Chapter 14

COMMUNICATION, CRYPSIS, AND MIMICRY AMONG CEPHALOPODS

Martin H. Moynihan and Arcadio F. Rodaniche

Every individual organism, group, and species has its own characteristic distribution in space and time. Distributions are partly determined by behavior. Among the most important behavior patterns of animals are signals that transmit information (true or false) from one individual to one or more others (of the same or different species). Almost any act can be a signal while subserving other functions as well, but there are certain particular patterns that have become specialized, modified in form or frequency, expressly and only to facilitate the transmission of information. These are usually called "displays." They are related to, and may interact with, equally and similarly specialized patterns designed to prevent or confuse communication. These might be called "antidisplays." Some patterns can be displays in some situations, antidisplays in others. In either or both cases, the kinds of specialization involved are usually called "ritualization. "

Cephalopods have many ritualized and unritualized patterns, arranged in adaptive sequences and repertoires.

The living cephalopods can be divided between two major systematic groups: One comprises a few species of Nautilus; the other includes the coleoids—many species of many genera of three or four orders—the squids, cuttlefishes, Vampyroteuthis, and octopi and argonauts. All are marine. Very little is known of the social behavior of Nautilus or of those coleoids that are confined to deep waters or open seas, but there have been fairly extensive studies of inshore and littoral forms, most notably species of Sepia, Octopus, Sepioteuthis, and Loligo s. 1. (references in Moynihan, 1975). These animals not only share a common remote ancestry but also show similarities in ways of life (see Lane, 1957; Packard, 1972; Wells, 1962).

They are large (considering the animal kingdom as a whole), more or less active, comparatively intelligent, with large, complex brains (Young, 1964, 1971), predaceous upon smaller organisms and preyed upon by larger ones. They tend to have rapid rates of development. Most squids are conspicuously gregarious. Other species have social liens of less conspicuous but not necessarily simpler types. Many are extremely abundant. All seem to have elaborate "courtship" and other sexual or partly sexual reactions. Some are supposed to reproduce only once in a lifetime, in "big bangs" (Gadgil and Bossert, 1970). All make special provisions for their large eggs. They may "incubate" the eggs and/or lay them in selected sites and conditions. They seldom or never show strictly parental protection or training of hatched young, but adults and immatures may associate with one another according to regular rules. Many species live in rich and varied environments of great diversity and appreciable instability or unpredictability. Some have important interspecific relations in addition to or apart from ordinary run-of-the-mill predator-prey responses. All have superb eyesight, tactile sense organs, some chemical sense(s), and statocysts that could be, but probably are not, used for hearing.

These are the rather broad constraints within which their communication systems have had to operate.

Visual signals are predominant. They include movements and postures and changes in colors and color patterns. The great majority of cephalopods differ from the most nearly comparable vertebrates, the fishes, in having more independently moveable appendages—eight or ten arms and tentacles in addition to fins. Inshore and littoral species also have several kinds of chromatophores plus leucophores and iridocytes. These permit partial or complete color changes of remarkable speed and precision.

The only species that has been studied at length in both the field and laboratory is Sepioteuthis sepioidea (Boycott, 1965; Moynihan and Rodaniche, in prep.). It may be typical of many inshore cephalopods in some aspects of its communication behavior. Its unritualized signals include a host of minor movements and intention movements of a few or all of the arms, the fins, and/or the body as a whole. Its ritualized patterns include more exaggerated or stereotyped movements of the same parts, e.g., spreading, curling, raising, or lowering of the arms, and many color changes—including general lightening or darkening, flushes of yellow or lavender pink, and production of more or less broad transverse bars, longitudinal stripes, spots of differing sizes, iridescent ocelli, irregular blotches, and even semigeometric arrangements of a variety of tones and hues. Some of the more spectacular ritualized patterns of this and other species are shown in the accompanying illustrations.¹

There are implications to be drawn from such behavior:

(1) The total numbers of "basic" or "major" ritualized patterns in the repertoires of many cephalopods seems to be of the same order of magnitude as in most bony fishes, birds, and mammals (approximately fifteen to thirty-five by rough definition). The reasons why kinds of display cannot proliferate endlessly without corresponding losses are discussed in Moynihan (1970). It is not surprising that cephalopods conform to the same general rules as other animals. Some or all of the inshore squids, cuttlefishes, and octopi are fortunate, however, in that the refinement of their chromatophores and associated organs permits unusual flexibility of combinations. A squid in the midst of a group, for instance, can transmit at least three or four different "messages" (in the sense of Smith, 1965), absolutely simultaneously, to completely different individuals and in different directions by assuming different color patterns on different parts of the body. It can also change any or all of the patterns and messages instantaneously whenever necessary or desirable. This may be the most flexible of ritualized visual systems. Perhaps only the most complex acoustic repertoires of certain birds and primates ("Songs," etc.) convey so much information so rapidly.

(2) Every species of cephalopod has some unique characteristics; but a few of the most elaborate and highly ritualized patterns (see legends of the figures) occur in essentially the same forms in all the coleoids so far studied in detail,including members of orders that diverged in the Mesozoic Era (probably in the Triassic). These patterns would seem to have been extremely conservative during the course of evolution. Some of them may have been conservative because they were designed to influence a great variety of "receivers," individuals of many different species and/or different ages, social classes, and sexes of the same species. Signals adapted to many kinds of receiver would be expected to change less frequently or more slowly on the average than signals adapted to only one or a few kind(s) of receiver (Moynihan, 1975)

(3) Another distinctive and very widespread feature of the ritualized systems of inshore cephalopods is a subtle and problematical phenomenon, an apparent relation between overt behavior and the background against which it is performed. All ritualized patterns of all animals may be considered to be expressions of "drives." The term "drive" itself is loose, and open to criticism when used in analytical (physiological) studies of causation. It is convenient, however, as descriptive shorthand for preliminary or superficial accounts, if it is employed in the sense of Thorpe (1951) as "the complex of internal and external states and stimuli (usually or normally) leading to a given behaviour." The role of the external factors usually appears to be largely or completely indirect, the external situation merely stimulating or affecting internal factors that are the immediate "triggers" of the resulting performance. In most cases and for many purposes, it is sufficient for an observer to state that pattern A is produced by a certain strength or range of internal motivation B, or a combination of B with internal motivation C. Whenever the external situation is such as to stimulate B and C appropriately, pattern A is bound to appear. Only A appears, not D or E or anything else and irrespective of the minor details of the external situation. This sort of simplified schema may also be applied to the ritualized patterns of cephalopods, but it does not always "fit" particularly well. Most cephalopods appear to "choose" among available patterns more frequently than do most other animals. This probably means that they sometimes have to pay more conscious attention to a greater variety of aspects of their surroundings.

Examples from Sepioteuthis sepioidea may show what we mean.

Individuals of this species have many alarm patterns. Two of the most common are bold Transverse Bars and Longitudinal Streaks (Figs. 1-4 and 7). Both are produced when the escape tendency or drive is more or less strong but impeded by—in conflict with—a counteracting and incompatible tendency, e.g., attack, hunger (feeding drive), or a gregarious or sexual attraction. Rather surprisingly, the internal factors involved seem to be identical in some performances of either type. Some individuals in Bar seem to be no more and no less likely to escape than are some individuals (of the same age and, presumably, sex) in Streak. The crucial causal difference between the two performances would appear, at times, to depend entirely on external circumstances. Bar patterns are assumed most frequently by individuals high up in open water. Streak patterns are relatively most common when individuals are low, near vegetation or coral or some other "broken" type of bottom. There are also intermediate situations: when a group of sepioidea is approached by some alarming stimulus in an intermediate habitat, some individuals may assume Bars and others Streaks before they all dash away together.

Fig. 1. An adult Sepioteuthis sepioidea. The animal shows several special color patterns, including Transverse Bars, a Fin Stripe, and a trace of Longitudinal Streak on the back. It has also assumed a Downward Pointing posture.

