Butterflies and MothsButterfly wing scales, magnified.(From Holland)The butterflies and moths both belong to the great order of scale-winged insects—the Lepidoptera. They are distinguished, however, by certain general characteristics, which hold true for the most part in both groups. The butterflies fly by day; the moths fly by night. All of the higher butterflies go into the chrysalis state without making a silken cocoon, while most of the higher moths make such a cocoon. The bodies of the butterflies are usually slender, while those of the larger moths are stout. The antennae of the butterflies are generally slender and commonly enlarged at the tip into a miniature club. The antennae of the larger moths are commonly feathery or are long and slender, tapering gradually toward the tip.The characteristic features that distinguish a moth froma butterfly are well illustrated in the plate opposite page 49, which shows one of the largest and most beautiful moths in the world. It is the Cynthia moth. As may be seen, the newly emerged moth is resting upon the silken cocoon in which it spent its period as a pupa or chrysalis. This cocoon was attached by the caterpillar to the twig from which it hangs at the time it spun the cocoon. The feathered antennae, the hairy legs, the thick thorax, and large abdomen—all show very clearly in this side view of the moth. As will be seen, the wings are large and very suggestive of those of a butterfly and have the characteristic eye-spots toward the tip and the crescent marks in the middle, which are so often found on the wings of the larger moths.Some of these large moths on cloudy days occasionally fly during daylight and, by the uninitiated, they are often mistaken for large butterflies. One who will notice their structure, however, will readily see the characteristic features of the moth.In the caterpillar stage, there are no hard and fast differences between the larvae of butterflies and those of the higher moths. In each case, the insect consists of a worm-like body, having a small head provided with biting jaws and simple eyes or ocelli. Back of the head are the three rings of the thorax, each of which bears a pair of jointed legs. Back of these three rings there are a considerable number of other body rings making up the abdomen, on the middle of which there are commonly four or five pairs of fleshy prolegs, not jointed but furnished at the tip with fine claws. At the hind end of the body there is another pair of prolegs similar in structure.The Scents of ButterfliesAndroconia from wings of male butterfliesMany students of American butterflies have occasionally mentioned the fact that certain species seem to give off a distinct scent which has frequently been spoken of as a pleasing fragrance, suggesting sandalwood or some other aromatic odor. The general subject as exemplified by butterflies of other lands has been studied for many years by Fritz Müller; and certain English entomologists have paid considerable attention to it. A translation of the Müller publications and an excellent summary of our present knowledge of the subject is published in Dr. Longstaff's book on butterfly hunting.The odors given off by butterflies are divided into two principal kinds, namely: first, those which are repulsive to the senses of man, and evidently for the purpose of protecting the butterflies from birds and other vertebrate enemies—these are found in both sexes; second, those which are evidently for the purpose of sexual attraction and confined to the male butterflies—these scents are usually attractive to the senses of man.The aromatic scents of the second group are generally produced by means of certain scales or hairs of many curious forms, which are scattered over the surface of the wings or are placed within certain special pockets, generally near the borders of the wings. These scales or hairs are called androconia. Some of them much magnified are represented in the picture above.Our knowledge of the scents given off by Americanbutterflies is very fragmentary, and it is highly desirable that many more observations should be made upon the subject. If collectors generally would make careful notes, both in the field and upon the freshly killed butterflies at home, we ought soon to be able greatly to extend our knowledge. By holding the butterfly with a pair of forceps, one can often determine whether the fragrance is emitted. It is often helpful also to brush the hairs or tufts where the androconia are attached, using a small, dry camel's hair brush for the purpose.Butterfly MigrationsMigration seems to be a general instinct in the animal world, developed when a species becomes enormously abundant. At such times this instinct apparently overcomes all others and the creatures move on regardless of obstacles and conditions that may mean certain death to the vast majority. Such migrations among mammals have often been recorded, one of the most notable examples being that of the little lemmings which migrate at periodical intervals in a way which has often been described. Among the insects such migrations have been frequently noticed, and the phenomenon has apparently been observed oftener among the butterflies than in any other group. Entomological literature during the last hundred years contains a great many records of enormous flights of butterflies over long distances, extending even from Africa into Europe or from one part of America to another far remote. As such migration is likely to happen whenever a species becomes extremely abundant it probably is Nature's way of providing for an extended food supply for the succeeding generations. That it results in the death of the great majority of the migrants is doubtless true, but it must lead to vast experiments in extending the geographic area inhabited by these species. Numerous examples of such migrating swarms will be found in the pages of this little book. (See plates, pages17,48-49,160.)Photographed from life by A. H. VerrillSeepage 12SWALLOWTAIL BUTTERFLY JUST OUT OF ITS CHRYSALIS.THE CHRYSALIS SKIN BELOWSeepage 72TIGER SWALLOWTAILPhotographed from lifeSeepage 7THE HAMMOCK CATERPILLAR ON ITS HAMMOCKThe migrations thus considered are only exceptionaloccurrences. There is, however, a regularly recurring annual migration on the part of some butterflies which is also a phenomenon of extraordinary interest. The most notable example is that of the Monarch which apparently follows the birds southward every autumn and comes northward again in spring. There is much evidence to indicate that in some slight degree other butterflies have a similar habit, although the present observations are inadequate to determine to what extent this habit has become fixed in most of these species.Hibernation or Winter LethargyThe ways in which butterflies spend the winter are always of peculiar interest to the naturalist. Here are creatures with four distinct stages of existence, each of which has the possibility of carrying the species through the season of cold. It is necessary to learn for each insect which stage has been chosen for the purpose, and if possible to find the reasons for the choice.As a rule the related members of a group are likely to hibernate in a similar stage. Thus most of the Swallowtails pass the winter as chrysalids while practically all the Angle-wings pass the winter as adults. This rule, however,has many exceptions, for one will often find closely related species which differ in the stage of hibernation.As one would expect, the conditions of hibernation vary greatly with the latitude. In the severe climate of the far north the conditions are likely to be more uniform than in the South where the milder climate permits greater variation to the insect. In some cases where a butterfly hibernates in only one stage in Canada it may pass the winter in two or more stages in Alabama or Florida.In many other orders of insects the egg is a favorite stage for hibernation. Even in the closely related moths it is often chosen by many species, but comparatively few butterflies pass the winter in the egg stage. The little Bronze Copper may serve as one example of this limited group.The conditions as to hibernation by the larvae of butterflies are very different from those of the egg. It has been estimated that probably half of all our species pass the winter in some stage of caterpillar growth. This varies all the way from the newly hatched caterpillar which hibernates without tasting food to the fully grown caterpillar which hibernates full fed and changes to a chrysalis in spring without eating anything at that time. A large proportion, however, feed both in fall and spring, going through the winter when approximately half grown.The Graylings and the Fritillaries are typical examples of butterflies which hibernate as newly hatched larvae. The eggs are laid in autumn upon or near the food plants and the caterpillars gnaw their way out of the shells and seek seclusion at once, finding such shelter as they may in the materials on the soil surface. In spring they begin to feed as soon as the weather permits and complete their growth from then on.The half-grown caterpillars may hibernate either as freecreatures under boards, stones, or in the turfy grass, or they may be protected by special shelters which they have provided for themselves in their earlier life. In the case of the latter each may have a shelter of its own or there may be a common shelter for a colony of caterpillars. Among the examples of those hibernating in miscellaneous situations without special protection the caterpillars of the Tawny Emperor, the Gray Emperor, the Pearl Crescent, and some of the Graylings are examples. Among those which hibernate in individual shelters the Sovereigns, among which our common Viceroy is most familiar, are good examples. Among those which hibernate in a tent woven by the whole colony for the whole colony the Baltimore or Phaeton butterfly is perhaps the best example.The caterpillars that hibernate when full grown may be grouped in a way somewhat similar to those which are half grown. Many species simply find such shelter as