloth. It is clear that the outdoor tank not only received higher PAR levels but also did so for a much longer duration.

Figures 7 and 8 compare the mean and maximum PAR values of the indoor exhibit versus the values of the outdoor tank with shade cloth. In this case, the outdoor tank tended to exhibit two to three times higher PAR values than the indoor tank between 07:15 and 10:15, and between 13:15 and 17:15. However, from 10:15 to 13:15 the indoor tank exhibited much higher levels than the outdoor tank.

Discussion
The factors that are involved in bulb tentacle development in E. quadricolor have been speculated upon for several years. Most of these have been based upon anecdotal cause and effect observations, speculation and rumors perpetuated via print and electronic media. To be sure, there are many instances where bulb tips disappear and reappear for no apparent reason. Combine this with the fact that there are at least two recognized forms of E. quadricolor (shallow and deepwater), and it becomes apparent that predicting the behaviour of these animals becomes problematic. This study, by no means the definitive word, offers documented evidence for the role of light in the development of bulb tips.

There were several sources of potential error. Although this investigation compared data over two distinct time periods (1999 and 2002), they were taken in the same month (August) and the differences in light levels between the indoor and outdoor locations are significant and not likely due to differences in the years. Another source of error may be that the light measurements taken in 2002 occurred several months after the animals were removed from the exhibit. This means that the light levels from October 2001 till May 2002 may have been slightly lower and the photoperiod would have been shorter than in August 2002. However, even under these conditions light levels would still have been greater than those recorded indoors in 1999 and bulb tips did appear in the outdoors exhibit during this time frame. Photoperiod would have been shorter than in August of 1999, but since these difference would have been most apparent at the beginning and end of the day, when the indoor exhibit would not have received direct light anyway, it is unlikely it would have had any great effect on the final results.

From this investigation it was readily apparent that both light intensity and duration were extended in the outdoor location compared to the indoor location. This makes it tempting to assume that both factors may be of importance, but they may also be mutually exclusive; unfortunately this cannot be determined from the data in this study.

Since being moved indoors in May 2002, many of the E. quadricolor in the new exhibit (basically in the same position as the old indoor exhibit) have retained their bulb tips but some show signs of loosing them. Several E. quadricolor were also placed in a 5500 gallon reef exhibit that receives more hours of direct sunlight than the new anemone exhibit. Several of the anemones in this exhibit have moved to various locations in the tank (shallow vs. deep, low current vs. high current) so it will be interesting to note if they retain their bulb tips and for how long.

Observations on Sexual Reproduction in Entacmaea and Stichodactyla
Reproduction in E. quadricolor takes two forms, sexual and asexual. Asexual reproduction has been well-documented by hobbyists and was a common occurrence in this exhibit as well resulting in a large population of these anemones such that the live rock was nearly invisible when all the anemones were expanded. Sexual reproduction also occurs in this tank (see Sprung and Delbeek, 1997) and to the best of my knowledge, is the only such population in captivity that regularly does so each year. Spawning usually occurs between April and May between 0700 and 0900 a few days after a full moon has occurred. In April 2002 we were fortunate to once again witness not only the sperm release but also the release of planulae. Sperm release can occur in the morning but we have also observed it occurring in the late afternoon and this was the case this year. The exhibit was found to be rather cloudy in the late afternoon of April 25^th and the incoming water flow was increased to ensure the sperm did not foul the water by being allowed to accumulate for too long and lowering oxygen levels. Four days later, on the morning of April 29^th numerous green sphere-like objects were observed floating in the water. Upon closer examination it was found that the tentacle tips of several colonies of E. quadricolor contained these same spheres. Occasionally a few were observed being released from the tips of the tentacles. From previous experiences we knew these to be anemone embryos (see Sprung and Delbeek, 1997 for photos of the embryos and settled juveniles). A few days later there were still some visible in the tentacles and these were observed to be actually rotating and moving within the tentacles, and were more oblong in shape. It appeared as if the embryos had developed into planulae in the anemone and would shortly be released as such into the water. From my observations it would appear that E. quadricolor is dioecious, with males releasing sperm into the water column to be taken in by the females where their eggs are fertilized internally. A few days later embryos/planulae are released into the water column via the tentacle tips. Sprung and Delbeek (1997) also documented the presence of zooxanthellae in juvenile E. quadricolor only a few days after settlement. This time a few embryos were placed under a microscope and were found to contain zooxanthellae as well. Therefore, not only does the female brood the embryos for a short period but she also supplies them with their initial supply of zooxanthellae. As a result it is probable that these planulae can either settle very quickly and metamorphose into fully functional anemones, or they can spend a great deal of time in the plankton till they find a suitable substratum to settle on, without the need for feeding.