23881081911 Encyclopædia Britannica, Volume 18 — MimicryReginald Innes Pocock

MIMICRY, in zoology, the deceptive and advantageous resemblance presented by defenceless and edible species of animals to other species of animals living in the same locality, which are harmful or distasteful and are consequently avoided by all or by a majority of the enemies of the class to which the mimetic and usually the mimicked species belong. Mimicry is a special form of protective resemblance, differing from ordinary protective resemblance as exemplified by the similarity of the resting goat-sucker to a piece of bark or of leaf- and stick-insects to the objects after which they are named, in that the imitated object belongs to the animal kingdom and not to the vegetable kingdom or to inorganic nature. Although, like protective resemblance, quite independent of affinity between the organisms concerned in the likeness, mimicry occurs most commonly between animals structurally similar, and therefore related, to one another, the relationship may be close or remote. For instance, the commonest and best-known cases are found in insects where both mimic and model may belong to the same genus, sub-family, family or order, or to different orders. More rarely it occurs between members of distinct classes of the same sub-kingdom, i.e. between spiders and ants or spiders and beetles; yet even in this case both mimic and model have in common certain fundamental structural points to which the finishing touches completing the mimetic likeness are superadded. Still more rarely mimicry exists between totally unrelated species like caterpillars and snakes or spiders and snails. But in no case does it appear that the modifications in shape and colour, which contribute to bring about a mimetic resemblance, are greater and more elaborate than those which result in the simpler examples of ordinary protective resemblance.

The principle of protective resemblance, for which the term mimicry, as above defined, was originally employed, was first explained by H. W. Bates. Subsequently the meaning of the word was extended by F. Müller to include cases of mutual resemblance between two or more noxious species inhabiting the same area. Hence the resemblances belonging to the first category are commonly termed “Batesian mimicry,” and those belonging to the second category “Müllerian mimicry,” or more properly “Müllerian resemblance.” The difference between the two phenomena is essential and evident; but without experimental information as to palatability it is impossible to know with certainty to which of the two a particular case of mimicry is to be assigned. Over and over again extended knowledge on this point and inferences drawn from other facts have shown the certainty or probability of examples of mimicry being in reality “Müllerian,” which were previously accepted without question as “Batesian.” A simple illustration will serve to explain these two aspects of mimicry and to show the advantage in the struggle for existence that mimicry confers upon the species concerned.

There is a common English Syrphid fly (Eristalis tenax) known as the drone-fly from its resemblance to a large hive or honey bee. Honey bees are protected from a large number of insect enemies because they sting and are distasteful. Insect-eating birds soon learn to associate distastefulness with the size, form and colour of the bees, and consequently leave them alone after one or more trials. But flies of the drone-fly kind cannot sting, and, so far as is known, are perfectly innocuous and edible. The advantage to the fly of its deceptive resemblance to the bee is theoretically perfectly evident and practically can be demonstrated by experiment. It is in the first place a matter of common knowledge that human beings who have been taught to avoid handling bees invariably fear to touch drone-flies, unless specially trained to distinguish the one from the others. Moreover, Professor Lloyd Morgan found that young birds that had tasted and rejected workers of the hive bee as unpalatable subsequently refused to taste not only drones, which have no sting, but also drone-flies. So far as our information at present extends the resemblance between these two insects is a simple case of mimicry in the Batesian sense of the word. That is to say, an edible species is protected by resembling one that is inedible. But if it be discovered, as is possible, that the drone-fly is also inedible, the mimicry must be ascribed to the Müllerian category, and the reason for it becomes less evident. In what way, it may be asked, are two or more distasteful species of insects, occurring in the same locality, benefited by resembling each other? The ingenious explanation suggested by Fritz Müller for similar cases met with in butterflies is probably the true answer. This explanation depends upon what is now an experimentally demonstrated fact that insectivorous birds, and probably other animals, have no instinctive knowledge of what insects are edible and what inedible. This knowledge is acquired by experience; and since it is not, at all events as a rule, taught by the first taste to any individual bird, it is reasonable to infer that a considerable amount of injury, sufficient to disable if not to kill, is annually inflicted upon insects belonging to species protected by distastefulness or kindred qualities. Now insects that possess noxious attributes, and the same is true of other animals, usually have a conspicuous warning coloration which appeals to the eyes of enemies and helps them to remember more easily the cause of an unpleasant experience, helps in fact to establish a psychical association between a particular style of coloration and a nasty taste or a painful wound. This being so, it is evident that if all the distasteful species in a given area are differently coloured, some individuals of all the species will be annually sacrificed to the experimental tasting of inexperienced foes before the numerous lessons have been learnt. But if all the species in question resemble each other the resemblance will be mutually beneficial to them because the association between the two attributes they have in common, namely distastefulness and a particular scheme of colour, will be rapidly established. One lesson only, instead of many, has to be learnt; and once learnt at the expense of a few individuals of one or two species it will thereafter be applied indiscriminately to all. This type of mimicry has been well defined by Professor E. B. Poulton as the unification of warning colours.

