Origins of Bioluminescent Communication

Independently, in a thousand and more phyletic lineages, individuals became luminescent. (The alternative explanation that all luminescent organisms trace their luminescent geneology back in unbroken succession to a common luminescent ancestor and that all contemporary nonluminescent beings are descended from photic dropouts is unreasonable.) The alleles required for light production either were part of the gene complex that Seliger and McElroy suggested was put together long, long ago and maintained until recently, or recently were fortuitously added in some context other than light production. How many genes constitute a "luminescent package"? How many genes might be from the remote past, and how many were selected in a pleiotropic context? In any event it was probably a substitution at a single locus that finally turned each light on. Regardless of context or antiquity of origin, when the light came on, it gave its bearers a new ecological status. Of interest here are organisms whose Darwinian fitnesses were improved by light emission and the reasons for that.

ENHANCEMENT OF REFLECTED LIGHT SIGNALS BY MEANS OF LUMINESCENCE

Luminescence might simply have enhanced visual signals that were already important in behavioral interactions. If it appeared in or near appropriate anatomical sites, then the luminescence might have emphasized, amplified, or highlighted already-existing signals, such as postures, movements, gestures, areas of pale color, or reflective surfaces, that had hitherto depended entirely on reflected light for their efficacy. This effect would be significant in poorly lit habitats, and especially so in transitional ones such as diurnal and nautical twilight zones. In dark places there is no reflected light to enhance, and in well-lit ones bioluminescent amplification seldom can bestow selective advantage. Twilight zones combine essential elements—organisms with well-developed photoreceptors and the benefit from amplification of reflected light. Perhaps this partially accounts for the facts that two- thirds of the fish of the mesopelagic zone are luminescent and that there are no luminous freshwater forms.

The appeal of this model is that initially only instructions for luminescence itself were required of the genome. Already present, regardless of whether the signal recipient was of the same or different species, were the visual organs, data-processing neural circuits, appropriate scanning or search patterns, "attention to detail," and all the other essential ecological, physiological, and behavioral adaptations. It is obvious that the augmentation of visual signals by luminescence might have ultimately permitted an ecological shift into a nocturnal activity period or a truly dark habitat and, when viewed after the fact, can indeed be recognized as a transitional stage.

The light emitted by the fruiting bodies of several species of fungi is bright enough to be seen at distances of several meters. Since a number of insects feed and oviposit on fungi it is easily imagined that insects might use the luminescence as a beacon. Furthermore, natural selection might favor the maintenance of luminescence in the fungus because the fungus used the insects for some vital function, such as transporting spores to other accumulations of decomposing organic matter (Lloyd, 1974). Such a mutualistic relationship, sans luminescence, exists between stinkhorn fungi and green- bottle flies in European woods. The caps of fresh fruiting bodies of the fungus are covered with a green black, shiny layer of spore slime. The flies orient to the color and the sheen of the cap, land on it, and eat slime and spores. The slime is dissolved by the saliva and digested in the gut. The spores are later deposited in the feces (Wickler, 1968:155). Consider for a moment a nonluminescent fungus whose shiny white or greenish white cap was a signal to its insectan, spore-carrying symbiont. Would not the first appearance of luminescence in the fruiting cap enhance its signal and make it operable when reflected light was inadequate, such as after sunset or on the floor of a dark forest or cave entrance?

The light emitted by the larviform females of glowworm beetles (Phengodidae) is bright and can be seen for several meters, and it emanates from several sources over the surfaces of the insects. Females of the genus Zarhipis, from the western United States, are brightly luminescent. Workers there concluded that the luminescence is not associated with mating behavior and that males use only pheromones for locating the female. The antennae of male glowworm beetles are feathery and large, presumably greatly increasing the number of chemoreceptors and sensitivity, and perhaps permitting stereo-detection of molecules. Rivers (1886) found that Zarhipis males are attracted to females during daylight but that they fly only in temperate heat, from 9 A.M. to 4 P.M., so that in hot weather they do not appear "until the sun is declining." Tiemann (1967) noted that earlier workers had "observed that males will approach females during the day in the humid coastal area. Although males were not observed to approach females during the day in the low relative humidities of the desert environment, they did come to the female at dusk . . . males were attracted to the females within 10 minutes . . . the glow from the females was barely visible in the twilight." If males orient visually at close range to the pale-banded females and perhaps specifically to the pale-band pattern, luminescence would, at its first occurrence, in the low ambient light of dusk, enhance the female's visual signals (Lloyd, 1971).

There are many other potential examples of this origin for luminescent signals, but the articulation of complete models depends on detailed knowledge about the behavior of both luminescent and nonluminescent members of the taxon involved. It is inviting to speculate that the shimmers and reflections of the silvery and iridescent bodies and bright-colored markings of some fish, cephalopods, and crustaceans, all being organisms with well-developed eyes, predisposed members of these groups for a subsequent evolution of luminescent signals through the transitional stage of enhancement of reflected signals. The "brighter-than-life" reflections from the lateral lines of some fish, such as the neon tetra of home aquaria, shine as though reinforced by luminescence. Other fish do have lanterns distributed along their lateral lines. Body undulations or other communicatively significant body movements may be highlighted, or transmitted in their entirety in the case of luminescence, by these optical phenomena. Lanterns on other fish outline the body, mark the opercula and fins, or produce a (shimmery?) glow over the lateral or ventral surface. Perhaps circular light organs in the region of the anal fin ("egg mimics") stimulate and direct the activity of males during courtship and fertilization. Table 1 lists various displays employed during mating and aggressive encounters in non- luminescent fish, and suggests lantern positions that would amplify them.