they may at or near the soil surface. The Clouded Sulphur is a good example of these. Others pass the winter in individual shelters made from a leaf or blade. Several of the larger Skippers are good examples of this condition. So far as I know none of our species pass the winter in colonial shelters when full grown.It would be natural to suppose that the great majority of butterflies would be likely to hibernate in the chrysalis state. Here is a quiet stage in which the insect is unable to move about or to take any food, in which it seems entirely dormant and as a rule is fairly well hidden from the view of enemies. We find, however, that only a rather small proportion of our butterflies has chosen this stage for survival through the winter. The most conspicuousexamples are the Swallowtails, nearly all of which hibernate in the chrysalis stage. Other examples are the various Whites, the Orange-tips, and isolated species like the Wanderer, and the Spring Azure and the American Copper. Practically all the butterflies that pass the winter as chrysalids have a silken loop running around near the middle of the body which helps to hold them securely through the long winter months. Apparently none of those chrysalids which hang straight downward are able to survive the winter.An adult butterfly seems a fragile creature to endure the long cold months of arctic regions. Yet many of our most beautiful species habitually hibernate as adults, finding shelter in such situations as hollow trees, the crevices in rocks, the openings beneath loose bark or even the outer bark on the under side of a large branch. It is significant that most of the adult-wintering Angle-wings are northern rather than southern species, some of them being found in arctic regions practically around the world. One of the few southern forms that hibernates as an adult is the Goatweed Emperor.These examples are all cases of true hibernation in a lethargic condition. There are certain butterflies, however, which pass the winter as adults that remain active during this period. Obviously this is impossible in latitudes where the winter is severe, and it involves migration to a warmer climate. The one notable illustration of this is the Monarch butterfly which apparently flies southward to the Gulf states at least and there remains until spring, when individuals come north again. The southward migration may be begun in Canada when the butterflies gather together in enormous flocks that remind one of thegathering of the clans with the migrating birds. This is one of the least understood of insect activities but it has been observed so often and over so long a period of years that there seems to be no questioning the general habit.Like everything else in relation to living things there are numerous variations in the prevailing modes of hibernation. In the case of many species one can find combinations of two or more stages in which the winter is passed. Probably if we could observe with sufficient care we might be able to find somewhere examples of almost any conceivable double combination—as egg with larva or chrysalis or adult—the insect hibernating in two of these stages. Many examples are known in which both chrysalis and adult of the same species pass the winter and also of those in which young and well-grown larvae pass the winter. As one would expect, the conditions as to such combinations are likely to be more variable in southern than in northern regions.Notwithstanding all the attention which has been paid to butterfly life-histories there is still some uncertainty in regard to the hibernation of many of our species. One of the most interesting series of observations which a young naturalist could undertake would be to learn positively how each species of butterfly in his locality passes the winter.Aestivation or Summer LethargyIn some species of butterflies there is a special adaptation to passing through the hottest part of the summer season in a state of lethargy which is suggestive of the torpor of the hibernating period. This phase of butterfly existencehas not been extensively studied and there are indications that it exists more generally than has been commonly supposed. It has been noticed even in northern New England that some of the Angle-wings seek shelter and become lethargic during August. Apparently this is an adaptation to single broodedness, helping to carry the species through the year without the exhaustion incident to the continued activity of the butterfly.In more southern regions, especially in the hot, dry climates where vegetation withers in midsummer, it is well known that some caterpillars become lethargic, remaining inactive until the fall rains start vegetation into growth. The Orange-sulphur butterfly is a good example of this.This summer lethargy offers excellent opportunity for careful study. Any observer who finds a butterfly hidden away in summer under boards, the bark of a tree, or in a stone pile should look carefully to see what species it is and how the butterfly behaves. Such observations should be sent to the entomological journals in order that our knowledge of the subject may be increased.Feigning DeathThe fact has long been noticed that various butterflies have the habit at times of feigning death and dropping to the ground where they may lie motionless for a considerable period. This habit is most easily observed in some of the Angle-wings, especially those which hibernate as adults. Those species have the under surfaces of their wings colored in various bark-picturing patterns and apparently live through the winter to some extent, restingbeneath the bark of large branches or upon the trunks of trees. Many of them also secrete themselves in hollow trees or beneath loose bark or in board piles or stone walls. It is probable, however, that during the long ages when these insects were adapting themselves to their life conditions, before man interfered with the natural order and furnished various more or less artificial places for hibernation, these butterflies rested more generally upon the under side of branches than they do now.Even in warm weather when one of these butterflies is suddenly disturbed it is likely to fold its legs upon its body and drop to the ground, allowing itself to be handled without showing any signs of life. This habit is doubtless of value, especially during hibernation or possibly during the summer lethargy or aestivation, the latter a habit which may be more general among these butterflies than is now supposed. As the insect lies motionless upon the ground it is very likely to blend so thoroughly with its surroundings that it becomes concealed, and any bird which had startled it from the branch above would have difficulty in finding it.Some very interesting observations have been made upon the death-feigning instincts of various other insects, especially the beetles. But no one so far as I know has yet made an extended study of the subject in relation to our American butterflies. It is an excellent field for investigation and offers unusual opportunities for photographic records. One of the pictures oppositepage 32shows a photograph which I took of a Mourning Cloak as it was thus playing 'possum. This species exhibits the instinct to a marked degree.ColorationThe caterpillars of butterflies and moths form a large part of the food of insect-eating birds. These caterpillars are especially adapted for such a purpose and in the economy of nature they play a very important part in keeping alive the feathered tribes. During the long ages through which both birds and insects have been developing side by side, there have been many remarkable inter-relations established which tend on the one hand to prevent the birds from exterminating the insects and on the other to prevent the insects from causing the birds to starve. The most important of these, so far as the caterpillars are concerned, are the various devices by which these insects protect themselves from attack, by hiding away where birds are not likely to find them, by clothing their bodies with spiny hairs, by other methods of rendering themselves distasteful, or by various phases of concealing coloration. On the whole, the examples of the latter are not so numerous or so easily found in the case of the larvae of butterflies as in those of moths.Perhaps the basal principle of concealing coloration is the law of counter-shading, first partially announced by Prof. E. B. Poulton, and later much more elaborately worked out by Mr. Abbott H. Thayer, and discussed at length by Mr. Gerald H. Thayer in his remarkable volume, "Concealing Coloration in the Animal Kingdom." The law of counter-shading is tersely stated in these words: "Animals are painted by nature darkest on those parts which tend to be most lighted by the sky's light andvice versa." As this law works out on most animals that live on or near the ground, the upper part of the body exposedto the direct light from above is dark; and the under part, shut off from the upper light and receiving only the small reflection from below, is enough lighter to make the appearance of the creature in its natural environment of a uniform tone from back to breast.Nearly all caterpillars illustrate this law of counter-shading. If they are in the habit of feeding or resting with their feet downward the back will be darker and the under side lighter, but if they are in the habit of feeding or resting in the opposite position these color tones will be reversed. One can find examples of such conditions almost any summer's day by a little searching of trees or shrubs.This law of counter-shading, however, is really only the basis for the coloration of caterpillars or other animals. It tends, chiefly, to make the creature appear as a flat plane when seen from the side, and may be said in a way to prepare the canvas upon which Nature paints her more distinctive pictures. A great many examples of color markings that tend to conceal the caterpillar amid its natural surroundings may be found by any one who will study the subject and it offers one of the most interesting fields for investigation. The