Since belief in the adequacy of the two theories, above outlined, to account for the facts they profess to explain, depends ultimately upon the testimony that can be brought forward of the usefulness of warning characters, of the deception of mimicry and of the capacity for learning by experience possessed by enemies, it is necessary to give some of the evidence that has been accumulated on these points. (1) In South America there are butterflies formerly grouped as Heliconidae which are conspicuously coloured, slow of flight and abundant in individuals so as to be susceptible of easy capture. They possess scent glands. By observation and experiment it was discovered independently by Messrs Bates, Wallace and Bell that they are not attacked by birds nor by many other enemies that prey upon unprotected Lepidoptera. (2) As the result of a series of trials made in Calcutta F. Finn came to the conclusion that young birds have no instinctive knowledge of the unpalatability of distasteful insects, but that experimental tasting soon teaches them to recognize and avoid species they have previously rejected with dislike, and that having once learnt the lesson they long remember it. (3) That birds may also be deceived by insects that mimic those they have found to be uneatable has been shown by the above-quoted experiment with the drone-fly and the honey-bees made by Professor Lloyd Morgan. He also found that chickens that had been given meal moistened with quinine and placed upon glass slips banded black and yellow, afterwards refused to touch meal moistened with water and spread upon the same slips, although they had previously eaten it with readiness off plain coloured slips. With two exceptions, these chickens that had learnt to associate black and yellow banding with a bitter taste also refused to touch the caterpillar of the cinnabar moth (Euchelia jacobaeae), which is banded with these colours. Moreover, young birds that had been taught by experience that these caterpillars are uneatable also left wasps untouched. (4) Guy Marshall once offered to a baboon a distasteful butterfly (Acraea anemosa), holding the insect in such a way as to display its bright red and black markings to the monkey. It was taken but rejected after being tasted. A specimen of another butterfly (Precis sesamus) which mimics the Acraea was then offered in the same manner. The baboon took it, held it in her hands for a few moments, and then let it escape uninjured without trying to taste it. But when another butterfly of the same species, but with the wings cut off, was offered to her she promptly ate it without showing any sign of dislike. The results of this experiment with the baboon and of those with the birds are precisely what would be expected if the theory of mimicry is true. Experiments to test distastefulness have also been made with various kinds of insectivorous Arthropoda, like spiders and mantises. These experiments have shown that Arthropods also have their likes and dislikes in the matter of insect-food and frequently refuse to eat insects which are warningly coloured and are distasteful to vertebrated enemies. They appear, however, to have no appreciation of mimetic and warning colours, and have therefore not influenced in any way the evolution of mimetic resemblances dependent upon hues and patterns. Nevertheless, as explained below, it seems to be highly probable that ant-imitating insects and spiders, when the resemblance is dependent to a greater extent upon size, shape and movement than upon tint, have acquired their mimetic likeness especially to protect them from the attacks of such insect-enemies as predaceous wasps of the family Pompilidae, flies of the family Asilidae, and from so-called parasitic hymenoptera of the family Ichneumonidae, as well as from other insect-eating Arthropods.

The term mimicry has also been applied to resemblances of a different kind from the two enumerated above—resemblances, that is to say, by which predaceous species are supposed to be enabled to approach or mix without detection with animals they prey upon or victimize in other ways. To this end the resemblance may be actually to the species victimized or preyed upon or else to a species which the species preyed upon does not fear. This phenomenon is termed “aggressive mimicry” as opposed to the Batesian and Müllerian phenomena, which are termed “protective mimicry.” A few possible cases of aggressive mimicry are enumerated in the following summary of some of the recorded cases of mimicry in different classes of the animal kingdom; but the phenomenon is of comparatively rare occurrence, and the supposed instances may be susceptible of other interpretations, excluding them altogether from mimicry, or bringing them under the Batesian or Müllerian interpretation of the phenomenon.

Among mammalia there are no certain cases of mimicry known. It has been claimed that the resemblance between some of the Oriental tree-shrews of the genus Tupaia and squirrels comes under the category of aggressive mimicry, the tupaias being enabled by their likeness to approach and pounce upon small birds or other animals which, mistaking them for the vegetable-feeding squirrels, make no effort to get out of the way. But this hypothesis cannot be accepted as furnishing a satisfactory explanation of the likeness. For in the first place there seems to be no good reason for thinking that the Tupaias feed to any considerable extent upon prey of that kind, and in the second place the resemblance is due to characters which may be merely adaptations to a similar mode of life. A long and bushy tail, for instance, is a useful balancer and is a not uncommon feature in mammals which lead an active arboreal life. Similarly the dull coloration of the two sets of animals is very possibly procryptic and serves to hide both shrews and squirrels from enemies. Hence there seem to be good reasons for regarding the likeness in question as due to similarity in habitat and not as mimetic.

In East and South Africa there is a genus of Mustelidae known as Ictonyx (Zorilla) which possesses a foetid odour and is warningly coloured with black and white bands after the manner of skunks. There also occurs in South Africa another member of this family (Poecilogale albinucha), which is very similarly coloured. It is possible that this resemblance is mimetic in the Batesian sense of the word, and that the Poecilogale, if inoffensive, profits by its likeness to the highly offensive and warningly coloured Ictonyx. But, on the other hand, Poecilogale may itself be a protected form since subcaudal stink-glands are commonly found in species of the, weasel tribe. If this be the case the two species probably furnish an instance of true Müllerian mimicry. In South America there is considerable superficial resemblance between the little bush dog (Speothos venaticus) of Guiana and Brazil and the large weasel-like animal of the same countries—the tayra (Galera barbara). The tayra is, when adult, black beneath and on the legs, and not uncommonly has a considerable quantity of greyish hair on the head. In these particulars, as well as in size and shortness of leg, the dog resembles the weasel; and since there are good reasons for believing that the latter is protected alike by ferocity and stink-glands, it is quite possible that the dog, of unusual coloration and form for the Canidae, is protected from the attacks of pumas, jaguars and ocelots by his likeness to the tayra.