About one-third of the known species of squid are luminescent. This may seem surprising at first, but not when one considers the nature of the skin of squids in general. In their mantles are pigment bodies (chromatophores) that can be expanded and contracted by means of muscle fibers. There are also reflectors or mirrors (irido- phores, iridocytes), some of which lie above the chromatophores and some below. Color changes can be made rapidly and are under the control of the central nervous system. They sweep across the body with "rapidity and variety more like that of an electric sign than an animal" (Lane, 1960:94). "It pulsated slowly, while the colors came and went over its body in such a way that new adjectives will have to be coined adequately to describe it—reds, blacks, browns, yellows, rolling, surging, springing into vision as the pigment spots contracted or expanded, a living, liquid palette" (Beebe, 1926). This would seem to predispose squids for the incorporation of luminescent amplification when chemical and ecological opportunities occur.

Some peculiar patterns of photophore arrangement found on fish and squids may be fully explicable only when the natures of the light- analyzing mechanisms of the signal recipients are completely understood. Counterparts of the optical illusions of the human visual system are to be expected, and some lantern arrangements may be but abstractions of their phylogenetic precursors and their reflected-light analogues or homologues.

Table 1

Some behavioral displays known from nonluminescent fish and the light-organ positions that could be significant if similar displays occur during mating, aggressive encounters, or other interactions of luminescent fish.

NOTE: Similar displays which also depended on reflected light, may have been used by the ancestors of luminescent fishes when photophores and associated signaling behavior were first evolving. Luminescence could have amplified weak reflected light in the mesopelagic zone of perpetual twilight, where presently two-thirds of the piscine inhabitants have light.

Sources: Norman, 1947; Harvey, 1952; Marshall, 1966.

RITUALIZATION—THE ACQUISITION OF BIOLUMINESCENT SIGNALS FROM EMISSIONS OF OTHER CONTEXTS

A number of different situations in which emitters and receivers are involved must be distinguished. Of these (see next section for listing and discussion) I distinguish true communication or communication sensu stricto on the basis of the effects of natural selection: In communication s.s., selection has brought about (enhanced and maintained) the emission and the mechanisms of production and reception in the specific context of information transfer that is being considered (Lloyd, 1971; Otte, 1974). An emission becomes ritualized—that is, adapted into a communicative context from another, stereotyped, and exaggerated—when, upon its detection, both the emitter and receiver benefit from the subsequent action the receiver makes on the basis of the information it derived from the emission. Or, when viewed from another perspective, when the emitter influences (manipulates) subsequent and mutually beneficial actions of the receiver. In the following examples it is speculated that communicative signals have been derived from emissions that were of original significance to the emitters in the context of illumination.

Fireflies of some species emit characteristic luminescence patterns when they land. In North America this is most commonly seen among fireflies of the genus Photuris (Lloyd, 1968), and it occurs in Pteroptyx and Luciola species in New Guinea and probably elsewhere (Lloyd, 1973a). Among Photuris spp. the sole function of the landing emissions seems to be illumination, and the use of such light is observed mainly in females. As a female approaches the ground her flash rate increases until finally the flashes fuse into a glow, which is discontinued only after landing. With practice an observer can quickly learn to recognize landing lights and will seldom confuse them with male advertisement patterns. Females land in areas where dozens of patrolling, advertising males are present, yet female landing flashes elicit no visible response from the males. The females are presumably not receptive to male sexual advances and probably have already mated. Landing flashes and glows could become incorporated into a signaling system if the females were sexually responsive: males that approached landing luminescence and emitted advertising flashes in the vicinity would have improved mating success.

Females of Pteroptyxfireflies, the synchronously flashing species of southeast Asia, emit light as they approach swarm trees and land on the foliage. These females are apparently un- mated and upon entering the swarms find mates. In one species, the 3-flicker pteroptyx (no. 22, Lloyd, 1973a: 1003), males sometimes pursue females in aerial chases over the foliage, bump or upset the aerodynamics of the females, force them to land, and land near them. A simple explanation for the origin of this behavior is that males that followed and landed near luminescing females increased their chances of mating with them.

Although the frequency of chases observed in the Pteroptyx species studied so far does not indicate that an aerial chase has become a ritualized and invariable part of their courtship, it has become so in the courtship of a species in a related genus. In another New Guinea firefly, the diamondback luciola (Luciola obsoleta), males and females perch in loose congregations, and each emits a variety of sexually distinct luminescent patterns. In late evening, about two hours after sunset and the onset of flashing activity, when a female takes flight she is pursued by flickering males. After traveling a few meters the female lands or is forced to land by the darting and bumping tactics of a pursuing male, which lands close by. Courtship then continues, additional luminescent signals are exchanged, and finally mounting and copulation occur.

Another possible example of ritualization of a noncommunicative emission concerns landing flickers of males of Pteroptyx fireflies. These flickers presumably function in illumination. Perched males of the 3-flicker pteroptyx begin emitting flickers when glowing females fly over them. The flicker that they emit at this time is different (in phrasing and flicker frequency) from the one used as an advertising pattern. A possible phy- logeny for the evolution of the response flicker from a landing flicker is: (1) Males (of an ancestral population) emitted a flicker while landing. Males emitted the flicker also when landing near females that had answered their advertising flash pattern. (2) The courted females approached the landing males by orienting to their landing flickers. Males, approaching by foot or short hopping flights, continued to use the landing flicker for illumination. (3) Since females oriented to and/ or approached the landing flickers of males, selection favored males that produced this flicker after each female answer, as well as when it was required for illumination during their locomotion, and/or during the time they were in visual, luminescent contact with females. (4) Perched males flickered, with the landing flicker, in response to the luminescence of approaching females with which they had no prior interaction, as now occurs in the 3-flicker pteroptyx.