chapter on caterpillars in the above-mentioned book by Mr. Thayer should serve as a starting point for any one taking up the subject.Butterflies differ from caterpillars and from most other animals in the fact that their coloring is chiefly shown upon the flat surfaces of the wings. Consequently, there is less opportunity for the various phases of counter-shading which is so commonly shown in the larger caterpillars. The bodies of nearly all butterflies do exhibit this phenomenon, but these bodies are relatively so small that counter-shadingplays but a little part in the general display.Upon the outstretched membranes of the butterflies' wings Nature through the long ages of development has painted a great variety of pictures. Those which tend to protect the insect by concealment amid its surroundings are most commonly spread on the under surface of the wings. Especially is this true in the case of those species which pass the winter as adults or which have the habit of resting upon the bark of trees, the sides of rocks, or the surface of the ground. We here find some of the most interesting examples of obliterative coloring that occur in nature. Some butterflies have taken on the look of tree bark, others the sombre appearance of weathered rocks, while still others are painted with the images of flowerets and their stems.Dazzling and Eclipsing ColorsMany of the butterflies, especially the Angle-wings, which are marked on the under surface in various protective colors, are admirable examples of that phase of animal coloring which is spoken of asdazzling coloration. This is apparently one of the most important protective devices to be found in Nature and the validity of it is now generally conceded by naturalists. One phase of it, which may be called eclipsing coloration, seems to have been first definitely formulated by the late Lord Walsingham, a famous English entomologist who enunciated it in an address as president of the Entomological Society of London. The most significant paragraphs in that address were these:"My attention was lately drawn to a passage in Herbert Spencer's 'Essay on the Morals of Trade.' He writes: 'As when tasting different foods or wines the palate is disabledby something strongly flavored from appreciating the more delicate flavor of another thing afterward taken, so with the other organs of sense a temporary disability follows an excessive stimulation. This holds not only with the eyes in judging of colors, but also with the fingers in judging of texture.'"Here, I think, we have an explanation of the principle on which protection is undoubtedly afforded to certain insects by the possession of bright coloring on such parts of their wings or bodies as can be instantly covered and concealed at will. It is an undoubted fact, and one which must have been observed by nearly all collectors of insects abroad, and perhaps also in our own country, that it is more easy to follow with the eye the rapid movements of a more conspicuous insect soberly and uniformly colored than those of an insect capable of changing in an instant the appearance it presents. The eye, having once fixed itself upon an object of a certain form and color, conveys to the mind a corresponding impression, and, if that impression is suddenly found to be unreliable, the instruction which the mind conveys to the eye becomes also unreliable, and the rapidity with which the impression and consequent instruction can be changed cannot always compete successfully with the rapid transformation effected by the insect in its effort to escape."Lord Walsingham then goes on to suggest that this intermittent play of bright colors probably has as confusing effect upon birds and other predaceous vertebrates as upon man; and that on this hypothesis such colors can be accounted for more satisfactorily than upon any other yet suggested. Since then the significance of this theory has been repeatedly pointed out by Professor Poulton, Mr.Abbott H. Thayer, and various other authorities upon animal coloring. The termsdazzlingandeclipsinghave been applied to the phenomenon.Shortly after Lord Walsingham propounded this theory I called attention[A]to its fitness in explaining some of the most interesting color phases shown by American insects, notably the moths and locusts which have brilliantly colored under wings and protectively colored upper wings.[A]Popular Science Monthly, 1898, "A Game of Hide and Seek." Reprinted in theInsect World, 1899.The animals of the north show numberless color phases of interest. One of the most curious of these is exhibited by several families of insects in which the outer wings are protectively colored in dull hues and the under wings brightly colored. For example, there are many species of moths belonging to the genus Catocala found throughout the United States. These are insects of good size, the larger ones measuring three inches in expanse of wings, and the majority of them being at least two thirds that size. Most of them live during the day on the bark of trees, with their front wings folded together over the back. The colors and markings of these wings, as well as of the rest of the exposed portions of the body, are such as to assimilate closely with the bark of the tree upon which the insect rests. In such a situation it requires a sharp eye to detect the presence of the moth, which, unless disturbed, flies only at night, remaining all day exposed to the attacks of many enemies. Probably the most important of these are the birds, especially species like the woodpeckers, which are constantly exploring all portions of the trunks of trees.The chief beauty of these Catocalas, as they are seenspread out in the museum cabinet, lies in the fact that the hind wings, which, when the moth is at rest in life, are concealed by the front ones, are brightly colored in contrasting hues of black, red, and white in various brilliant combinations. These colors, in connection with the soft and blended tones of the front wings, make a very handsome insect.It is easy to see that when one of these Underwing Moths is driven to flight by a woodpecker or other bark-searching bird it would show during its rapid, irregular flight the bright colors of the under wings which would be instantly hidden upon alighting and the very different coloring of the upper wings blending with the bark would be substituted. Consequently, the bird would be very likely to be baffled in its pursuit.Coloration of LocustsOn the rocky hills and sandy plains of New England there are several species of grasshoppers or locusts that also illustrate these principles. If you walk along a strip of sandy land in summer, you start to flight certain locusts which soon alight, and when searched for will be found closely to assimilate in color the sand upon which they rest. On a neighboring granite-ribbed hill you will find few if any of this species of locust, but instead there occur two or three quite different species, which when at rest closely resemble the lichen-covered rocks. This resemblance is very striking, and is found in all stages of the insect's existence. If now you go to a lowland meadow, still another color phase will be found to prevail—the green grass is swarming with the so-called "long-horned" grasshoppers,which are green throughout with linear bodies, and long, slender legs and antennae.Each of these three groups of insects is adapted to its particular habitat. All are constantly persecuted by birds, and have been so persecuted for unnumbered ages in the past. In every generation the individuals have varied, some toward closer resemblance to environment, others in an opposite direction. The more conspicuous insects have been constantly taken, and the least conspicuous as constantly left to reproduce. Were the three groups to change places to-day, the green grasshoppers from the meadows going to sandy surfaces, the sand-colored locusts going to rocky hills, and the "mossbacks" from the hills to the lowland meadows, each would become conspicuous, and the birds would have such a feast as is seldom spread before them.The species living on sand and rocks are often "flushed" by birds. Those which flew but a few feet would be likely to be captured by the pursuing bird; those which flew farther would stand a better chance of escaping. Similarly, those which flew slowly and in a straight line would be more likely to be caught than those which flew rapidly and took a zigzag course. As a consequence of the selection thus brought about through the elimination of those which flew slowly along the straight and narrow way that led to death, you will find that most locusts living in exposed situations when startled fly some distance in a rapid, zigzag manner.But still another element of safety has been introduced by some species of these locusts through the adoption of the color tactics of the Catocala moths. The under wings of the common Carolina locust—the species most abundantalong the highway—are black, bordered with yellowish white. The base of the hind wings of a related species living on the Western plains is bluish, while in the large coral-winged locust of the Eastern states the hind wings are red, bordered with black. In nearly all of the species of these locusts frequenting open localities where they are liable to disturbance by birds or other animals, the hind wings exhibit contrasting colors in flight. Most of them also fly in a zigzag line, and alight in a most erratic manner. Many times I have had difficulty in determining the exact landfall of one of these peculiar creatures, and I believe Lord Walsingham's suggestion is well exemplified in them. (Seepage 33.)The most famous example of a combination of this dazzling coloring of the upper wing surface with a definite protective coloring of the under wing surface is the Kallima butterfly which is illustrated in almost every book dealing with animal coloration. The under wing surface bears a striking resemblance to a leaf and the hind wings project to form a tail which looks like the petiole of the leaf, and there is also a mark running across the wings which mimics the midrib. When