Below the Edge of Darkness Read online

  To provide additional insurance against the entry of unauthorized photons, I hung a flap of black plastic over the slider as a light baffle and cut it off at exactly my height. A lot of good-natured ribbing went on in the Case Lab and, at five foot two, I had come in for my share of short-people jokes, so I hung a sign on the door that read, ALL YE WHO MUST STOOP TO ENTER ARE TOO DAMN TALL. I was a little unsure how Case would respond to my grab for territory, so I was much relieved when I came in Monday morning to find that he had got there ahead of me and written a big A+ on my sign.

  When Virginia Woolf extolled the benefits of having a room of one’s own, she was focused on the need for carving out space and time for writing. I now had space and time for doing science on my own terms. The sense of freedom it afforded was intoxicating. I could sit in the dark with my dinoflagellates and observe their light-making one on one. It was thrilling the first time I managed to record a sequence of flashes from a cell that I knew had had no previous stimuli.

  This was something that no one had ever seen before, and it was wonderfully weird. In response to a series of stimuli, the first flash was more than ten times brighter and fifteen times faster than the subsequent ones. To understand its underlying cause, I needed to be able to visualize the inner workings of the cell. Although I could see a great deal just looking through a microscope, much of what was happening with the flash was so fast, it required slow-motion playback for careful analysis. But that would call for a high-resolution image intensifier, which was, unfortunately, exorbitantly expensive.

  The funding for my research came from a modest grant that Case had from the Office of Naval Research. In those days, ONR funded basic science projects related to areas of Navy concern. Their interest in bioluminescence was strategic. The last German submarine sunk in World War I was detected non-acoustically, when it was clearly outlined by the luminescence it stimulated. In the cat-and-mouse games of anti-submarine warfare, bioluminescence had significance for both the hunter and the hunted.

  ONR had been funding Case’s research on fireflies in the hopes of gaining a better understanding of bioluminescence, and my research on dinoflagellates greatly heightened their interest. Case undoubtedly had exactly this response in mind when he suggested the project. The upshot was that in the next grant cycle, when he reported on my findings and requested funding to purchase a bunch of new equipment, including a high-end image intensifier, it was granted.

  Now I had my dark room and all the toys I could possibly want, including a way to see what was going on inside P. fusi. What I found in there was like a microscopic curiosity cabinet, but rather than being dark, dusty, and filled with dead things, it was illuminated with twinkling starlight that revealed the inner workings of a cell that seemed to be constantly rearranging the furniture.

  I could see that the difference between the first flash and subsequent flashes was a consequence of the summed activity of microflashes from the scintillons. The first flash was so bright and fast because the scintillons were activated in synchrony, while in subsequent flashes they were asynchronous, appearing like a scintillating star field where the microflashes increased in number to a peak and then decreased in number more slowly, sometimes with individual scintillons emitting light more than once during the course of the flash.

  I was discovering more about the electrical activity that triggered the flash. Unlike most animal neurons, where the action potential is initiated by sodium ions rushing through membrane channels, in P. fusi it was hydrogen ions. To trigger the action potential, all I needed to do was bump the cell in a way that distorted the membrane. And I found that the action potential that triggered the light output was present in both day-phase and night-phase cells, even though no flash was produced in day phase.

  The shutting down of bioluminescence was accompanied by a total rearrangement of the cell’s interior. During the day, the scintillons left the periphery of the cell and migrated in to cluster around the nucleus, near the center of the cell, while the opaque chloroplasts—the organelles where photosynthesis occurs—migrated outward, spreading across the surface of the cell like a light-gathering solar array, maximizing light capture. By contrast, at night the opposite arrangement occurred, with the chloroplasts clumping around the nucleus, while the scintillons spread across the cell’s surface to optimize light emission. At sunrise and sunset, the region around the nucleus was like rush hour at a futuristic spaceport, with organelles shuttling back and forth in what amounted to highly organized chaos.

  These cells don’t have brains, much less eyes, nor are they able to swim. They float where the water takes them. What possible reason could there be for a single cell to require such complex control mechanisms and peculiar flash patterns?

  At one time, the purpose of light emission in dinoflagellates was so unfathomable that it was simply deemed to be a by-product of some other cellular function. But clearly a daily rhythm that assures that bioluminescence occurs only at night, when it will be visible, suggests otherwise.

  The possibility that the bioluminescence serves a defensive function became apparent in 1972, with an elegant experiment that took advantage of the circadian rhythm in dinoflagellate luminescence to demonstrate that insect-like creatures called copepods, which are common dinoflagellate predators, grazed less on bioluminescent night-phase cells, as compared with non-bioluminescent day-phase cells. Why should bioluminescence prevent copepods from feeding?

  Most copepods find their prey by setting up a feeding current, which they create by beating paintbrush-like appendages, drawing phytoplankton toward their mouths. This sounds indiscriminate, like water disappearing down a drain, but it’s not. Copepods can actually manipulate their feeding currents and sort the particle stream into rejected and accepted food.