A few cases of mimicry have been recorded in birds. The common cuckoo and some other species inhabiting Africa and Asia closely resemble sparrow-hawks. Some cuckoos are singular for their habit of using the nests of smaller birds to lay their eggs in, so that the young may be reared by foster-parents; and it has been suggested that the object of the likeness exhibited to the hawk is to enable the cock cuckoo either to frighten the small birds away from their nests or to lure them in pursuit of him, while the hen bird quietly and without molestation disposes of her egg. The fact that both sexes of the cuckoo resemble the hawk does not necessarily prove this suggested explanation to be false; but if it be true that the smaller passerine birds are duped by the similarity to the bird of prey, it may be that the cuckoos themselves escape molestation from larger hawks on account of their resemblance to the sparrow-hawk. Another species of this group, the black cuckoo of) India, apparently mimics the black drongo-shrike (Dicrurus ater), the resemblance between the two species being very close. The drongo is a fierce and powerful bird which will not tolerate a strange bird of the size of a cuckoo near its nest, yet on account of its resemblance to the drongo, the hen cuckoo is enabled, it has been claimed, to lay her egg in the nest of the drongo, which mistakes the cuckoo for one of its own kind. In this case also both sexes of the cuckoo mimic the drongo, whereas according to the theory it would be necessary for the hen bird alone to do so. This suggests that the resemblance to the pugnacious drongo may be beneficial in protecting the defenceless cuckoo from enemies.

Some observations, however, of Guy Marshall on the inedibility of certain birds suggest that the resemblance between cuckoos and hawks on the one hand and cuckoos and drongos on the other may be susceptible of another explanation in full agreement with the theory of mimicry as propounded by Bates. He found that a South African drongo (Dicrurus (Buchanga) assimilis) was rejected after one or two attempts to eat it by a hungry mongoose (Herpestes galera) which had been starved for purposes of the experiment. The drongo is blue and black and is, he believes, warningly coloured. The same mongoose also refused to eat a kestrel (Cerchneis rupicoloides) and a hobby (Falco subbuteo), although it devoured certain other birds that were given to it. It is clearly possible, therefore, that cuckoos which mimic drongos and hawks may be protected from those enemies which find these birds distasteful.

One of the most perfect cases of mimicry in birds is presented by a Madagascar thrush or babbler (Tylas eduardi), which resembles feather for feather a shrike (Xenopirostris polleni), from the same island. The Tylas has departed from the normal coloration of its group to take on that of the shrike, a comparatively powerful and pugnacious bird. Analogous cases are supplied by the mimicry that exists between some of the orioles (Mimeta) and the friar-birds (Philemon or Tropidorhynchus) of the Austro-Malayan Islands. The friar-birds are noisy and pugnacious species of the group of honey-eaters, and mob hawks and other birds of prey, which leave them unmolested. The general style of coloration of orioles is gaudy yellow and black, rendering them invisible in sunlit foliage, and quite different from the more sombre hues of the friar-birds; but in the islands of Bourou, Timor and Ceram the orioles have not only assumed the tints of friar-birds in general, but in each of the islands named a species of oriole has acquired the little peculiarities in colour of plumage possessed by the friar-bird of the same locality. There seem to be no reasons for doubting that these are cases of genuine protective mimicry.

Apparently the only instances of mimicry known amongst reptiles occur amongst snakes; and in all the cases quoted by Wallace harmless snakes mimic venomous species. In tropical America the genus Elaps, which is both poisonous and warningly coloured, is a model for several innocuous snakes. In Guatemala Elaps fulvus is mimicked by Pliocerus equalis; in Mexico Elaps corallinus by Homalocranium semicinctum, and in Brazil, Elaps lemniscatus by Oxyrhopus trigeminus. In South Africa the harmless egg-eating snake (Dasypeltis scaber) is very like the Cape adder (Bitis atropos); and in Ceylon the harmless Colubrine Lycodon aulicus is alleged to mimic Bungarus ceylonicus, an ally of the deadly krait of India. Considering, however, the numbers of venomous and innocuous snakes that occur in most tropical countries, it might be supposed that mimicry in this order of reptiles would be of commoner occurrence than appears to be the case. It must be remembered, however, that apart from size and colour all snakes resemble each other in a general way in their form and actions. They present a strong family likeness which is not found in any other terrestrial vertebrated animals with exception of some lizards and possibly Caecilians amongst the Amphibia. So close indeed is the similarity that many monkeys, apes and human beings have an apparently instinctive fear of all snakes and do not discriminate between poisonous and non-poisonous forms. Hence it may be that innocuous snakes are in many instances sufficiently protected by their likeness in shape to poisonous species that close and exact resemblance in colour to particular species is superfluous.

As a possible instance of mimicry in fishes, A. T. Masterman recalls the fact that two species of weever (Trachinus draco and T. vipera), have the same habitat in British waters as certain species of soles (e.g. Solea vulgaris). The weevers are poisonous and the venom is concentrated principally in the six spines of the first dorsal fin. These spines are sharp and connected by a black membrane which projects, when the fish is disturbed, as a danger signal, it is believed, above the surface of the sand in which the fishes lie hid awaiting prey. For protective purposes soles, which are edible, also lie buried in or on the sand which they match in colour, with the exception of the right or upper pectoral fin which has a large black patch. When disturbed the soles raise this black fin and, as a rule, hold it rigid so that it becomes a very conspicuous object. If the view that the sole is protected by the blackness of the pectoral fin resembling the blackness of the dorsal fin of the weever, be correct, these fishes furnish an instance of Batesian mimicry. Furthermore, there is a common littoral fish in the Mediterranean (Uranoscopus scaber), belonging to the same family as Trachinus, exhibiting the same habits and living on the same ground, which also has a jet black erectile dorsal fin, and is believed to be poisonous. It is probable that the resemblance between Uranoscopus and Trachinus with respect to the colour of the dorsal fin is mutually beneficial to the two fishes. If so, the likeness must be regarded as an instance of Müllerian mimicry.

It is amongst Arthropods, however—and especially amongst insects—that mimicry, both Batesian and Müllerian, occurs in greatest profusion and perfection.