the butterfly is flying the brilliant colors of the upper surface are visible, but when it alights these are instantly replaced by the sombre tone of the under surface, so that apparently the insect completely disappears and in its place there is only a leaf attached to a branch in a most natural position. In Dr. Longstaff's book there is an illustration of another tropical butterfly,Eronia cleodora, which resembles on its under surface a yellow disease-stricken leaf but on its upper surface gives a brilliant combination of black and white. This insect alights upon the leaves which it resemblesand is a striking example of both dazzling and mimicking coloration.Many of our own butterflies, notably the Angle-wings, are excellent examples of a similar combination. In flight they reveal conspicuous colors which are instantly hidden upon alighting and then one only sees the bark-like or dead leaf-like under surface as may be seen in the plate oppositepages 160-161. The iridescence upon the upper wing surface of many butterflies, whose under wing surface is colored in concealing tones, is doubtless also of great use to the insect in a similar way. There is a splendid opportunity here for some observer to study this phase of butterfly activity and to get photographs of the insects amid their natural surroundings.In their book upon "Concealing Coloration" the Messrs. Thayer have called attention to many interesting phases of dazzling coloration. They show that bright marks like the eye-spots or ocelli, which form so prominent a feature on the wing surfaces of many butterflies, really helped to conceal the insect amid its natural surroundings, by drawing the eye away from the outlines of wings and body so that the latter tend to disappear. Their discussion of this subject opens up another vast field for outdoor observations of absorbing interest, in which there is great need for many active workers.Seepage 23BUTTERFLY FEIGNING DEATH, HANGING TO BARK BY ONE FOOTSeepage 17BUTTERFLY IN HIBERNATING POSITIONCaterpillar feeding upon leaf of milkweedCaterpillar hung up for the change to the chrysalisThe transition stageThe ChrysalisPhotographs from life.See pages8-10,233THE MONARCH BUTTERFLY: CHANGE FROM CATERPILLAR TO CHRYSALIS.Chrysalis showing butterfly ready to emergeThe empty chrysalisButterfly just out of chrysalisSide view a little laterPhotographs from life.See pages10-13,235THE MONARCH BUTTERFLY: THE CHANGE FROM CHRYSALIS TO ADULT.Seepage 30THE CAROLINA LOCUSTAbove, with wings expanded as in flightSelective Color SenseOne who collects the Underwing moths soon discovers that the light colored species which resemble the bark of birch trees are likely to be found upon the trunks of those trees, and that the dark colored kinds which resemble the bark of maple trees are likely to be found upon the trunks of these. Obviously, were this not true the protective coloring would avail but little and it is evident that these moths are able to select a background which is of advantage in helping to conceal them.There is much evidence to show that in a similar way the butterflies are able by means of a well-developed color sense to select the places where they alight. One of the most notable examples is that of a South American species,Peridromia feronia. This is a silvery gray butterfly which alights head downward upon the bark of certain palm trees that have silvery gray stems and remains there with its wings fully expanded so that it utilizes the background in much the same way that the Underwing moths do. "When disturbed they will return to the same tree again and again."One who will observe the habits of our Angle-wings and other butterflies which have obliterative coloring of the under wing surface can easily learn that these insects select rather carefully the places where they alight. It will be found that as a rule each species utilizes a background that blends with its own coloring. It is probable that this habit is much more common in other groups of butterflies than has been realized. Much evidence of this sort has been collected regarding the butterflies of Europe and other countries, as well as near our own borders in America.Warning Coloration and MimicryThe colors of a great many animals, including a considerable percentage of American butterflies and their larvae, have been commonly explained by the theory ofwarning colors. According to this theory animals whichwere for any reason not edible by birds and mammals have developed various striking combinations of color such as black and yellow, red and black, or black and white, in order to advertise to their foes their inedible qualities. This theory has been very generally accepted by naturalists and will be found expounded at length in many books published during the last quarter century.The whole subject of the validity of warning coloration has recently been brought up for reconsideration by the illuminating investigations of Mr. Abbott H. Thayer and discussed at length in the book upon "Concealing Coloration" already mentioned. In an appendix to this book dated 1908 Mr. Thayer states that he no longer holds the belief that "there must somewhere be warning colors." He has convincingly shown that a large proportion of the animals which were supposed to be examples of this theory are really illustrations of concealing coloration. But there yet remain various facts which have been conclusively proven that apparently require the theory of warning colors to explain them. Here is another field in which there is a real need for much careful investigation under conditions that are rigidly scientific.Along with the theory of warning coloration the theory ofmimicryhas been propounded. According to this if a butterfly in a given region shows warning coloration, having developed such coloration because it is distasteful to birds and mammals, it may be mimicked by another butterfly in the same region belonging to another group, the latter butterfly being edible, but benefiting by its resemblance to the distasteful species, because birds or mammals mistake it for the latter and do not attempt to catch it. The most notable example of such mimicry inNorth America is that of the Monarch butterfly, which is supposed to be the distasteful species, and the Viceroy butterfly, which is supposed to mimic it. Several other instances of mimicry are found among our own butterflies, while in South America, Africa, and Asia there are numberless examples.HeliotropismIt has long been known that the green surfaces of plants respond to the stimulus of the sun's rays in a most remarkable manner. This response has commonly been calledheliotropismand it has been carefully studied by botanists all over the habitable world. More recently, the fact has been observed that many animals respond in certain definite ways to the stimulus of direct sunshine and the same term has been applied in this case. Very little attention has been given to the subject of heliotropism until within a few recent years. But the observations which have been made by Parker, Longstaff, Dixey, and others open up a most interesting field for further observation. An admirable summary of our present knowledge of the subject has been published by Dr. Longstaff in his book "Butterfly Hunting in Many Lands."One of the earliest observations upon this subject was that published in my book "Nature Biographies" which appeared in June, 1901, concerning the habit in the Mourning Cloak: "On a spring-like day early in November (the 8th) I came across one of these butterflies basking in the sunshine upon the ties of a railway track. It rested with its wings wide open. On being disturbed, it would fly a short distance and then alight, and I was interested to noticethat after alighting it would always turn about until the hind end of its body pointed in the direction of the sun, so that the sun's rays struck its wings and body nearly at right angles. I repeatedly observed this habit of getting into the position in which the most benefit from the sunshine was received, and it is of interest as showing the extreme delicacy of perception toward the warmth of sunshine which these creatures possess."A little later, some very elaborate observations were made upon this habit of the Mourning Cloak by Prof. G. H. Parker of Harvard University. Professor Parker noticed that during the warm spells in winter the butterflies came out of their hiding places and after alighting, always placed themselves with their heads away from the direction of the sun and their bodies lying nearly at right angles to the sun's rays. By experiment, he found that they adjusted themselves to this position as soon as they were fully exposed to direct sunshine, even if at the time of alighting they were in a shadow. He found that this movement was a reflex action through the eyes, for when the eyes were blinded no such adjustment took place. He called itnegative heliotropism.Dr. Longstaff uses the termorientationfor this adjustment of the butterfly to the sun's rays and he finds it is a very general habit, especially with the Angle-wings, for the butterfly thus to orient itself after alighting, in such a way that the hind end of the body points toward the sun. This occurs not only with those species which keep their wings spread open when they alight but also with those in which the wings are closed together and held in a vertical position on alighting.Various explanations of this phenomenon have beenoffered but apparently none of them are yet generally accepted. Were the habit confined to butterflies like the Mourning Cloak, it would seem easy to prove that a main advantage was found in the benefit derived from the heat rays of the sun. Were it confined to those species which always fold their wings on alighting, it would seem easy to believe that it was a device for reducing the shadow cast by the insect to its lowest terms. It has also been suggested that the habit is for the purpose of revealing to the fullest extent the markings of the butterfly. Evidently there is here an ample field for further investigation before definite conclusions are reached.