BIOLUMINESCENT SIGNALS FROM BEHAVIORALLY SIGNIFICANT ENVIRONMENTAL PHOTIC STIMULI

Many organisms have evolved specific, yet poorly understood, positive responses to light. Hence, the success of luminescent, lochetic¹ fungus gnats (Fulton, 1939; Gatenby and Cotton, 1960), the popularity of light trapping among insect fanciers, certain poaching and hunting techniques, and the use of bright lights by marine biologists as well as anglers for attracting specimens and prey. A variety of behavior is undoubtedly involved. Some moths may use celestial light sources for bearings and thus maintain straight flight over some distance; if they take certain bearings on a streetlight they will fly a spiral route into the light. Artificial lights may activate neural circuits and behavior that evolved in the context of surface seeking in aquatic animals, entrance seeking in cave animals, or in relation to dawn and the beginning of flight activity in winged, diurnal organisms. If mates are brought together by mutual attraction to some form of natural illumination, then the advent of luminescence could lead to bioluminescent signaling. Luminescence in such cases might provide a concentrated light stimulus and focus the attraction of one of the pair. With the paucity of behavioral/ecological data on luminescent forms this phenomenon remains in the realm of speculation, and I am unable to find a suggestive example or to postulate one that is superior to explanations along other lines. The surface swarming of luminescent syllid worms such as the Bermuda fireworm might have originated in this manner, but the luminescent enhancement of a previously existing signal (say bright and dark bands of surface ripples as seen from below) seems more likely at present. Phototactic responses of shed gametes might predispose an organism to develop luminescent signals in this context.

Classification of Bioluminescent Emissions and Interactions

The term "communication" has been used loosely to include a number of different kinds of interactions (Sebeok, 1968; Otte, 1974). Still other relationships that organisms have with components of their environments, that no one would consider communication, involve similar or identical sensory, neural, and behavioral attributes and have importance in the evolution of communication. I believe that it is worthwhile to list these phenomena and try to distinguish precisely among them. A classification that focuses on adaptation and that is based on the action of natural selection is useful and relevant. Otte (1974) used this approach in a discussion of "communicative" interactions including mimicry, deceit, and intra- and interspecific signaling. For each interaction he considered the effects on each participant 's survival or reproductive success, and scored it positive (+), negative (-), or no effect (0). Others (e.g., Odum, 1959:226) had previously used this sort of notation when considering various kinds of ecological interaction among organisms, but had focused, with an evolutionarily aloof perspective, on the consequences for population size. In the classification presented here primary attention is paid to the long-term effects of selection on the emitting attributes (and emission ) of the emitter, and the reception attributes of the receiver, in the specific context under consideration. Attributes include sensory and effector mechanisms with their underlying neural circuits of emission analysis and production. The effects on individuals considered by Otte result in the population changes discussed by Odum and the long-term changes in the emissive-perceptive attributes of concern in this discussion; the mundane interactions of individual organisms determine differential genetic survival and long-term evolutionary changes.

I SIGNALS-COMMUNICATION SENSU STTUCTO (+/+)

The single category of interaction with which most researches working on communication are concerned involves a transmitter and a reciever in separate individual. and natural selection acts on the mechanisms of both to facilitate and enhance the transmission-reception in the given context (Table 2). This is the category for which I reserve the term "communication" (Lloyd, 1971), since it best agrees with previous expressions of selection-conscious workers and vernacular usage.

Bioluminescent examples of this category would include mating communication of fireflies (Lloyd, 1971), cuttlefish (Sepia: Girod, 1882), octopods (Tremoctopus: Lane, 1960), fish (Nicol, 1969:391), and fireworms (Odontosyllis enolpa: Harvey, 1952); shoaling in fish and crustaceans (Nicol, 1969:391); gamete orientation and attraction in sedentary or sessile marine organisms (hypothetical, occurrence unknown); emissions that are involved in interactions of mutual benefit between members of different species such as fungi and insects (discussed above); and warning lights in fireflies (undemonstrated) (Lloyd, 1971) and poisonous dinoflagellates (hypothetical).

II. SELF-SIGNALS—ILLUMINATION (+/+)

This category differs from the one above in that the transmitter and receiver mechanisms, in a given emission pathway, are in the same individual. Autocommunication can at least tentatively be distinguished from other intra- individual information-transferring phenomena, such as hormones and neural feedback, by the occurrence of the informationally significant alterations of the carrier energy during its passage between emitter and receiver mechanisms. Taxa for which illumination lights, the biolu- minescent analogues of echolocation in bats and active electroreception in fish (Bullock, 1973), have been suggested are fireflies (Waller, 1685; Lloyd, 1968), squids (Lane, 1960:72, 113), and fish (Harvey, 1952:523).

III. FALSE SIGNALS—AGGRESSIVE MIMICRY (+/-)

The exploitation of a receiver in which an emission activates mechanisms that evolved and/or are maintained in the context of true communication is but one of several kinds of interactions in which the receiver is exploited. A false signal possesses those properties, and is presented in those circumstances that enable it to be sensed, neurally processed, and responded to in a manner appropriate to true signals. Selection acts upon the mechanisms of reception to promote the discrimination of false signals from true ones.