In insects of the order Orthoptera, departure from the normal in form and colour, carrying with it similarity to other living things, usually takes the line of protective resemblance to parts of plants. This is well exemplified by the leaf-insects (Phyllium) and stick-insects (Bactra), where the likeness to the models after which they are named is procryptic; and also by various species of tropical Mantidae which resemble flowers for the purpose of alluring insects within striking distance and perhaps also for concealing their identity from enemies. Some cases of genuine mimicry, however, are known in the order; Perhaps the best is that of the Sudanese Locustid (Myrmecophana fallax), which is strikingly ant-like. The head is large, the neck slender, the antennae short and the legs longish, and the appearance of the long stalk-like waist of the ant is produced by a patch of whitish hair on each side of the forepart of the abdomen which has the effect of cutting away the parts of the segments so covered, leaving a narrow dark-coloured median area to represent the waist. This at least is the method of disguise suggested by examination of the dried insect; but representatives of the same or an allied species found in Mashonaland were observed in the living state to be green with the antlike parts represented in black pigment. These parts were quite conspicuous against the green of the plants frequented by the insects, wherever the green portions were rendered invisible by the same background. Ant-mimicry has also been recorded in the case of the larva of one of the Indian species of Mantidae. Again, several species of this order have become profoundly modified in form in imitation of inedible beetles. In the Philippines, a cricket (Scepastus pachyrhynchoides), has taken on the shape and coloration of a species of Apocyrtus, a hard and inedible weevil (Curculionidae); and Phoraspis, a kind of grasshopper similarly resembles ladybirds (Coccinellidae). A species of beetle (Caria dilatata) of this family in Borneo is mimicked by a species of a genus allied to Gammarotettix not only in shape and coloration but also in the habit of remaining still when disturbed. In the same island a species of Gryllacris mimics Pheropsophus aquatus, a “Bombardier” beetle which ejects a puff of volatile formic acid when attacked; and Condylodera tricondyloides mimics different species of tiger-beetles (Cicindelidae) at different stages of its growth. Finally the larva of one of the Bornean Mantidae, which is a floral simulator in its pupal and adult stages, closely resembles in its black and red coloration the larva of the stinking and warningly coloured bug Eulyes amoena.

Comparatively few cases of mimicry in the Neuroptera have been observed. There are records, however, of species of Mantispa mimicking the wasp Polistes in North America and Borneo and Belonogaster in South Africa; and other species of the genus imitate parasitic hymenoptera of the genera Bracon and Mesostenus.

Coleoptera (beetles) supply instances of mimicry of ants, wasps and Ichneumonids, and some defenceless forms of this order mimic others that are protected. A good illustration of wasp-mimicry is furnished by a large heteromerous beetle (Coloborhombus fasciatipennis) from Borneo which is remarkably like a large wasp (Mygnimia aviculus) from the same island. The front wings of the wasp have a conspicuous white patch near the tip and a patch similar in size and colour is present on the wings of the beetle, which, unlike the majority of beetles, habitually keeps its wings extended, and since the elytra are exceptionally short the wings are not covered by them when folded. The resemblance also extends to the general form of the body and to the length and thickness of the wings and antennae. The elytra are equally reduced, and apparently for the same purpose, in an Australian Longicorn beetle (Esthesis ferrugineus), which, like so many wasp-like Hymenoptera, has the body banded red and black. This beetle probably mimics the Australian hornet (Abispa australis). In the European Longicorn (Clytus arietis), on the other hand, the elytra are of normal length and are banded with yellow stripes. The beetle, moreover, is of slender build and all its actions are suggestively wasp-like. This may, however, be an instance of Müllerian rather than of Batesian mimicry, the beetle being itself inedible; for Shelford has stated his conviction that the Bornean representatives of the sub-family (Clytinae), to which Clytus arietis belongs, are all highly distasteful and are warningly coloured, as are members of this sub-family from other parts of the world.

In the Philippine Islands several species of Longicorns of the genus Doliops mimic hard inedible weevils (Curculionidae) of the genus Pachyrhynchus. The antennae of these weevils are short and end in a knob; those of the Longicorns are very much larger, but the weevil-like look is produced by the presence of a knob-like swelling upon the third joint, the terminal portion of the antenna being so extremely fine as to be almost invisible. Similar modification of the antennae in the Longicorn Estigmenida variabilis brings about the resemblance between this beetle and a beetle, Estigmena chinensis, one of the Phytophaga of the family Hispidae. Numerous instances of mimicry in this order of insects have recently been recorded from Borneo by R. W. C. Shelford, a large number of them being in all probability Müllerian.

Instances of ant-mimicry, unique in the method employed to bring about the resemblance, are supplied by some insects of the Homopterous group of the Rhynchota, belonging to the family Membracidae. In one of these (Heteronotus trinodosus), the dorsal area of the forepart of the thorax is developed into a plate which projects backwards over the body of the insect, which retains its normal form, and conceals all but the head, wings and legs. This shield if shaped in such a manner as to resemble closely the body of an ant, the median portion of the shield being deeply constricted in imitation of the waist and the terminal portion sub-globular like the abdomen of the ant. This insect comes from Central America. Still more curious is the mimicry of another of these insects from Venezuela which is found in company with a leaf-cutting ant (Oecodoma cephalotes) of that country. When pursuing their operations of leaf-storage, these ants present the appearance of a crawling crowd of leaf-particles, fragments of leaves being carried by the insects in such a way as to conceal to a great extent the insect underneath, of which little more than the dark coloured legs project beyond the burden. The immature form of the above-mentioned species of Membracidae mimics both ant and leaf-particle. The legs and lower part of the body are dark coloured, but the dorsal surface of the thorax and abdomen is coloured green and is raised so as to form a crest with jagged edges exactly reproducing the irregular margin of a fragment of leaf cut out by the mandibles of the ant. In Borneo the Homopteron Issus bruchoides mimics a species of Curculionid beetle of the genus Alcides.