Butterflies and Moths
Butterfly wing scales, magnified.(From Holland)
Butterfly wing scales, magnified.(From Holland)
The butterflies and moths both belong to the great order of scale-winged insects—the Lepidoptera. They are distinguished, however, by certain general characteristics, which hold true for the most part in both groups. The butterflies fly by day; the moths fly by night. All of the higher butterflies go into the chrysalis state without making a silken cocoon, while most of the higher moths make such a cocoon. The bodies of the butterflies are usually slender, while those of the larger moths are stout. The antennae of the butterflies are generally slender and commonly enlarged at the tip into a miniature club. The antennae of the larger moths are commonly feathery or are long and slender, tapering gradually toward the tip.
The characteristic features that distinguish a moth froma butterfly are well illustrated in the plate opposite page 49, which shows one of the largest and most beautiful moths in the world. It is the Cynthia moth. As may be seen, the newly emerged moth is resting upon the silken cocoon in which it spent its period as a pupa or chrysalis. This cocoon was attached by the caterpillar to the twig from which it hangs at the time it spun the cocoon. The feathered antennae, the hairy legs, the thick thorax, and large abdomen—all show very clearly in this side view of the moth. As will be seen, the wings are large and very suggestive of those of a butterfly and have the characteristic eye-spots toward the tip and the crescent marks in the middle, which are so often found on the wings of the larger moths.
Some of these large moths on cloudy days occasionally fly during daylight and, by the uninitiated, they are often mistaken for large butterflies. One who will notice their structure, however, will readily see the characteristic features of the moth.
In the caterpillar stage, there are no hard and fast differences between the larvae of butterflies and those of the higher moths. In each case, the insect consists of a worm-like body, having a small head provided with biting jaws and simple eyes or ocelli. Back of the head are the three rings of the thorax, each of which bears a pair of jointed legs. Back of these three rings there are a considerable number of other body rings making up the abdomen, on the middle of which there are commonly four or five pairs of fleshy prolegs, not jointed but furnished at the tip with fine claws. At the hind end of the body there is another pair of prolegs similar in structure.
The Scents of Butterflies
Androconia from wings of male butterflies
Androconia from wings of male butterflies
Many students of American butterflies have occasionally mentioned the fact that certain species seem to give off a distinct scent which has frequently been spoken of as a pleasing fragrance, suggesting sandalwood or some other aromatic odor. The general subject as exemplified by butterflies of other lands has been studied for many years by Fritz Müller; and certain English entomologists have paid considerable attention to it. A translation of the Müller publications and an excellent summary of our present knowledge of the subject is published in Dr. Longstaff's book on butterfly hunting.
The odors given off by butterflies are divided into two principal kinds, namely: first, those which are repulsive to the senses of man, and evidently for the purpose of protecting the butterflies from birds and other vertebrate enemies—these are found in both sexes; second, those which are evidently for the purpose of sexual attraction and confined to the male butterflies—these scents are usually attractive to the senses of man.
The aromatic scents of the second group are generally produced by means of certain scales or hairs of many curious forms, which are scattered over the surface of the wings or are placed within certain special pockets, generally near the borders of the wings. These scales or hairs are called androconia. Some of them much magnified are represented in the picture above.
Our knowledge of the scents given off by Americanbutterflies is very fragmentary, and it is highly desirable that many more observations should be made upon the subject. If collectors generally would make careful notes, both in the field and upon the freshly killed butterflies at home, we ought soon to be able greatly to extend our knowledge. By holding the butterfly with a pair of forceps, one can often determine whether the fragrance is emitted. It is often helpful also to brush the hairs or tufts where the androconia are attached, using a small, dry camel's hair brush for the purpose.
Butterfly Migrations
Migration seems to be a general instinct in the animal world, developed when a species becomes enormously abundant. At such times this instinct apparently overcomes all others and the creatures move on regardless of obstacles and conditions that may mean certain death to the vast majority. Such migrations among mammals have often been recorded, one of the most notable examples being that of the little lemmings which migrate at periodical intervals in a way which has often been described. Among the insects such migrations have been frequently noticed, and the phenomenon has apparently been observed oftener among the butterflies than in any other group. Entomological literature during the last hundred years contains a great many records of enormous flights of butterflies over long distances, extending even from Africa into Europe or from one part of America to another far remote. As such migration is likely to happen whenever a species becomes extremely abundant it probably is Nature's way of providing for an extended food supply for the succeeding generations. That it results in the death of the great majority of the migrants is doubtless true, but it must lead to vast experiments in extending the geographic area inhabited by these species. Numerous examples of such migrating swarms will be found in the pages of this little book. (See plates, pages17,48-49,160.)
Photographed from life by A. H. VerrillSeepage 12SWALLOWTAIL BUTTERFLY JUST OUT OF ITS CHRYSALIS.THE CHRYSALIS SKIN BELOW
Photographed from life by A. H. Verrill
Seepage 12
SWALLOWTAIL BUTTERFLY JUST OUT OF ITS CHRYSALIS.THE CHRYSALIS SKIN BELOW
Seepage 72TIGER SWALLOWTAIL
Seepage 72
TIGER SWALLOWTAIL
Photographed from lifeSeepage 7THE HAMMOCK CATERPILLAR ON ITS HAMMOCK
Photographed from life
Seepage 7
THE HAMMOCK CATERPILLAR ON ITS HAMMOCK
The migrations thus considered are only exceptionaloccurrences. There is, however, a regularly recurring annual migration on the part of some butterflies which is also a phenomenon of extraordinary interest. The most notable example is that of the Monarch which apparently follows the birds southward every autumn and comes northward again in spring. There is much evidence to indicate that in some slight degree other butterflies have a similar habit, although the present observations are inadequate to determine to what extent this habit has become fixed in most of these species.
Hibernation or Winter Lethargy
The ways in which butterflies spend the winter are always of peculiar interest to the naturalist. Here are creatures with four distinct stages of existence, each of which has the possibility of carrying the species through the season of cold. It is necessary to learn for each insect which stage has been chosen for the purpose, and if possible to find the reasons for the choice.
As a rule the related members of a group are likely to hibernate in a similar stage. Thus most of the Swallowtails pass the winter as chrysalids while practically all the Angle-wings pass the winter as adults. This rule, however,has many exceptions, for one will often find closely related species which differ in the stage of hibernation.
As one would expect, the conditions of hibernation vary greatly with the latitude. In the severe climate of the far north the conditions are likely to be more uniform than in the South where the milder climate permits greater variation to the insect. In some cases where a butterfly hibernates in only one stage in Canada it may pass the winter in two or more stages in Alabama or Florida.
In many other orders of insects the egg is a favorite stage for hibernation. Even in the closely related moths it is often chosen by many species, but comparatively few butterflies pass the winter in the egg stage. The little Bronze Copper may serve as one example of this limited group.
The conditions as to hibernation by the larvae of butterflies are very different from those of the egg. It has been estimated that probably half of all our species pass the winter in some stage of caterpillar growth. This varies all the way from the newly hatched caterpillar which hibernates without tasting food to the fully grown caterpillar which hibernates full fed and changes to a chrysalis in spring without eating anything at that time. A large proportion, however, feed both in fall and spring, going through the winter when approximately half grown.
The Graylings and the Fritillaries are typical examples of butterflies which hibernate as newly hatched larvae. The eggs are laid in autumn upon or near the food plants and the caterpillars gnaw their way out of the shells and seek seclusion at once, finding such shelter as they may in the materials on the soil surface. In spring they begin to feed as soon as the weather permits and complete their growth from then on.