Female fireflies of the genus Photuris mimic the mating signals of females in the genera Photinus, Pyractomena, Photuris, and Robopus, attract the males of these species, seize them, and devour them (Lloyd, 1965; Farnworth 1973). The females of some species are able to mimic the flashed responses of more than one prey species, and individual females switch appropriately from one response to another, depending on the characteristics of flash pattern they are answering. Some males of the prey species respond to the false signals in the same manner that they would to true signals from their own females, and are caught (Lloyd, 1975).

If the luminescent lures of female angler fish are also used in courtship for signaling to their own tiny males (Harvey, 1952:529), if there is great competition among males for mates, and if species with similar signaling systems occur together, the occurrence of aggressive mimics among them would not be surprising. However, the attraction of other prey to the lure would not involve false signals, but false clues as discussed below.

IV. CUES—EMANATIONS OF INDIFFERENT EMITTERS (0/+)

Organisms perceive and process stimuli from sources that are themselves in no way affected by the outcome of the detection. Whether the emanation is detected is irrelevant to the emitter, and the emitter may even be inanimate (a cow pie that is used for food by scarab beetles), dead (a deceased cat that attracts staphylinid beetles that prey upon carrion-feeding larvae), or living (a leafy branch in a shaft of sunlight to which mat- ing-swarms of insects are orienting). The biological significance of cues to the receiver may be that they are, after detection and neural processing, translated as "of no significance (now)." In other words, they can be disregarded. For example, a shore bird lands beside a log that emits stimuli that are translated as "a neutral object, a possible perch, of no negative value ('danger to me')." The detection by the receiver organism is of biological significance to the receiver in that context, and selection maintains and enhances the reception mechanisms. It would be unadaptive for a shore bird to respond to all such logs in the same manner as to a crocodile!

Pteroptyx fireflies of southeast Asia gather in trees, sometimes in great numbers. Males emit their flashes in synchrony with nearby males, and as a consequence, when there are enough males so that continuity is maintained over an entire tree, mass synchrony occurs. The flash rate is characteristic of a species, and emerging adults in the vicinity fly to trees that are pulsing with the appropriate characteristics; a pulsing tree is like a beacon. As flying fireflies approach a firefly- tree from a distance they will at first see only the entire tree. At some point during their approach their compound eyes will begin to resolve parts of the tree, then small clusters of flashing males, and finally individual males. Only when individual males, each with its own "halo" of neighbors with which it is interacting, are resolved by incoming females, are they finally in competition for them. Selection is not maintaining the beacon effect in the beacon context; selection produces group synchrony because it favors individuals that synchronize with their near neighbors. Mass synchrony is a consequence (effect) of natural selection and not a goal (in the sense of Williams, 1966:9; Lloyd, 1973a, 1973c). The emission of a treeful of fireflies is therefore a cue and not a signal. Each individual male, within his own halo, is signaling. (A useful discussion of underlying genetic concepts and arguments may be found in Williams's [1966] extremely valuable little book.)

TABLE 2

Emission-reception phenomena.

NOTE: Emitters are living, dead, or inanimate; receivers are living organisms. Selection signs indicate positive (+), negative (-), or no effect (0) on the mechanisms in the specific context being considered (see text). "Em" indicates emission mechanisms and "Rm" receiver mechanisms, in the sense of Williams (1966:9) when applicable; otherwise they indicate emission and receiver mode.

V. FALSE CUES—MIMICRY OF CUES (+/-)

The exploitation of a receiver in which an emission activates mechanisms that evolved and/or are maintained in the context of cuing (Table 1) has, like exploitation by false signals, the plus-minus polarity of selection signs. Selection on the emitter mechanisms favors the enhancement of the deception; selection on the receiver favors the discrimination of false cues from their true cue models. In sequel to the shore bird example, selection will favor those behavioral and morphological traits of crocodiles that promote their adjudication as logs by the reception mechanisms of the birds; and at the same time, in the race of measure-countermeas- ure, selection favors receiver mechanisms that do not make the mistake of computing crocs as a "something nothing."

Hastings (1971) presented suggestive evidence that ventral luminescence of the pony fish (Leiognathus equulus) simulates the shimmery water surface above and makes it difficult for predators below to detect; there are other explanations and considerations (Nicol, 1969: 392). In the special terminology of protective coloration literature (Robinson, 1969:229), this would be a form of eucrypsis, and more specifically homochromy.

Fireflies of some species pupate underground or in dead, rotting logs. They are luminescent and will, upon mechanical stimulation, turn on their lights. These lights may protect the fireflies by activating, in the potential predators that come upon them, the avoidance responses that normally keep the predators from moving into daylight (where desiccation or their own predators might overcome them; Lloyd, 1973d).

Beebe's (1926) description of prey capture by myctophid fish suggests a false-cue context, with the luminescence perhaps mimicking surface illumination: "Five separate times when I got fish quiet and wonted to a large aquarium, I saw good-sized copepods and other creatures come within range of the ventral light, then turn and swim close to the fish, whereupon the fish twisted around and seized several of the small beings."

VI. CLUES—EMISSIONS OF BETRAYAL (-/+)

Virtually every living thing is exploited by some other organism. Exploiters perceive and use emanations of their victims, at the least to make contact with them. It seems reasonable to use the term "clue" for such emissions since in detective fiction parlance a clue is evidence that leads to the undoing of the individual responsible for the clue's production. An insectivorous bird detects and computes visual properties of a caterpillar, then devours the insect. The cylindrical form itself is an important clue to the detection, as evidenced by the various ways it is disguised by protective coloration (Cott, 1957). On the other hand, emissions of predators, such as the circular eye form (Cott, 1957), that reveal their presence to quarry are also clues.