Fig. 2. A "courting party" of three adult Sepioteuthis sepioidea. The animal on the upper right is a female in Pied coloration. This pattern is "discouraging." It is highly ambivalent, produced by both hostile and sexual tendencies. The female has also assumed a Downward Pointing posture, which in itself is hostile. The animal in the center is a male. He shows Longitudinal Streaks on the back. He must be somewhat alarmed, nervous, and intimidated by the female. The animal on the lower left may be another male. He is in a semi-Dark color pattern, possibly also slightly alarmed

Fig. 3. An Octopus vulgaris with Longitudinal Streaks. (After Cowdry, 1911.)

The partial interchangeability of Bar and Streak may be related to the fact that both patterns are conspicuous in some circumstances and cryptic or mimetic in others. They tend to be equally conspicuous whenever noticed, i.e., when functioning as conventional displays, in a wide range of environments. But their chances of being effectively cryptic or mimetic may be very different in different environments. The elaboration of mechanisms and strategies to permit crypsis and mimicry among cephalopods has been a subject for comment from the time of Aristotle (more recent accounts of interest include Lane, 1957; L. Tinbergen, 1939; Holmes, 1940; Packard and Sanders, 1971; Cousteau and Diolé, 1973). This elaboration may reflect the great vulnerability of cephalopods to predation (Clarke, 1966; Moynihan, 1973, 1975). It certainly has had many consequences for their display behavior.

Fig. 4. Two color patterns of Sepia officinalis. (After Holmes, 1940.) Left: Longitudinal Streaks with Fin Stripes. Right: The so-called Dymantic display (a term applied to patterns that emphasize eyespots—in this and some other cases false eyespots). The Dymantic of Sepia is usually or often combined with dark borders to the fins.

Both Dymantic and Longitudinal Streaks are among the most obviously conservative of cephalopod displays. They occur in the repertoires of some species of at least three different orders: Teuthida, Sepiida, Octopida.

Fig. 5. The Dymantic of Octopus vulgaris. (After Wells, 1962.) Here the emphasis is on the real eyes. The dark borders to the mantle and arms probably are functionally equivalent and historically related to the corresponding features of the fins of Sepia and other forms. Homologies can be indirect and devious within the Cephalopoda.

Another pattern of sepioidea, the Dark (Figs. 2, 7, and 8), is remarkable in several ways. It is sometimes assumed by individuals resting or sleeping a few meters below the surface of the water during the middle of the day. In these circumstances it may, at least conceivably, have nothing to do with communication. It could be simply a means of maximizing heat absorption. (As far as we know, cephalopods are ectothermic. But recent discoveries of endothermic species among classes of vertebrates that were supposed to be "cold-blooded"—Chondrich-thyes and Osteichthyes as well as Reptilia—may add a note of caution and indicate that methods of temperature regulation are varied even in the sea.) An apparently identical Dark pattern can also be assumed by disturbed or frightened individuals. It may seldom occur or never have as high an intensity as the most extreme Bars or Streaks, but its motivation may overlap those of either one or both of the other two patterns at somewhat lower levels of alarm or anxiety.

Fig. 6. A young adult or subadult Sepioteuthis sepioidea . The coloration of this individual is more or less "ordinary" (unritualized) over most of the body, but there are traces of a Transverse Bar across the front part of the back and two well-developed Dymantic spots on either side toward the rear.

Fig. 7. Alarm patterns of very young Sepioteuthis sepioidea. From top to bottom: Arms and tentacles held in an Upward V position, with some indications of Transverse Bars; Transverse Bars and Upward Curling of the arms; Dark coloration with some Bars on the belly and Upward Pointing; Upward Pointing with an extremely broad Fin Stripe, perhaps incorporating some components of Longitudinal Streak; Bars with Fin Stripe and Upward V. (After photographs taken in the San Bias.)

This is a small sample of the possible combinations of signals.