  So one hypothesis is that the flash serves as a warning of toxicity, causing the copepod to reject cells that flash. Many bioluminescent dinoflagellates are toxic, and being toxic is a good way to deter predators, but only if predators recognize the prey that they need to avoid. If they have to nibble on the prey every time to determine its toxicity, then nobody wins—the prey is damaged or dies and the predator gets sick or dies. Far better for both players if the predator can learn to recognize toxic prey from a distance, through the prey’s advertising its toxicity. The vibrant orange-and-black wings of monarch butterflies, which are toxic to birds, are one such advertisement. It’s a very clear visual signal, effectively saying, Don’t eat me or you’ll be sorry! Perhaps the dinoflagellate flash conveys the same message.

  An alternative hypothesis is that the flashing of bioluminescent dinoflagellates functions as a burglar alarm. Just the way the beeping horn and strobing lights on a car’s alarm system serve to expose a burglar, forcing him to flee or risk capture, the dinoflagellate flashing might expose a copepod—which would otherwise be swathed in a protective cloak of darkness—to detection and consumption by its visual predators such as fish.

  The fear screams of prey, heard in some frogs, birds, and monkeys, are examples of these putative burglar alarms. A frog being eaten by a water shrew emits screams loud enough to attract the attention of a hawk that might attack the shrew and cause it to drop the frog. The point of a burglar alarm is for prey, in the face of imminent death, to use whatever attention-grabbing means are available, such as sound, light, or odor, to attract the attention of another predator that may attack their attacker. This not only offers the prey an opportunity for escape but has the added benefit of possibly permanently removing the predator from the scene.

  I have spent most of my career observing bioluminescent flashes and trying to understand what information they convey. My early experience with the strange flash patterns of P. fusi taught me to recognize that not all flashes are alike. There are big differences, and those differences have meaning. Although many discussions of bioluminescence in dinoflagellates assume that flashes serve the same function for all dinoflagella
tes, I think not. Unlike P. fusi, most dinoflagellates that can emit light do so in the form of one or two very dim, short flashes. Many of the dinoflagellates that emit such flashes are toxic, which suggests that, for them, the flash is a warning of toxicity. In such a case, a tiny pinprick of light in the darkness is all that is needed, because it’s a private conversation between the prey and its predator. The dimness of the flash benefits both, allowing the prey to conserve energy and the predator to avoid detection by its visual predators, which means it doesn’t need to stop feeding and swim away every time it sees a flash.

  By contrast, the first flash of P. fusi is more than one hundred times brighter and, as such, is in no way a private conversation. It is a scream for help that will readily draw the attention of visual predators and continue to do so with its subsequent numerous and prolonged flashes, illuminating the offending attacker, making it easy prey for its predators. In the face of such exposure, it behooves the predator to stop feeding and swim away.

  Nature is frequently far more complicated than we appreciate, and that complexity can be a source of confusion when we make simplifying assumptions—like the supposition that one bioluminescent dinoflagellate is like any other. There have been a number of experiments carried out over the years to test the burglar alarm hypothesis versus the toxic warning one, with sometimes conflicting results. I believe that these can best be explained by the different species and different concentrations of dinoflagellates used.

  One of the dinoflagellates used in these experiments is one of those with a dim flash, Lingulodinium polyedra.*8 Oddly, while copepods refuse to eat L. poly in its bioluminescent night phase, they gobble it up like candy during its day phase. In other words, it’s not toxic, which seems to disprove the toxic-warning hypothesis. However, it’s important to remember that nature is full of cheats—organisms that send false signals. There are many examples of animals that mimic the appearance of toxic or otherwise dangerous animals in order to take advantage of predators’ learned avoidance behavior. The brilliant red, black, and yellow banding pattern of coral snakes advertises their toxicity to predators like hawks and coyotes. King snakes, though not toxic, masquerade as coral snakes with a similar banding pattern,*9 thereby acquiring protection from predators without expending the energy needed to synthesize the toxin.

  Also, in some experiments with toxic dim-flashing dinoflagellates, copepods behave as they would with a burglar alarm: They stop feeding and swim away. However, in those experiments the concentration of dinoflagellates used was very high. My suspicion is that when the dinoflagellates become too abundant, even though the flashes stimulated by the copepod feeding are dim, there are so many of them that the copepod risks visual exposure and so executes an escape.

  Long after my graduate student days, thanks to the experimental studies of a couple of excellent grad students in my own lab, Kathleen Cusick and Karen Hanley, working with dim emitters like L. poly, we were able to demonstrate that in a mixture of bioluminescent and non-bioluminescent plankton, copepods feed selectively on the non-luminescent food at low cell densities. However, above a certain threshold number of bioluminescent dinoflagellates, they stop feeding on the non-luminescent prey and swim away. On the other hand, we found that with bright emitters like P. fusi, a single flash is enough to attract the attention of fish predators, so its bioluminescence functions as a burglar alarm at any cell density.