In the early evolution of flash communication in fireflies one selective agent that brought about the shortening of long glows, the simplest emission, into short pulses, could have been a visually orienting predator that approached perched, glowing individuals. Today wolf spiders prey upon fireflies and seem to orient to their lights (Lloyd, 1973b).

VII. FALSE CLUES—BAIT, BOGUS BONANZAS, ENTRAPMENT (+/-)

The exploitation of a receiver in which an emission activates mechanisms that evolved and/or are maintained in the context of cluing (Table 1) has, like exploitation by false signals and false cues, the plus-minus polarity of selection signs. False clues mimic clues. Selection on emitters favors enhancement of the deception, and on receivers favors its discrimination.

The luminescent lures of ceratoid and stoma- toid fish may emit false clues by mimicking worms or other prey (Marshall, 1966:176, 303; Harvey, 1952:529). The autotomy of luminous segments by marine annelids (Polynoe: Harvey, 1952:208) seems to involve the use of false clues. Haswell's quote (Harvey, 1952:209) says it all:

When certain species of Polynoe are irritated in the dark a flash of [bioluminescent] light runs along the scales, each being illuminated with a vividness which makes it shine out like a shield of light, a dark spot near the centre representing the surface attachment where the light-producing tissue would appear to be absent. The irritation communicates itself from segment to segment, and if the stimulus be sufficiently powerful, flashes of [bioluminescence] may run along the whole series of elytra, one or more of which then become detached, the animal meanwhile moving away rapidly and leaving behind it the scale or scales still glowing with [bioluminescent] light. The species in which the phenomenon of [bioluminescence] occurs are species characterized by the rapidity of their movements, and also by the readiness with which the scales are parted with; and it seems not at all unlikely that the [bioluminescence] may have a protective action, the illuminated scales which are thrown off distracting the attention of the assailant in the dark recesses which the Polynoidae usually frequent.

VIII. NOISE—HINDRANCES FROM INDIFFERENT EMITTERS (0/-)

Physical disturbances in the environment often interfere with the ability of receivers to receive and process the significant energy components that make up such phenomena as signals, cues, and clues. In the simplest meaning of "noise," the physical disturbance is in the same channel (sound, light, etc.) as the masked significant energy, and the noise jams the sensors by producing a background of some intensity.

Bright moonlight, as it shines on the vegetation where firefly females are perched, is a hindrance to males' perceiving luminescent responses of females. It probably also interferes with females' ability to detect flashes of males against the night sky. Selection has certainly resulted in adaptations such as filters, shades, or screens that reduce the reception of moonlight— that improve signal-to-noise ratios. Selection has brought about and maintains behavioral adaptations that result in the elimination of the noisy sun and moon—luminescent fireflies do their signaling after sunset and are far less active on moonlit nights.

IX. FALSE NOISE—EMITTERS WITH VESTED INTEREST (+/-)

False noise is emitted by living organisms that exploit or somehow place the receiver at a disadvantage and benefit themselves. It contrasts with noise in that the latter may be emitted by living, nonliving, inanimate, or dead emitters— in the case of noise, no advantage can or does accrue to the emitter. With false noise the emission interferes with the functioning of the receiver because it, like noise, produces a background energy level that disrupts the reception of important (to the receiver) energy phenomena.

Upon tactile stimulation the squid Heteroteu- this dispar discharges masses of mucus that become brilliantly luminescent. It has been suggested that this discharge may "baffle" pursuers and that it would be "disconcerting" to predators (Lane, 1960:107, 112). The luminous discharge may function in different modes of predator defense. If the secretion maintains cohesion and is mistaken for the body of the squid, while the squid itself escapes, we have a false clue. If the luminous cloud prevents the squid's detection because the resulting increased background light level exceeds the intensity of light clues (luminescent or reflected) that the predator would use for attack orientation, the luminous discharge is a false noise.

X. AMBIGUOUS SIGNALS—MISTAKEN IDENTITY WITH MUTUAL DETRIMENT (-/-)

In this interaction the receiver is negatively affected because the emission that is received mistakenly activates mechanisms that belong to another signaling context or species. In the subsequent interactions the emitter is also harmed. An important class of this category results in interspecific mating. Selection can be expected to improve mating-signal discrimination as well as to intensify existing but minor signal differences between species living in the same area and active at the same time. Such interactions commonly occur when sibling species² come into contact following (extrinsic) isolation that has led to their speciation and probably had an important role in the development of the timing differences found in mating signals of several species of fireflies (Fig. 1).

XI. INDIFFERENT FAZERS—DYSFUNCTION CAUSED BY INDIFFERENT EMITTERS (0/-)

The received energy of an emission, because of its intensity or chemical properties, may cause temporary or permanent impairment of function of receiver mechanisms. The response is not an evolved one—i.e., it is pathological. If the interaction is of no consequence to the emitter, selection will act only on the receiver. Bright rays of a flashlight will occasionally cause a flying firefly to become disoriented and spiral to the ground, temporarily out of competition for a mate. If an average male lives for but two evening flights, each thirty minutes long, and he loses ten minutes of flight time, he realizes an appreciable loss. Upon crashing to the ground, if he gets caught by a predator or trapped in a pool of water or in a spider web, his loss is even greater. Lightning flashes may have once brought about similar dysfunction, and selection could already have resulted in, and presently maintain, protective mechanisms for dealing with brief, high-intensity light bursts.

XII. FAZERS—EXPLOITATIVE DYSFUNCTION (+/-)

The effect of this emission on the receiver is the same as in the above category, but the emitter is not indifferent to the effects upon the receiver—it causes them and capitalizes on them.