If these suggestions are correct, then Darks can be produced by different factors at different times. The hypothesis is unattractive (too easy—explaining too much or too little), but it is plausible and certainly not logically impossible.

Fig. 8. More alarm patterns, as shown by older Sepioteuthis sepioidea (again after photographs in the field). Left: Full Dark with Downward Curling. Right: Slight Dark with Upward Curling.

Fig. 9. An adult Sepioteuthis sepioidea with a "Rear Light" (a patch of pale color, not luminescent) at the "tail." This may be a low-intensity indication of the Pied pattern.

Darks tend to be rather conspicuous. They are cryptic only in situations that are very special indeed. Disturbed squids often release blackish ink, which may hang in the water as a blob, somewhat contorted or "strung out" by the effects of currents. An individual that has inked usually shoots away from the scene in some disruptive or pale color pattern. Presumably predators often focus on the ink and fail to notice the escape of their quarry. Insofar as the ink serves as a decoy, something that looks like an animal or edible object, it is an image or mimic of the prey. On one occasion, one of us saw and blocked the retreat of two young sepioidea in very shallow water over pure white sand. They released ink and then, instead of going disruptive or pale, turned Dark and arranged their arms in "contorted" patterns. In these positions they looked like their own blobs of ink. Other predators might have been confused or distracted, not knowing which "blobs" were real or worth attacking. It would seem that some squids can mimic their own (self-produced) mimics!

Fig.10. A mixed school of squids, adult and sub-adult Loligo ("Doryteuthis") plei and Sepioteuthis sepioidea. From left to right: a plei with Center Light (paling) pattern in a Head-down posture; another plei with Center Light; a plei in the "ordinary" coloration of the species; a sepioidea in "ordinary"; a female sepioidea with a trace of Pied. The Pied and Center Light may be partly homologous.

Fig. 11. An adult male Sepioteuthis sepioidea in Lateral Silver color pattern. This pattern is sometimes assumed by courting males "defending" their females against intruders or possible rivals. It is the pale side of the body that is directed toward opponents. The performance is usually effectively repellent.

Fig. 12. A Flamboyant display, with Upward V Curling, by a young Octopus vulgaris. Vs with Upward Curling or Pointing may be as widespread as Dymantic and Longitudinal Streak patterns. Compare with Fig. 7. (After Packard and Sanders, 1969, 1971.)

There probably are additional patterns of sepioidea and other cephalopods that are interchangeable in much the same way(s) and perhaps to the same extent as the Bar, Streak, and Dark. The quality or potential is not, however, characteristic of all cephalopod displays. Many are as largely determined by internal factors and as little interchangeable as are most ritualized patterns of vertebrates and arthropods. And, of course, even vertebrate displays, if not those of arthropods, show some variation in this respect. The primary difference between major systematic groups must be quantitative rather than qualitative. The peculiarity of cephalopods in this context can be summarized crudely. Some of the ritualized patterns of some cephalopods, perhaps most of the littoral and inshore species, seem to be more often or directly controlled by (adapted to) their physical environments than are most of the corresponding patterns of other animals. The performance of a ritualized pattern by a cephalopod may entail more calculation, the weighing of pros and cons, than is usually true of other animals.

Fig. 13. A juvenile Octopus chierchiae eating a crab. This individual was captured on the Pacific coast of central Panama and maintained in the laboratory of the Smithsonian Tropical Research Institute on Naos Island. The drawing is based on a photograph taken in the laboratory.

Many cephalopods display while attacking and eating prey. Among the special patterns shown here are Stripes on the body, Darkening of the forward arms, and the beginning of a Longitudinal Streak along the side.

Fig. 14. Adult female Octopus oculifer ( from the Pacific coast of central Panama, photographed in the laboratory at Naos). This color pattern is an extreme of conspicuousness, presumably homologous with the Zebra Stripes of other species. The raising of the forward arms may be an intention movement or low-intensity version of some sort of Upward V Curling. The performance as a whole is certainly hostile, probably aggressive.