  Further adding to the complexity of these interactions are recent findings demonstrating that copepods release chemical cues that cause some bioluminescent dinoflagellates to increase their light output. These chemicals can also lead certain toxic dinoflagellates to increase toxin production. In other words, dinoflagellates can sense the presence of predators and adjust their defenses accordingly. With every new discovery, the world of sea sparkle becomes “curiouser and curiouser!” *10

  Relationships between living creatures are extraordinarily complex. Trying to sort out these associations to determine what part different bioluminescent signals play in a laboratory environment is fraught with pitfalls. It’s hard enough to keep one critter happy and healthy in a laboratory setting; when you’re dealing with three or more, in order to sort out multitrophic connections, the challenges quickly multiply.

  To comprehend the meaning of any animal’s visual signals, it’s essential to put yourself in their shoes—or chelae or flippers. In other words, you need to be able to imagine what their world looks like. That’s a grand challenge, because even though it’s part of our world, it’s a place very few people have seen in its natural state.

  Skip Notes

  *1  Scientists, as a rule, are pretty picky about the naming of things. Every organism is given two names. The first is the genus and is always capitalized; the second is the species and is never capitalized. In scientific publications (and in this book) both names are italicized and written out the first time they appear. Subsequently, either to save ink or to cut down on writer’s cramp, the genus is abbreviated to a single letter.

  *2  To be fair to sharks (and dinoflagellates, for that matter), over the past decade the worldwide average of fatalities from shark attacks has been six per year.

  *3  Armor in dinoflagellates consists of cellulose plates that form a protective shell. Some plates are smooth, but some sport elaborate pores and grooves that make them look very much like the armor plating on a crocodile.

  *4  One of a seemingly endless array of euphemisms for seasickness, which are amusing to everyone but the victim.

  *5  Motorboats stir up sediments that can increase algae growth due to nutrient resuspension and decrease sunlight for dinoflagellates. A two-stroke engine, like the one I grew up water skiing behind, can release 25 to 35 percent of its unburned gas and oil into the water. Which explains why, when motorboats were eventually banned on the pond I grew up on, it became a much healthier ecosystem with cleaner water and abundant birdlife.

  *6  A further abbreviation—unauthorized but commonly used.

  *7  That’s 22.4 miles per hour, which doesn’t warrant a speeding ticket, even in Boston, which recently instituted a 25-mile-per-hour default speed limit to try to escape its title as “the most exciting city in America in which to drive.”

  *8  Formerly known as Gonyaulax polyedra. As I mentioned earlier, scientists are very picky about the naming of things. The proper assigning of names is actually the foundation of biology, but that doesn’t mean that the taxonomists who come up with those names are well loved. In fact, the renaming of this particular organism raised some considerable ire in the bioluminescence community because so much research had been published on it under its old name.

  *9  It’s not a perfect disguise, which is why learning this poem can save your life: “Red touch yellow—kills a fellow. Red touch black—venom lack.” On the other hand, if you’re not into poetry, just try to remember that the one with the black nose is bad news.

  *10  Almost as trippy as Alice in Wonderland!

  Chapter 4

  THE STARS BELOW

  The knot in my stomach tightened as I walked out onto the fantail to look yet again at the steel cable disappearing into the depths behind the ship. Dangling from the end of that cable, some eight hundred feet below, was friend and fellow scientist José Torres, taking his turn in a deep-sea diving suit called Wasp. The sun was getting low in the sky and I was starting to worry that I might have to wait until the next day for my first deep dive. The Wasp crew said they were saving the best for last, but I wasn’t fooled: I knew they were saving the shortest for last.

  Developed by the offshore oil industry for the purpose of diving on oil rigs to depths of two thousand feet, the suit is called Wasp because it looks like a big yellow insect with a transparent head, a yellow tube body, and metal Michelin Man arms with pincers instead of hands. There are no legs for walking on the bottom. Instead, the suit
has thrusters on the outside, controlled by foot switches attached on the inside to a metal plate that can be adjusted to accommodate the stature of the occupant. It took time to pump up the floor; consequently, going in order of height and saving the shortest for last made sense, but for me it felt like that interminable stretch of time leading up to Christmas, when, as a kid, the more desperately I wanted it to come, the more drawn out its advent became.

  It had all started innocently enough, three years earlier, in 1981, when Dr. Case scored a new toy for the lab. The Optical Multichannel Analyzer (OMA) was the most sensitive spectrometer ever developed for measuring the color of dim, transient light sources. From the moment it came into the lab, I couldn’t keep my hands off it.

  Traditionally, color is measured by dispersing light with a prism or a diffraction grating, either of which will transform white light into a rainbow. The prism or grating is rotated so the rainbow is scanned across a sensitive light detector that measures the spectrum, one narrow bit of the rainbow after the next. This works well for measuring spectra of relatively stable light sources, but not for measuring brief flashes of bioluminescence. The OMA eliminated the need for scanning by replacing the single photomultiplier tube with a row of seven hundred solid-state detectors that could measure the contributions of different parts of the rainbow all at once instead of consecutively. It was an awesome technological breakthrough that opened up a whole new way of looking at bioluminescence.