The half-grown caterpillars may hibernate either as freecreatures under boards, stones, or in the turfy grass, or they may be protected by special shelters which they have provided for themselves in their earlier life. In the case of the latter each may have a shelter of its own or there may be a common shelter for a colony of caterpillars. Among the examples of those hibernating in miscellaneous situations without special protection the caterpillars of the Tawny Emperor, the Gray Emperor, the Pearl Crescent, and some of the Graylings are examples. Among those which hibernate in individual shelters the Sovereigns, among which our common Viceroy is most familiar, are good examples. Among those which hibernate in a tent woven by the whole colony for the whole colony the Baltimore or Phaeton butterfly is perhaps the best example.
The caterpillars that hibernate when full grown may be grouped in a way somewhat similar to those which are half grown. Many species simply find such shelter as they may at or near the soil surface. The Clouded Sulphur is a good example of these. Others pass the winter in individual shelters made from a leaf or blade. Several of the larger Skippers are good examples of this condition. So far as I know none of our species pass the winter in colonial shelters when full grown.
It would be natural to suppose that the great majority of butterflies would be likely to hibernate in the chrysalis state. Here is a quiet stage in which the insect is unable to move about or to take any food, in which it seems entirely dormant and as a rule is fairly well hidden from the view of enemies. We find, however, that only a rather small proportion of our butterflies has chosen this stage for survival through the winter. The most conspicuousexamples are the Swallowtails, nearly all of which hibernate in the chrysalis stage. Other examples are the various Whites, the Orange-tips, and isolated species like the Wanderer, and the Spring Azure and the American Copper. Practically all the butterflies that pass the winter as chrysalids have a silken loop running around near the middle of the body which helps to hold them securely through the long winter months. Apparently none of those chrysalids which hang straight downward are able to survive the winter.
An adult butterfly seems a fragile creature to endure the long cold months of arctic regions. Yet many of our most beautiful species habitually hibernate as adults, finding shelter in such situations as hollow trees, the crevices in rocks, the openings beneath loose bark or even the outer bark on the under side of a large branch. It is significant that most of the adult-wintering Angle-wings are northern rather than southern species, some of them being found in arctic regions practically around the world. One of the few southern forms that hibernates as an adult is the Goatweed Emperor.
These examples are all cases of true hibernation in a lethargic condition. There are certain butterflies, however, which pass the winter as adults that remain active during this period. Obviously this is impossible in latitudes where the winter is severe, and it involves migration to a warmer climate. The one notable illustration of this is the Monarch butterfly which apparently flies southward to the Gulf states at least and there remains until spring, when individuals come north again. The southward migration may be begun in Canada when the butterflies gather together in enormous flocks that remind one of thegathering of the clans with the migrating birds. This is one of the least understood of insect activities but it has been observed so often and over so long a period of years that there seems to be no questioning the general habit.
Like everything else in relation to living things there are numerous variations in the prevailing modes of hibernation. In the case of many species one can find combinations of two or more stages in which the winter is passed. Probably if we could observe with sufficient care we might be able to find somewhere examples of almost any conceivable double combination—as egg with larva or chrysalis or adult—the insect hibernating in two of these stages. Many examples are known in which both chrysalis and adult of the same species pass the winter and also of those in which young and well-grown larvae pass the winter. As one would expect, the conditions as to such combinations are likely to be more variable in southern than in northern regions.
Notwithstanding all the attention which has been paid to butterfly life-histories there is still some uncertainty in regard to the hibernation of many of our species. One of the most interesting series of observations which a young naturalist could undertake would be to learn positively how each species of butterfly in his locality passes the winter.
Aestivation or Summer Lethargy
In some species of butterflies there is a special adaptation to passing through the hottest part of the summer season in a state of lethargy which is suggestive of the torpor of the hibernating period. This phase of butterfly existencehas not been extensively studied and there are indications that it exists more generally than has been commonly supposed. It has been noticed even in northern New England that some of the Angle-wings seek shelter and become lethargic during August. Apparently this is an adaptation to single broodedness, helping to carry the species through the year without the exhaustion incident to the continued activity of the butterfly.
In more southern regions, especially in the hot, dry climates where vegetation withers in midsummer, it is well known that some caterpillars become lethargic, remaining inactive until the fall rains start vegetation into growth. The Orange-sulphur butterfly is a good example of this.
This summer lethargy offers excellent opportunity for careful study. Any observer who finds a butterfly hidden away in summer under boards, the bark of a tree, or in a stone pile should look carefully to see what species it is and how the butterfly behaves. Such observations should be sent to the entomological journals in order that our knowledge of the subject may be increased.
Feigning Death
The fact has long been noticed that various butterflies have the habit at times of feigning death and dropping to the ground where they may lie motionless for a considerable period. This habit is most easily observed in some of the Angle-wings, especially those which hibernate as adults. Those species have the under surfaces of their wings colored in various bark-picturing patterns and apparently live through the winter to some extent, restingbeneath the bark of large branches or upon the trunks of trees. Many of them also secrete themselves in hollow trees or beneath loose bark or in board piles or stone walls. It is probable, however, that during the long ages when these insects were adapting themselves to their life conditions, before man interfered with the natural order and furnished various more or less artificial places for hibernation, these butterflies rested more generally upon the under side of branches than they do now.
Even in warm weather when one of these butterflies is suddenly disturbed it is likely to fold its legs upon its body and drop to the ground, allowing itself to be handled without showing any signs of life. This habit is doubtless of value, especially during hibernation or possibly during the summer lethargy or aestivation, the latter a habit which may be more general among these butterflies than is now supposed. As the insect lies motionless upon the ground it is very likely to blend so thoroughly with its surroundings that it becomes concealed, and any bird which had startled it from the branch above would have difficulty in finding it.
Some very interesting observations have been made upon the death-feigning instincts of various other insects, especially the beetles. But no one so far as I know has yet made an extended study of the subject in relation to our American butterflies. It is an excellent field for investigation and offers unusual opportunities for photographic records. One of the pictures oppositepage 32shows a photograph which I took of a Mourning Cloak as it was thus playing 'possum. This species exhibits the instinct to a marked degree.
Coloration
The caterpillars of butterflies and moths form a large part of the food of insect-eating birds. These caterpillars are especially adapted for such a purpose and in the economy of nature they play a very important part in keeping alive the feathered tribes. During the long ages through which both birds and insects have been developing side by side, there have been many remarkable inter-relations established which tend on the one hand to prevent the birds from exterminating the insects and on the other to prevent the insects from causing the birds to starve. The most important of these, so far as the caterpillars are concerned, are the various devices by which these insects protect themselves from attack, by hiding away where birds are not likely to find them, by clothing their bodies with spiny hairs, by other methods of rendering themselves distasteful, or by various phases of concealing coloration. On the whole, the examples of the latter are not so numerous or so easily found in the case of the larvae of butterflies as in those of moths.
Perhaps the basal principle of concealing coloration is the law of counter-shading, first partially announced by Prof. E. B. Poulton, and later much more elaborately worked out by Mr. Abbott H. Thayer, and discussed at length by Mr. Gerald H. Thayer in his remarkable volume, "Concealing Coloration in the Animal Kingdom." The law of counter-shading is tersely stated in these words: "Animals are painted by nature darkest on those parts which tend to be most lighted by the sky's light andvice versa." As this law works out on most animals that live on or near the ground, the upper part of the body exposedto the direct light from above is dark; and the under part, shut off from the upper light and receiving only the small reflection from below, is enough lighter to make the appearance of the creature in its natural environment of a uniform tone from back to breast.