In the Hemipterous group of the Rhynchota ant-mimicry is illustrated by the larva of a British species of Reduviidae (Nabis lativentris) in which the forepart of the abdomen is furnished on each side with a patch of white hairs leaving a central narrow dark portion in imitation of the waist of the ant; and also by an East African species (Myrmoplasta mira) which in its general form exhibits a close resemblance to an ant (Polyrrhacis gagates) which occurs in the same neighbourhood. Another instance in this group is supplied by a Bornean species of Reduviidae which mimics a species of the genus Bracon, one of the parasitic Hymenoptera.

Typical dipterous insects (flies) closely resemble in general form aculeate Hymenoptera belonging to the families of bees and wasps. The changes in colour and structure required to complete the resemblance to particular species are comparatively slight and much less complicated than those needed to produce a likeness to other protected insects. Hence we find that the majority of flies that mimic insects of other orders have bees or wasps for their models. Many of the Syrphidae are banded black and yellow and present a general resemblance to wasps, especially when they alight, the resemblance being enhanced by a twitching action of the abdomen imitating the similar action so familiar in species of stinging hymenoptera. These flies are characterized by a peculiar method of flight. They commonly hang poised in the air, then dart with lightning swiftness to another spot and poise themselves again. This habit, the origin of the name “hover-flies,” is probably connected with their mimetic coloration. If they flew like ordinary flies their resemblance to Hymenoptera would be obscured by the rapidity of their flight and they might be caught on the wing by insectivorous birds or other insects; but when poised they display their coloration. When the latter is lost during flight, the rapidity of their movement defies pursuit. The particular likeness to a honey-bee presented by one member of this family, the drone-fly (Eristalis tenax), has been already referred to. But the likeness probably goes deeper than superficial resemblance that appeals to the eye; for spiders which distinguish flies from bees by touch and not by sight, treat drone-flies after touching them, not in the fearless way they evince towards blue-bottles (Calliphora), but in the cautious manner they display towards bees and wasps, warily refraining from coming to close quarters until their prey is securely enswathed in silk. This forcibly suggests that the drone-fly mimics a honey-bee not only in appearance but also in the feel of its hairs or the nature of its buzz. Other flies of the genus Volucella, larger and heavier in build than Eristalis, resemble humble-bees in colour and form, and it was formerly supposed that the purpose of this similarity was to enable the flies to enter with impunity the nests of the humble-bees and to lay their eggs amongst those of the latter insects. But it has been ascertained that the species of Volucella which behave in this manner also visit for a like purpose the nests of wasps, which they do not resemble. Hence it is probable that this case of mimicry is purely of a protective and not of an aggressive nature and serves to save the flies from destruction by insectivorous enemies. The same explanation no doubt applies to the mimicry, both in Borneo and South Africa, of hairy bees of the family Xylocopidae by Asilid flies of the genus Hyperechia, and also to other cases of mimicry of Hymenoptera as well as of inedible beetles of the family Lycidae by Diptera. Numerous other cases of mimicry between Diptera and Hymenoptera might be cited.

The Lepidoptera furnish more instances of mimicry, both Batesian and Müllerian, than any other order of insects. In the majority of cases both model and mimic belong alike to the Lepidoptera, and it is often uncertain whether both are inedible (Müllerian mimicry) or whether inedibility is the attribute only of the model (Batesian mimicry). A large number of cases that were formerly regarded as belonging to the latter category are now suspected of belonging rather to the former. Sometimes Lepidoptera mimic protected members of other orders of insects—such as Coleoptera, Hymenoptera and Hemiptera; but perhaps the most singular illustrations of the phenomenon known in the order are exemplified by the larvae of the hawk-moth Chaerocampa, which imitate the heads of snakes. Professor Poulton long ago suggested, and supported the suggestion by experimental evidence on a lizard, that the larvae of two British species, C. elpenor and C. porcellus, are protected by the resemblance to the heads of snakes presented by the anterior extremities of their bodies which are ornamented with large eye-like spots. When the larvae are disturbed the similarity is produced with startling suddenness by the telescopic contraction of the anterior segments in such a manner as to suggest a triangular, pointed head with two large dorsal eyes. Subsequent observers (A. Weismann, Lady Verney) have shown by experimenting upon birds that this suggestion is correct; and Guy Marshall found that baboons which are afraid of snakes are also afraid of the snake-like larva of the South African Chaerocampa osiris. Finally Shelford states that the anterior end of a Bornean species (C. myodon) offers a striking and detailed resemblance to the head of a snake (Dendrophis picta).

Instances of ant-mimicry in this order are sometimes confined to the larval stage. The early larval stage of the “Lobster Moth” (Stauropus fagi) for example, presents a general resemblance, due to a combination of shape, colour, attitude and movements, to black ants, the swollen head and the caudal disk with its two tentacles representing respectively the abdomen and antenna-bearing head of the model. A parallel case of mimicry exists at Singapore between the larva of a Noctuid moth and the common red tree-ant (Oecophylla smaragdina). In this case also the posterior end of the larva represents the anterior end of the ant. Another instance of mimicry affecting the larval form is supplied by the moth Endromis versicolor, the caterpillars of which resemble the inedible larvae of saw-flies. The resemblance that certain moths—e.g. Trochilium apiforme, crabroniforme—present to bees and wasps is effected in the main by the loss of the scales from the wings, leaving these organs transparent. It is important to note that the scales are present when the moths first emerge from the pupa-case, but are loosely attached and fall off with the first flight.