Fig. 1. Firefly signals. Fireflies within each section are closely related. Their signals have diverged since speciation, probably in many cases because of inappropriate matings. Horizontal axis = time; vertical lines mark one-second intervals. Black symbols at left are male advertising flashes. Numbers indicate approximate length of intervals between consecutive male advertisements (in sec). Common names have been used for species not having latin binomens.

P. consanguineus group (eastern U.S.): The male pattern in three species is composed of two flashes, and the differences in their timing is critical and prevents interbreeding. Female answers occur 1 sec (black triangles) after the second flash of these patterns. Two other species in the group have single flashes and long female delays. The latter species may have recently lost the second flash. If one presumes the 1-sec female delay to have been derived from a common ancestor (it occurs in four of the six studied species of the group), the position of the "lost" male flash (hollow symbols with question mark) can be inferred on the basis of the timing of the female answers. The hypothetical ancestral signals are very much like signals of extant species. Unique codes among the species of this group may have been achieved by timing changes and flash losses.

P. ardens group (northeastern U.S.): Male advertisement patterns vary within individuals as indicated by solid and lined symbols. Note the mathematical relationship between the two species with respect to flash length, number, and period, and advertisement period. Since speciation have these parameters changed independently, or together as some basic component of their nervous systems was altered?

P. pennsylvanica group (northeastern U.S.): The timing of the two patterns is similar but flash duration and relative intensity differ.

Black luciola group (New Guinea): The flicker frequency of the little black luciola is about half that of its relative. The fireflies of the L. peculiaris group (New Guinea) emit continuous (c) trains of flashes. Males of the first two species give this pattern as they fly in search of mates, and they are attracted to penlight flashes that occur immediately after each of their flashes. The feeble nature of alternate flashes in L. huonensis suggests that these are being lost and that originally the timing was more like that of L. peculiaris. The flashes of L. obsoleta have a similar period, but they are even more interesting from another standpoint: L. obsoleta's signaling system is completely different, and these flashes do not appear in courtship until the male has reached the female and mounted her. The stages of courtship that precede mounting include aggregation of many individuals, sedentariness with flickering and flashing unlike the mounting (c_m) flashing pattern, aerial chases, and finally an intimate walking-luminescing interaction. If the c_m pattern is homologous with the flash trains of its relatives, how and whence have the complex preliminaries of L. obsoleta gotten into the act?

P. pyralis group (Texas): The structure of the flashes of males and females of both species are similar; it is the female response delay that prevents interbreeding. * Hollow symbols marked m indicate male advertisement flashes; solid symbols are the female answers.

The ventral surface of the dogfish shark (Spinax niger) is covered with light organs. C. F. Hick- ling (1928, in Harvey, 1952:498) made the following observations and speculations, the latter of which place this predatory "tactic" here:

When a luminescent specimen is held so that one's line of vision is perpendicular to the ventral surface of the fish, the luminescence is plainly visible. When the fish is then rotated slightly to left or right about its long axis, the light disappears. This observation seems to offer some explanation of the function of the luminescence of Spinax.

The complex lantern-like structure of each individual organ seems designed to throw out a parallel beam of light, and to prevent scattering of the rays; the arrangement of the axes of all the organs parallel to the median vertical axis of the fish, seems to aim at precisely the effect described above, namely, that the luminescence will only shine upon objects immediately beneath the ventral surface.

The mouth of Spinax is situated remarkably far behind the tip of the snout, so that Spinax can obviously only seize objects immediately beneath (in the relative sense) its mouth. But it is only when an object is immediately below the ventral surface of the fish that the light from the luminous organs flashes fully upon it. One may therefore suggest that the sudden flash of light, at the moment of attack, may cause the prey of Spinax to hesitate for just that fraction of a second in which the mouth can make a successful snatch.

XIII, XIV. SCANNERS AND LOCATORS—SELF- SIGNALS ONE WAY, CLUE PRECURSORS (+/0,+/-)

In the self-signal category (II above) the emitter and receiver mechanisms were in the same individual. The two categories considered here are identical with the self-signal during the outward leg of its propagation. In many situations the objects that are impinged upon are not really receivers—they have no receiver mechanisms and they merely intercept passing emissions (scanners, +/0). The locator (+/-) emission is a scanner that has been incorporated into a prey- tracking system and is of significance to both emitter and "receiver." It is important to consider these two categories, even though they do not involve true receivers, because of their evolutionary relationship to emissions and interactions that are of interest. A scanner becomes a locator—bats used their cries for orientation before they used them for tracking prey—and the selective pressure that a locator exerts on prey can lead to the development of receiver mechanisms in the prey. For example, a portion of the moth body that originally reflected the locator back to the bat ultimately became a detector (tympanum) in the moth. When the locator emission began to be detected and used by members of moth populations, it ceased to be solely a locator, but part of it also became a clue and balanced the selection signs—i.e., locator +/- versus clue -/+—of the bat-moth interaction.

It is common among some marine organisms such as squids, deep-sea fishes, and shrimps (Eu- phausiidae) to have photophores around, on, or in the eyes, and it has been suggested that these function in illumination. The eye lanterns of some squids and fishes shine into the eyes (Harvey, 1952:287; Lane, 1960:73). It would be interesting to know if the latter actually shine upon photosensitive tissue, or if they are instead reflected from a tapetum that somehow aims or focuses the light rays and permits accurate range or direction-finding of nearby prey.