Fig. 15. A high-intensity dispute between two adult male Sepioteuthis sepioidea in a courting group. Both animals have their arms in Spread positions. The lower individual shows Zebra Stripes. This type of coloration, with or without Spreads, is still another of the conservative patterns.

Fig. 16. Zebra Stripes of Sepia officinalis. (After Tinbergen, 1939.)

This may be a reason why some or most coleoid cephalopods have relatively larger brains per body weight (Packard, 1972) than do the ecologically equivalent fishes.

The predominance of visual displays among coleoids is a problem in itself. As Tavolga (1968) and others have pointed out, aquatic environments are not ideal for visual communication. Light beams extinguish rapidly with distance in water and are scattered and obscured by suspended particles and plankton. It is obvious in the field that visual signals can be used only for communication over limited areas. The visual patterns of S. sepioidea, for instance, convey much information very rapidly, and sometimes broadly (in the sense of going off in different directions), but they do not transmit very far.

One may ask, therefore, why so many coleoids place such reliance on an imperfect channel, or why they do not make more use of other kinds of communication. There would appear to be two answers. First, visual signals are adequate. They work, and work well within their sphere. Second, they may have fewer drawbacks than the possible alternatives—for animals of coleoid habits and structures.

The deficiencies of visual communication are often minimized by two features. Many coleoids are so gregarious and cohesive that they do not need more than short- to medium-distance signals among themselves. Most of them also prefer the clearest waters available. The fact that cryptic or mimetic patterns cannot be seen at great distances is not a drawback.

Some of the possible alternatives, olfactory or other chemical signals, are slow and/or extremely diffuse. Tactile signals are very short range. Both kinds of signal may be too clumsy for animals as mobile as most coleoids, or be suitable for only a few types of social encounters. Coleoids do, in fact, use them for certain interactions, such as copulations, when a degree of close contact and a pause in other activities are perhaps inevitable.

Nautilus depends on touch, taste, and smell in a greater diversity of circumstances. It is also comparatively sluggish (Cousteau and Diolé, 1973).

The deep-water coleoids may have to use their chemical and tactile senses more frequently and extensively than do their littoral or surface relatives. It is suggestive, however, that many of them have evolved numerous or elaborate light organs. This would seem to indicate that they have retained at least part of the visual acuity and some of the preferences that are characteristic of the group as a whole.

Acoustic signals would have many theoretical advantages—they could be sent over long as well as short distances and refined to extreme precision—and it is not immediately evident why cephalopods do not use them. There may be an element of evolutionary "accident." Perhaps appropriate mutations did not occur among cephalopods at the right times. But there may also have been functional considerations. Much of the mobility of cephalopods is dependent on jet propulsion, a method of locomotion that seems to work best when the body is very flexible. Flexibility is made easier when hard parts are reduced. (This, in turn, may be a reason why most of the earlier shelled cephalopods were replaced by the largely unshelled coleoids.) In the absence of many hard parts, it may be difficult to evolve noise-making organs such as stridulating devices. Most coleoids lack structures like the "swim bladders" that are important in the sound reception of many fishes.

It is also possible that sounds are more likely than visual signals to attract the attention of potential predators. Some accounts of the acoustic behavior of fishes, e.g., Tavolga (1960), imply that many of the species that produce sounds most frequently are particularly well protected. The more vulnerable cephalopods may not have been able to take the same risks.

COMMENT

Two general aspects of the communication systems of cephalopods are of interest from a comparative point of view. (1) These systems are highly complex and sophisticated, probably as advanced as those of any other animals apart from man and (possibly) subhuman forms such as chimpanzees. Cephalopods are invertebrate, but they are not backward or primitive in their social behavior. (2) The repertoires of many coleoid cephalopods are distinguished by a maximum use and development of a single channel or medium of expression. They illustrate the intricacy and efficiency that limited systems can attain in conditions that are both favorable per se and narrowly bound in scope.

1. All photographs were taken in the field near the San Bias Islands along the Caribbean coast of Panama unless otherwise noted.