Of the multitudes of cases of mimicry between different species of Lepidoptera, a few only can be selected for description. These cases, however, have a peculiar interest and importance for they have been studied in fuller detail than any others and the discovery of a particular instance in South America first suggested to Bates the theoretical explanation of this bionomical phenomenon. On the Amazons and in other parts of South America there are butterflies of the group Ithomiinae which are distasteful and have all the characters of specially protected species, being conspicuously coloured, slow of flight, careless of, exposure and abundant in individuals. The wings are transparent and are black-bordered and black-barred, the anterior wing having two black bars and the posterior one. This type of colouring is also found in genera of quite distinct sub-families of butterflies, namely in Danainae and Pierinae, as well as in some diurnal moths, all of which occur in the same district as the Ithomiinae. The following species may be cited as instances of this type of pattern: Methona confusa, Thyridia psidii, Eutresis imitatrix and Dirgenna dero (Ithomiinae); Itura ilione and I. phenarete (Danainae); Dismorphia orise (Pierinae); Anthomyza buckleyi (moth of the family Pericopidae) and Castnia linus (moth of the family Castniidae). So alike in form, colour and mode of flight are those Lepidoptera that when on the wing it is almost or quite impossible to distinguish one from the other, and the resemblance between members belonging to different sub-families cannot be assigned to affinity. Microscopical examination of the wings, moreover, has shown that the transparency of the wings, common to all, has been acquired by a different modification of the scales in each of the genera exhibiting the Ithomiine type of coloration. That the Danaine and Ithomiine species are distasteful is known. Itura, for example, belonging to the former, has protrusible scent-emitting processes at the end of the abdomen; and Thyridia has scent-producing tufts of hair on the edge of the posterior-wing. Bates offered no satisfactory explanation of the resemblance between these two genera and others of the same protected sub-families; but he did not hesitate to ascribe the resemblance to them presented by the Pierine, Dismorphia (Leptalis) orise, to mimicry, believing Dismorphia to be unprotected and noting that it departed widely in the matter of coloration from typical members of the sub-family to which it belongs. Although mimicry in the Lepidoptera has been carried to a greater extreme in South America than in any other country of the world, remarkable instances of it have taken place in the Ethiopian and Oriental regions. A classical and highly complex case first investigated and explained by R. Trimen is that of Papilio dardanus which is widely distributed in Africa and is represented by several sub-species or geographical races. The most primitive of these is antinorii from Abyssinia, which is non-mimetic and has the two sexes nearly alike. The males, of the other subspecies are much like the males of antinorii; but the females are widely different and mimic various species of inedible butterflies belonging to the protected groups of the Danainae and Acraeinae. One of these sub-species, merope, which ranges from the west coast to Victoria Nyanza, is polymorphic and occurs under three forms, namely (ahippocoon, which mimics the Danaine Amauris niavius; (btrophonius, which mimics the Danaine Limnas chrysippus; (cplanemoides, which mimics the Acraeine Planema poggei. Oddly enough one or more of these forms may occur in other sub-species. For example, the sub-species cenea which occurs in south and south-east Africa not only has the form cenea mimicking two Danaines, Amauris echeria and A. albimaculata, but also the hippocoon form which resembles a local race of Amauris niavius, known as dominicanus. The sub-species polytrophus from the Kikuyu Escarpments also has the planemoides and cenea forms and another form trimeni, which is intermediate between the unmodified female of antinorii and hippocoon, and like the latter is mimetic of Amauris niavius dominicanus. Finally the sub-species tibullus from the east coast has the cenea-form, the trimeni-form and probably the planemoides-form. The study of this intricate case is not yet completed and it is at present unknown whether it is an instance of Batesian or Müllerian mimicry. Special attention may be drawn to two phenomena connected with it, both of not uncommon occurrence in mimetic Lepidoptera. The first is the occurrence of mimicry only in the female sex. The reason for this is to be found in the greater need of protection of the female which is slower in flight than the male and is exposed to special danger of attack when resting to lay her eggs. The second noteworthy phenomenon is the mimicry of more than one protected species by members of a single species. This is a not uncommon occurrence, and in the case of Batesian mimicry the explanation is probably this. When an edible species gains protection by mimicking a distasteful one, there is a likelihood of its increasing in numbers until it equals or surpasses its model in this respect. Were this to take place the purpose of the mimicry would be abortive, because enemies would probably not refrain from slaughter if even every alternate capture proved palatable. It is advantageous therefore that the numbers of the mimetic species should be fewer than those of the model; and this appears to be achieved in some cases by the individuals of the mimetic species dividing themselves between two or more models.

Spiders furnish numerous instances of mimicry. Though simple in kind, many of these are as perfect illustrations of the phenomenon as any found in the animal kingdom.

Amongst the orbweavers of the family Argyopidae there are species belonging to the genera Cyclosa and Cyrtophora which closely resemble small snail-like gastropods as they cling to the underside of leaves with their legs drawn up. Other members of the same family—like Araneus coccinella, and Paraplectana thorntoni—imitate beetles of the family Coccinellidae which are known to be distasteful; and certain genera of the family Salticidae (Homalattus and Rhanis) closely resemble small hard-shelled beetles.

The most perfect cases, however, are exhibited by those species which imitate ants. The structural modifications required to convert a spider into the image of an ant are of a more complicated character than those that serve the same purpose in an insect. All insects have the same regional division of the body into head, thorax and abdomen, the same number of legs, a pair of antennae and a segmented abdomen. Spiders on the contrary have no antennae, no separate “head,” an unsegmented abdomen and an additional pair of legs. In the majority of ant-imitating spiders the forepart of the cephalothorax is constricted on each side to resemble the neck of the insect, and in many cases the similarity is increased by the presence of a stripe of white hairs which has the optical effect of cutting out an extra piece of integument, exactly as occurs in analogous cases in insects. Narrowing of the posterior portions of the spider's cephalothorax and sometimes of the anterior end of the abdomen reproduces the slender waist of the ant, and frequently transverse bands of hairs represent the segmentation of this region in the insect. The legs become slender and those of the first or of the second pairs are held up and carried in front of the head to simulate the antennae of the ant. Added to this the spiders commonly copy to the life the mode of progression and the restless activities of their models.