XV-XVIII. QUASI EMISSIONS AND CIPHERS (0/-; 0/0)

Additional categories are theoretically plausible, if not always empirically identifiable. Quasi emissions are of no significance to emitters (0/-) in the contexts being focused upon. They inappropriately activate receiver mechanisms that belong to a diversity of contexts. Stray emissions could be mistaken for signals, clues, or cues (Table 1), and receivers could react to them as though they were the real thing.

For example, I once observed a male Photuris congener in the grasp of a large lycosid spider. While the spider devoured him, he continued to emit a flashing pattern that resembled that of congener females, and he attracted two more males. When the decoyed males arrived the spider seized them and subsequently ate them too. The consequences of the emission were irrelevant to the first captive and unfortunate for the respondents that mistakenly accepted quasi signals as signals. As a second example, consider the plight of the tiny male of the reticulate firefly (Phausis reticulata) that was attracted to the glow of a single lantern of the gigantic, relatively speaking, larviform female of the plumose glowworm beetle (Phengodes plumosa: McDermott, 1958:15). The activated neural circuits of both individuals in this case were appropriate to a signal context. The male lost time and energy, but apparently the female, hardly more than tickled by the attentive male, merely continued to glow and advertise, unaffected—another example of a quasi signal (0/-).

COMBINATIONS

An emitter has simultaneous interactions of different kinds with other organisms. The flash of a female firefly given in response to the flash pattern of a male of her species is a signal, in that context; to a female perched below in dense vegetation that mistakenly flashes an answer it is a quasi signal; but if selection has favored this sort of response in females, because in a percentage of such circumstances they get an opportunity to get a mate that they otherwise would not have seen or begun courtship with, it is a clue. To an immigrating female firefly that, as a consequence of seeing the interaction, identified the locale as a potential oviposition site, it would be a cue.

Fig. 2. Lycosa rabida eating Photuris sp. Lycosid spiders are important predators of fireflies, and evidence suggests that they seek grounded luminescing individuals. Predators behaving in this manner could have brought about the evolution of the flash from the glow. The flashes of the firefly in the context in which the spider was the receiver were clues; the firefly's "intended" receiver may have been itself (self-sig- nals), a mate (signals), or prey (false signals).

Fig. 3. Firefly melee. A grappling, struggling group of four fireflies, including two agressive mimic females and two prey males. Probably one female attracted both males (false signals) and the second female observed the interaction and attacked (clue?).

The wolf spider (Lycosa rabida) in Fig. 2 is grasping and sucking on a Photuris firefly that it probably located by cluing on the firefly's illumination self-signals, predatory false signals, or mating signals. The bunch of 4 fireflies in Fig. 3 was composed of two Photuris males of one species and a female of each of two other Photuris species. I found this group by investigating bright flares emitted by one of the males—the flares are characteristic emissions (function unknown) of these males when they are seized by predators. I believe that one of the females attracted both males by answering them with false signals, and that the other female observed the combination of emissions of a male and the female and launched her attack.

Male fireflies recognize courtship interactions taking place between other individuals (Lloyd, 1969, 1973c) and probably use the information for exploitation. In such cases there are two emitters, and both are essential for the receiver to obtain the information. In the case of sexual interloping, it is the female response to the other male that gives the interloper the essential information for identifying what is happening; the male pattern that had been observed was by itself not significant and was stored information at the time the female flash was emitted. With respect to the voyeur male, to what categories do the emissions of the courting pair belong? The courting male's pattern, when received by the voyeur, is a clue when the outcome of this is competition that diminishes his chances to copulate with the female. In some species selection has resulted in reducing this clue, and courting males greatly reduce the intensity of their mating flashes when they approach answering females. The category or identity of the flash of the courted female that reaches the voyeur male is unknown since its significance to the female is not understood. In attracting a second male either she might get the opportunity to select the "better" male (i.e., the one that would genetically bestow upon her sons the better attributes with respect to mate competition) or she might decrease her chances of getting either male because of her longer exposure to predators, getting knocked from her perch, or losing contact with both males, as a consequence of their fighting after both reached her. Until the statistical probabilities of these outcomes are known the interactions cannot be classified.

Similar uncertainties are involved in cases of the flare flashes of males captured by aggressive mimic females or spiders. If in a significant percentage of the encounters these flashes attract additional predators (and in a predator melee the firefly sometimes escapes), the selection signs for the prey-second-predator emissive- receptive interaction will be different from what it would be if the flare flash were simply a consequence of a poison that has a pathological effect on flash physiology.

Interacting organisms can have different interactions with each other simultaneously. The echo-locating cry of a bat is a locator when it strikes the body of the moth and the portion of the bat's emission that strikes the moth tympanum provides a clue for the moth.

EVOLUTIONARY RELATIONSHIPS

Quite obviously the interactions observed between organisms today have not always existed, but got that way via various evolutionary routes. Through mutual adaptation and unilateral exploitation and escape, interactions of one kind have passed, through time, into others. The fazer of the dogfish, before it became so intensely bright or well aimed, may have functioned in illumination of the sea-bottom (self-signal), or attracted prey (false cue?), or, like the light of the pony fish, have been used for hiding (false cue). It has perhaps also become involved in signaling systems. The false signals of Photuris females were probably derived from at least two sources, self-signals and their own sexual signals. An expected evolutionary source of false clues, such as those the self-fragmenting polynoid worm scatters in its wake when attacked, would be the clues their ancestors had emitted, to their detriment, in earlier times. Certain signals that fireflies use in courtship may have been adapted from illumination luminescence (self-signals), but the origin of firefly luminescent signaling, if the universality of larval luminescence among the lampyrids can be used as an indication, is obscured by our ignorance of the present function of larval lights. The most comprehensive statement that can be made about the origins and functions of biolumines- cent emissions is that nothing is known of most of them, little is known of a few of them, and we have not learned of the best of them.