Nearly all caterpillars illustrate this law of counter-shading. If they are in the habit of feeding or resting with their feet downward the back will be darker and the under side lighter, but if they are in the habit of feeding or resting in the opposite position these color tones will be reversed. One can find examples of such conditions almost any summer's day by a little searching of trees or shrubs.
This law of counter-shading, however, is really only the basis for the coloration of caterpillars or other animals. It tends, chiefly, to make the creature appear as a flat plane when seen from the side, and may be said in a way to prepare the canvas upon which Nature paints her more distinctive pictures. A great many examples of color markings that tend to conceal the caterpillar amid its natural surroundings may be found by any one who will study the subject and it offers one of the most interesting fields for investigation. The chapter on caterpillars in the above-mentioned book by Mr. Thayer should serve as a starting point for any one taking up the subject.
Butterflies differ from caterpillars and from most other animals in the fact that their coloring is chiefly shown upon the flat surfaces of the wings. Consequently, there is less opportunity for the various phases of counter-shading which is so commonly shown in the larger caterpillars. The bodies of nearly all butterflies do exhibit this phenomenon, but these bodies are relatively so small that counter-shadingplays but a little part in the general display.
Upon the outstretched membranes of the butterflies' wings Nature through the long ages of development has painted a great variety of pictures. Those which tend to protect the insect by concealment amid its surroundings are most commonly spread on the under surface of the wings. Especially is this true in the case of those species which pass the winter as adults or which have the habit of resting upon the bark of trees, the sides of rocks, or the surface of the ground. We here find some of the most interesting examples of obliterative coloring that occur in nature. Some butterflies have taken on the look of tree bark, others the sombre appearance of weathered rocks, while still others are painted with the images of flowerets and their stems.
Dazzling and Eclipsing Colors
Many of the butterflies, especially the Angle-wings, which are marked on the under surface in various protective colors, are admirable examples of that phase of animal coloring which is spoken of asdazzling coloration. This is apparently one of the most important protective devices to be found in Nature and the validity of it is now generally conceded by naturalists. One phase of it, which may be called eclipsing coloration, seems to have been first definitely formulated by the late Lord Walsingham, a famous English entomologist who enunciated it in an address as president of the Entomological Society of London. The most significant paragraphs in that address were these:
"My attention was lately drawn to a passage in Herbert Spencer's 'Essay on the Morals of Trade.' He writes: 'As when tasting different foods or wines the palate is disabledby something strongly flavored from appreciating the more delicate flavor of another thing afterward taken, so with the other organs of sense a temporary disability follows an excessive stimulation. This holds not only with the eyes in judging of colors, but also with the fingers in judging of texture.'
"Here, I think, we have an explanation of the principle on which protection is undoubtedly afforded to certain insects by the possession of bright coloring on such parts of their wings or bodies as can be instantly covered and concealed at will. It is an undoubted fact, and one which must have been observed by nearly all collectors of insects abroad, and perhaps also in our own country, that it is more easy to follow with the eye the rapid movements of a more conspicuous insect soberly and uniformly colored than those of an insect capable of changing in an instant the appearance it presents. The eye, having once fixed itself upon an object of a certain form and color, conveys to the mind a corresponding impression, and, if that impression is suddenly found to be unreliable, the instruction which the mind conveys to the eye becomes also unreliable, and the rapidity with which the impression and consequent instruction can be changed cannot always compete successfully with the rapid transformation effected by the insect in its effort to escape."
Lord Walsingham then goes on to suggest that this intermittent play of bright colors probably has as confusing effect upon birds and other predaceous vertebrates as upon man; and that on this hypothesis such colors can be accounted for more satisfactorily than upon any other yet suggested. Since then the significance of this theory has been repeatedly pointed out by Professor Poulton, Mr.Abbott H. Thayer, and various other authorities upon animal coloring. The termsdazzlingandeclipsinghave been applied to the phenomenon.
Shortly after Lord Walsingham propounded this theory I called attention[A]to its fitness in explaining some of the most interesting color phases shown by American insects, notably the moths and locusts which have brilliantly colored under wings and protectively colored upper wings.
[A]Popular Science Monthly, 1898, "A Game of Hide and Seek." Reprinted in theInsect World, 1899.
[A]Popular Science Monthly, 1898, "A Game of Hide and Seek." Reprinted in theInsect World, 1899.
[A]Popular Science Monthly, 1898, "A Game of Hide and Seek." Reprinted in theInsect World, 1899.
The animals of the north show numberless color phases of interest. One of the most curious of these is exhibited by several families of insects in which the outer wings are protectively colored in dull hues and the under wings brightly colored. For example, there are many species of moths belonging to the genus Catocala found throughout the United States. These are insects of good size, the larger ones measuring three inches in expanse of wings, and the majority of them being at least two thirds that size. Most of them live during the day on the bark of trees, with their front wings folded together over the back. The colors and markings of these wings, as well as of the rest of the exposed portions of the body, are such as to assimilate closely with the bark of the tree upon which the insect rests. In such a situation it requires a sharp eye to detect the presence of the moth, which, unless disturbed, flies only at night, remaining all day exposed to the attacks of many enemies. Probably the most important of these are the birds, especially species like the woodpeckers, which are constantly exploring all portions of the trunks of trees.
The chief beauty of these Catocalas, as they are seenspread out in the museum cabinet, lies in the fact that the hind wings, which, when the moth is at rest in life, are concealed by the front ones, are brightly colored in contrasting hues of black, red, and white in various brilliant combinations. These colors, in connection with the soft and blended tones of the front wings, make a very handsome insect.
It is easy to see that when one of these Underwing Moths is driven to flight by a woodpecker or other bark-searching bird it would show during its rapid, irregular flight the bright colors of the under wings which would be instantly hidden upon alighting and the very different coloring of the upper wings blending with the bark would be substituted. Consequently, the bird would be very likely to be baffled in its pursuit.
Coloration of Locusts
On the rocky hills and sandy plains of New England there are several species of grasshoppers or locusts that also illustrate these principles. If you walk along a strip of sandy land in summer, you start to flight certain locusts which soon alight, and when searched for will be found closely to assimilate in color the sand upon which they rest. On a neighboring granite-ribbed hill you will find few if any of this species of locust, but instead there occur two or three quite different species, which when at rest closely resemble the lichen-covered rocks. This resemblance is very striking, and is found in all stages of the insect's existence. If now you go to a lowland meadow, still another color phase will be found to prevail—the green grass is swarming with the so-called "long-horned" grasshoppers,which are green throughout with linear bodies, and long, slender legs and antennae.
Each of these three groups of insects is adapted to its particular habitat. All are constantly persecuted by birds, and have been so persecuted for unnumbered ages in the past. In every generation the individuals have varied, some toward closer resemblance to environment, others in an opposite direction. The more conspicuous insects have been constantly taken, and the least conspicuous as constantly left to reproduce. Were the three groups to change places to-day, the green grasshoppers from the meadows going to sandy surfaces, the sand-colored locusts going to rocky hills, and the "mossbacks" from the hills to the lowland meadows, each would become conspicuous, and the birds would have such a feast as is seldom spread before them.
The species living on sand and rocks are often "flushed" by birds. Those which flew but a few feet would be likely to be captured by the pursuing bird; those which flew farther would stand a better chance of escaping. Similarly, those which flew slowly and in a straight line would be more likely to be caught than those which flew rapidly and took a zigzag course. As a consequence of the selection thus brought about through the elimination of those which flew slowly along the straight and narrow way that led to death, you will find that most locusts living in exposed situations when startled fly some distance in a rapid, zigzag manner.