The likeness presented varies considerably in degree from a general resemblance to several species, such as is seen in the Salticid spider (Peckhamia picata) of North America, to a close similarity to particular species. To this category belong Myrmarachne plataleoides, one of the Salticidae, and Amyciaea forticeps, one of the Thomisidae which in India imitate and live with the vicious little red ant (Oecophylla smaragdina); also Myrmarachne providens, which mimics the red and black Indian ant (Sima rufonigra); and the South American species of Clubionidae, e.g. Myrmecium nigrum, which is an accurate copy of the large black ant (Pachycondyla villosa).

Sometimes it is only the males of a species of spider that mimic ants, as in the case of Ildebaha mutilloides and I. myrmicaeformis, two South American species of the family Argyopidae, in which the females are protected by strong spine-armature. The males are without these protective spines and are exposed to special dangers as they wander in search of the webs of the females. In South Africa too the males of a species of Eresidae (Seothyra) resemble and are found in company with a large ant (Camponotus fulvopilosus), which is common on the veld. Like the males of Ildebaha, those of Seothyra, wander about by day in search of the females which live concealed in burrows. Many other spiders belonging to the Theridiidae and Linyphiidae also mimic ants; but it is needless to enumerate them, the most perfect examples of this phenomenon being found in the families Clubionidae and Salticidae.

Ant-mimicking spiders have been seen now and again to devour their models. It has therefore been suggested by some and taken for granted by others that the resemblance comes under the category of aggressive mimicry and that the ants are deluded by this resemblance into regarding the spiders as members of their own species. That the ants do not destroy them is certain; but that they are deceived by the superficial similarity of the spiders to themselves is highly improbable, for these insects are capable of distinguishing a strange ant belonging to the same species if it comes from another colony. Moreover, the above-suggested explanation does not coincide with the explanation of the likeness to ants shown by certain insects such as Myrecophana fallax, the ant and leaf-like Membracid Homopteron and the larvae of the lobster-moth (Stauropos fagi), which are plant-eaters. It is probable that one explanation—namely, that of protection—covers all cases of ant-mimicry; and this explanation lies in all probability in the immunity from the attacks of most insectivorous enemies that ants enjoy, and especially from predaceous wasps of the family Pompilidae which annually destroy thousands upon thousands of spiders to feed their larvae; and since more than one observer has testified to the fear and abhorrence these wasps have of ants, it is needless to look farther for the benefit ant-mimicry is to spiders. These wasps, moreover, also provision their nurseries with caterpillars, grasshoppers and other insects. Hence it may be inferred that the insects which imitate ants profit in the same way that spiders do from this form of mimicry.

In the above-cited historical instance of mimicry amongst some South American Lepidoptera which formed the foundation of Bates theory, species of butterflies, belonging to the Ithomiine genus Itura and the Danaine genus Thyridia, both unpalatable forms, resemble each other. This is a very simple case of the possession of the same type of coloration by two or more protected insects inhabiting the same district. The significance of this phenomenon, as already stated, was first explained by Fritz Müller; but although the term “Müllerian mimicry” has been assigned to this and similar instances, they are not strictly speaking cases of mimicry at all but of warning coloration. Poisonous or noxious animals usually have some special advertising attribute, sometimes the display of conspicuous coloration, as in the skunk; sometimes the emission of sound as in the rattlesnake; sometimes a combination of the two, as in the common porcupine and the large black scorpions of Africa and India. Such characters have been termed by Professor Poulton “aposematic.” Neither of the above-mentioned animals is mimicked; but where two or more noxious animals, inhabiting the same district, resemble each other, both being aposematically or warningly coloured, the likeness is said to be “synaposematic.” Synaposemasy is Müllerian mimicry. Finally, the likeness of an edible species to a warningly coloured inedible one in the same locality is termed “pseudaposematic,” in allusion to the pretentiousness or falsity of the warning signal. Pseudaposemasy is Batesian mimicry.