References

Beebe, W., 1926. The Arcturus Adventure. New York: Putnam. 290pp.

Brown, F. A., Jr., 1973. Bioluminescence. In: Comparative Animal Physiology, 3d ed., C. L. Prosser, ed. Philadelphia: W. B. Saunders, p.965.

Bullock, H. T., 1973. Seeing the world through a new sense: electroreception in fish. Amer. Sei., 61:316-25.

Cott, H. B., 1957. Adaptive Coloration in Animals. London: Methuen. 508pp.

Ewan, J., and Ewan, N., 1970. John Banister and His Natural History of Virginia 1678-1692. Urbana: University of Illinois Press. 487pp.

Farnworth, E. G., 1973. Flashing behavior, ecology and systematics ofjamaican lampyrid fireflies. Ph.D. diss. University of Florida.

Fulton, B. B., 1939. Lochetic luminous dipterous larvae. J. Elisha Mitchell Sei. Soc., 55:289-93.

Gatenby, J. B., and Cotton, S., 1960. Snare building and pupation in Bolitophila luminosa. Trans. Roy. Soc. New Zealand, 88:149-56.

Girod, P., 1882. Recherches sur la poche du noir des cephalopodes des cotes de France. Arch. Zool. exp. gen., 10:1-100.

Harvey, E. N., 1952. Bioluminescence. New York: Academic Press. 649pp.

Hastings, J. W., 1971. Light to hide by: ventral luminescence to camouflage the silhouette. Science, 173:1015-17.

Lane, F. W., 1960. Kingdom of the Octopus. New York: Sheridan House. 299pp.

Lloyd, J. E., 1965. Aggressive mimicry in Photuris: firefly femmes fatales. Science, 149:653-54.

Lloyd, J. E., 1966. Studies on the flash communication system in Photinus fireflies. Univ. Mich. Mus. Zool. Misc. Pub., no. 130. 95pp.

Lloyd, J. E., 1968. Illumination, another function of firefly flashes. Entomol. News, 79:265-68.

Lloyd, J. E., 1969. Flashes, behavior, and additional species of nearctic Photinus fireflies (Coleoptera: Lampyridae). Coleop. Bull., 23:29-40.

Lloyd, J. E., 1971. Bioluminescent communication in insects. Ann. Review Entomol., 16:97-122.

Lloyd, J. E., 1972. Chemical communication in fireflies. Environ. Entomol., 1:265-66.

Lloyd, J. E., 1973a. Fireflies of Melanesia: bioluminescence, mating behavior and synchronous flashing. Environ. Entomol., 2:991-1008.

Lloyd, J. E., 1973b. Firefly parasites and predators. Coleop. Bull., 27:91-106.

Lloyd, J. E., 1973c. A model for the mating protocol of synchronously flashing fireflies. Nature, 245:268-70.

Lloyd, J. E., 1973d. Fireflies—commonplace beetles and larvae by day but tiny flashing lanterns on summer evenings. Animals, 15:220-25.

Lloyd, J. E., 1974. Bioluminescent communication between fungi and insects. Florida Entomol., 57:90.

Lloyd, J. E., 1975. Aggressive mimicry in Photuris fireflies: signal repertoires by femmes fatales. Science, 187:452-53.

McDermott, F. A., 1958. The fireflies of Delaware. Soc. Nat. Hist. Delaware. 36pp.

McElroy, W. D., and Seliger, H. H., 1962. Biological luminescence. Sei. Amer., Dec., 76-89.

Marshall, N. B., 1966. The Life of Fishes. New York: World Publishing Co. 402pp.

Nicol, J. A. C., 1969. Bioluminescence. In: Fish Physiology, W. S. Hoar, and D.J. Randall, eds. New York: Academic Press, pp.355-99.

Norman, J. R., 1947. A History of Fishes. London: Ernest Benn, Ltd. 463pp.

Odum, E. P., 1959. Fundamentals of Ecology, 2d ed. Philadelphia: W. B. Saunders Co. 546pp.

Otte, D., 1974. Effects and functions in the evolution of signaling systems. Ann. Rev. Systematics and Ecology, 5:385-417.

Rivers, J. J., 1886. Description of the form of the female lampyrid (Zarhipis riversi Horn). Amer. Natur., 20:648-50.

Robinson, M. H., 1969. Defenses against visually hunting predators. In: Evolutionary Biology, 3, T. Dobzhansky, M. Hecht, and W. Steere, eds. New York: Appleton-Century-Crofts. 309pp.

Sebeok, T. A., ed., 1968. Animal Communication. Bloom- ington: Indiana University Press. 686pp.

Seliger, H. H., 1975. The origin of bioluminescence. Photochemistry and Photobiology, 21:355-61.

Shawver, L. J., 1974. Trilobite eyes: an impressive feat of early evolution. Sei. News, 105:72-73.

Tiemann, D. L., 1967. Observations on the natural history of the western banded glowworm Zarhipis integripennis (LeConte). Proc. California Acad. Sei., 35:235-64.

Waller, R., 1685. Observations of the Cicindela volans, or flying glowworm. Phil. Trans. Roy. Soc., 15:841-45.

Wickler, W., 1968. Mimicry in Plants and Animals. New York: McGraw-Hill. 155pp.

Williams, G. C., 1966. Adaption and Natural Selection. Princeton: Princeton University Press. 307pp