An important phenomenon connected with insect mimicry is the convergence of several species in the same area towards a common type of coloration and shape, exhibited by one or more than one protected form. The resemblance shows various grades of completeness; and the convergent mimics may be themselves noxious, or edible and innocuous. In other words the insects entering into the combination may furnish instances of Batesian and of Müllerian mimicry. Very commonly different species of aculeate Hymenoptera, inhabiting the same district, form the centres of mimetic attraction for insects of various orders, so that a considerable percentage of the insect-fauna can be arranged in groups according to the pattern of the particular model the species have copied. Good illustrations of this law have been discovered by Guy Marshall in Mashonaland. He found on the same day on a bud of vetch, specimens of black ants (Camponotus sericeus and C. cosmicus), black ant-like Hemipterous insects (Megapetus atratus), and the ant-like Orthopteron (Myrmecophana fallax) (cf. supra). In this little coterie the ants are beyond question the models towards which the bug and the grasshopper have converged in appearance. Since many of the insects of the order Hemiptera are distasteful, the mimicry of the bug (Megapetus) is in this case probably Müllerian or synaposematic; the grasshopper (Myrmecophana), on the other hand, is probably edible and the mimicry is Batesian or pseudaposematic. This is a simple case consisting of a small number of component species. Others are more complex, numerous species being involved. In Mashonaland, for instance, a large number of genera and species of Hymenoptera belonging to the Apidae, Eumenidae, Sphegidae, Pompilidae, Scoliidae, Tiphiidae and Mutillidae, resemble each other in having black bodies and dark blue wings. The same style of coloration is found in Coleoptera of the families Cetoniidae and Cantharidae; in Diptera of the families Asilidae, Bombylidae, Tabanidae and Tachinidae; in Hemiptera of the family Reduviidae and in Lepidoptera of the family Zygaenidae. In this instance the Hymenoptera, of which the coloration is synaposematic, form together a composite model which the other insects have mimicked. Of the latter, the Lepidopteron (Tascia homochroa) is distasteful, as also are the beetles of the family Cantharidae (e.g. Lytta moesta). Probably the bugs too (e.g. Harpactor tristis) are protected. The mimicry of these insects therefore is synaposematic; but some, at all events, of the flies like the Bombylid Exoprosopa umbrosa, probably form pseudaposematic elements in the group. Into another category Hymenoptera enter not as models but as mimics, the models being inedible Malacodermatous beetles mostly belonging to the genus Lycus and characterized by orange coloration set off by a large black patch upon the posterior end of the elytra and a smaller black spot upon the thorax. Towards this Lycoid centre have converged Coleoptera (beetles) of the sub-order Lamellicornia (Copridae), Phytophaga; Heteromera (Cantharidae) and Longicornia; Hemiptera of the families Pyrrhocoridae, Lygaeidae, and Reduviidae; Lepidoptera of the families Arctiidae and Zygaenidae; Diptera of the family Asilidae; and lastly Hymenoptera of the families Braconidae, Pompilidae, Crabronidae and Eumenidae. With the exception of the Asilid fly and perhaps some of the Longicorn and Phytophagous beetles, which are probably protected Batesian mimics, all the other species constituting the above-mentioned assemblage are, it is believed, Müllerian or synaposematic mimics. In the three cases cited above, with the exception of the first, the synaposematic mimics are vastly in excess of the pseudaposematic; this appears to be the general rule elsewhere. Frequently the groups are composed solely of protected species, so far as is at present known; and sometimes solely, in all probability, of unprotected species with exception of course of the model. An example of the latter occurs in Singapore where the vicious red spinning-ant (Oecophylla smaragdina) is mimicked by the larva of a Noctuid moth and by spiders belonging to two distinct families, namely, Saltiicus plataleoides (Salticidae) and Amyciaea forticeps (Thomisidae), there being no reason to suppose that either the moth larva or the spiders are protected forms. Mimetic aggregations of species similar to those mentioned above have been found in other countries; but the instances cited are sufficient to show how widespread are the influences of mimicry and how profoundly it has modified the insect fauna of various parts of the world.

Bibliography.—H. W. Bates, Trans. Linn. Soc. (Lond., 1862); id. The Naturalist on the Amazons (1879); T. Belt, The Naturalist in Nicaragua (2nd ed. 1888); F. A. Dixey, Rep. Brit. Assoc. (1894) p. 692; id. Tr. Ent. Soc. (London, 1894), p. 249; id., op. cit. (1896); p. 65; id., op. cit. (1897), p. 317; also Proc. Ent. Soc. (Lond. 1897), pp. xx.–xxxii. and xxxiv.–xlvii.; F. Finn, Journ. Asiatic Soc. Bengal, lxiv., (1895); lxv. (1896) and lxvi. (1897); E. Haase, Bibliotheca zoologica, (1891–1893, Stuttgart; English trans. by C. M. Child, 1896); G. A. K. Marshall, Trans. Ent. Soc. (London, 1902), pp. 287-584 (annotated by E. B. Poulton); A. T. Masterman, Journ. Linn. Soc., xxx., 239-244 (1908); R. Meldola, Proc. Ent. Soc. London (1877), p. 12; id. Ann. Mag. Nat. Hist. (5)x. (1882); C. Lloyd Morgan, Habit and Instinct (London, 1896); id. Animal Behaviour, pp. 164-165 (London, 1900); F. Müller, Kosmos (May 1879), p. 100; (trans) Proc. Ent. Soc. London (1879), xx.; A. Newton, A Dictionary of Birds, p. 572-575 (London, 1893–1896); E. G. Peckham, Occasional Pap. Nat. Hist. Soc. Wisconsin, i. (1889); R. I. Pocock, Journ. Linn. Soc. Zool., pp. 256-270 (1909); id. Proc. Zool. Soc. London, 944-959 (1909); E. B. Poulton, Proc. Zool. Soc. London (1887), 191-274; id. The Colours of Animals, 216-244 (1890); id. “Natural Selection the Cause of Mimetic Resemblance and Common Warning Colour,” Journ. Linn. Soc. Zool., xxvi. (1898); revised and amplified in Essays on Evolution, pp. 220-270 (1908); id. “Mimicry and Natural Selection,” Verhandl. d. V. internat. zool. Congr. Berlin (Jena, 1902); revised in Essays on Evolution, 271-292; id. “The Place of Mimicry in a Scheme of Defensive Coloration,” Essays on Evolution, pp. 293-382 (1908); W. P. Pycraft, The Story of Bird Life, pp. 32-33 (“Mimicry”), (1899); M. Roelofe, C.R. Soc. Ent. Belg. (2), No. 59 (1878); R. Shelford, Proc. Zool. Soc. (1902), ii. part 2, 230-284; R. Trimen, Trans. Linn. Soc., xxvi. p. 497 (1870); A. R. Wallace, Proc. Zool. Soc. (1863), pp. 26-28; id. Trans. Linn. Soc., xxv. pp. 19-22 (1866); id. The Malay Archipelago, ii. (London, 1869); id. Contributions to the Theory of Natural Selection, pp. 103-106 (London, 1875); id. Darwinism, pp. 239-265 (London, 1889); A. Weismann, The Evolution Theory, Eng. trans. (London, 1904).  (R. I. P.)