But still another element of safety has been introduced by some species of these locusts through the adoption of the color tactics of the Catocala moths. The under wings of the common Carolina locust—the species most abundantalong the highway—are black, bordered with yellowish white. The base of the hind wings of a related species living on the Western plains is bluish, while in the large coral-winged locust of the Eastern states the hind wings are red, bordered with black. In nearly all of the species of these locusts frequenting open localities where they are liable to disturbance by birds or other animals, the hind wings exhibit contrasting colors in flight. Most of them also fly in a zigzag line, and alight in a most erratic manner. Many times I have had difficulty in determining the exact landfall of one of these peculiar creatures, and I believe Lord Walsingham's suggestion is well exemplified in them. (Seepage 33.)
The most famous example of a combination of this dazzling coloring of the upper wing surface with a definite protective coloring of the under wing surface is the Kallima butterfly which is illustrated in almost every book dealing with animal coloration. The under wing surface bears a striking resemblance to a leaf and the hind wings project to form a tail which looks like the petiole of the leaf, and there is also a mark running across the wings which mimics the midrib. When the butterfly is flying the brilliant colors of the upper surface are visible, but when it alights these are instantly replaced by the sombre tone of the under surface, so that apparently the insect completely disappears and in its place there is only a leaf attached to a branch in a most natural position. In Dr. Longstaff's book there is an illustration of another tropical butterfly,Eronia cleodora, which resembles on its under surface a yellow disease-stricken leaf but on its upper surface gives a brilliant combination of black and white. This insect alights upon the leaves which it resemblesand is a striking example of both dazzling and mimicking coloration.
Many of our own butterflies, notably the Angle-wings, are excellent examples of a similar combination. In flight they reveal conspicuous colors which are instantly hidden upon alighting and then one only sees the bark-like or dead leaf-like under surface as may be seen in the plate oppositepages 160-161. The iridescence upon the upper wing surface of many butterflies, whose under wing surface is colored in concealing tones, is doubtless also of great use to the insect in a similar way. There is a splendid opportunity here for some observer to study this phase of butterfly activity and to get photographs of the insects amid their natural surroundings.
In their book upon "Concealing Coloration" the Messrs. Thayer have called attention to many interesting phases of dazzling coloration. They show that bright marks like the eye-spots or ocelli, which form so prominent a feature on the wing surfaces of many butterflies, really helped to conceal the insect amid its natural surroundings, by drawing the eye away from the outlines of wings and body so that the latter tend to disappear. Their discussion of this subject opens up another vast field for outdoor observations of absorbing interest, in which there is great need for many active workers.
Seepage 23BUTTERFLY FEIGNING DEATH, HANGING TO BARK BY ONE FOOT
Seepage 23BUTTERFLY FEIGNING DEATH, HANGING TO BARK BY ONE FOOT
Seepage 23
Seepage 17BUTTERFLY IN HIBERNATING POSITION
Seepage 17BUTTERFLY IN HIBERNATING POSITION
Seepage 17
Caterpillar feeding upon leaf of milkweedCaterpillar hung up for the change to the chrysalisThe transition stageThe ChrysalisPhotographs from life.See pages8-10,233THE MONARCH BUTTERFLY: CHANGE FROM CATERPILLAR TO CHRYSALIS.
Photographs from life.See pages8-10,233THE MONARCH BUTTERFLY: CHANGE FROM CATERPILLAR TO CHRYSALIS.
Photographs from life.
See pages8-10,233
Chrysalis showing butterfly ready to emergeThe empty chrysalisButterfly just out of chrysalisSide view a little laterPhotographs from life.See pages10-13,235THE MONARCH BUTTERFLY: THE CHANGE FROM CHRYSALIS TO ADULT.
Photographs from life.See pages10-13,235THE MONARCH BUTTERFLY: THE CHANGE FROM CHRYSALIS TO ADULT.
Photographs from life.
See pages10-13,235
Seepage 30THE CAROLINA LOCUSTAbove, with wings expanded as in flight
Seepage 30THE CAROLINA LOCUSTAbove, with wings expanded as in flight
Seepage 30
Selective Color Sense
One who collects the Underwing moths soon discovers that the light colored species which resemble the bark of birch trees are likely to be found upon the trunks of those trees, and that the dark colored kinds which resemble the bark of maple trees are likely to be found upon the trunks of these. Obviously, were this not true the protective coloring would avail but little and it is evident that these moths are able to select a background which is of advantage in helping to conceal them.
There is much evidence to show that in a similar way the butterflies are able by means of a well-developed color sense to select the places where they alight. One of the most notable examples is that of a South American species,Peridromia feronia. This is a silvery gray butterfly which alights head downward upon the bark of certain palm trees that have silvery gray stems and remains there with its wings fully expanded so that it utilizes the background in much the same way that the Underwing moths do. "When disturbed they will return to the same tree again and again."
One who will observe the habits of our Angle-wings and other butterflies which have obliterative coloring of the under wing surface can easily learn that these insects select rather carefully the places where they alight. It will be found that as a rule each species utilizes a background that blends with its own coloring. It is probable that this habit is much more common in other groups of butterflies than has been realized. Much evidence of this sort has been collected regarding the butterflies of Europe and other countries, as well as near our own borders in America.
Warning Coloration and Mimicry
The colors of a great many animals, including a considerable percentage of American butterflies and their larvae, have been commonly explained by the theory ofwarning colors. According to this theory animals whichwere for any reason not edible by birds and mammals have developed various striking combinations of color such as black and yellow, red and black, or black and white, in order to advertise to their foes their inedible qualities. This theory has been very generally accepted by naturalists and will be found expounded at length in many books published during the last quarter century.
The whole subject of the validity of warning coloration has recently been brought up for reconsideration by the illuminating investigations of Mr. Abbott H. Thayer and discussed at length in the book upon "Concealing Coloration" already mentioned. In an appendix to this book dated 1908 Mr. Thayer states that he no longer holds the belief that "there must somewhere be warning colors." He has convincingly shown that a large proportion of the animals which were supposed to be examples of this theory are really illustrations of concealing coloration. But there yet remain various facts which have been conclusively proven that apparently require the theory of warning colors to explain them. Here is another field in which there is a real need for much careful investigation under conditions that are rigidly scientific.
Along with the theory of warning coloration the theory ofmimicryhas been propounded. According to this if a butterfly in a given region shows warning coloration, having developed such coloration because it is distasteful to birds and mammals, it may be mimicked by another butterfly in the same region belonging to another group, the latter butterfly being edible, but benefiting by its resemblance to the distasteful species, because birds or mammals mistake it for the latter and do not attempt to catch it. The most notable example of such mimicry inNorth America is that of the Monarch butterfly, which is supposed to be the distasteful species, and the Viceroy butterfly, which is supposed to mimic it. Several other instances of mimicry are found among our own butterflies, while in South America, Africa, and Asia there are numberless examples.
Heliotropism
It has long been known that the green surfaces of plants respond to the stimulus of the sun's rays in a most remarkable manner. This response has commonly been calledheliotropismand it has been carefully studied by botanists all over the habitable world. More recently, the fact has been observed that many animals respond in certain definite ways to the stimulus of direct sunshine and the same term has been applied in this case. Very little attention has been given to the subject of heliotropism until within a few recent years. But the observations which have been made by Parker, Longstaff, Dixey, and others open up a most interesting field for further observation. An admirable summary of our present knowledge of the subject has been published by Dr. Longstaff in his book "Butterfly Hunting in Many Lands."
One of the earliest observations upon this subject was that published in my book "Nature Biographies" which appeared in June, 1901, concerning the habit in the Mourning Cloak: "On a spring-like day early in November (the 8th) I came across one of these butterflies basking in the sunshine upon the ties of a railway track. It rested with its wings wide open. On being disturbed, it would fly a short distance and then alight, and I was interested to noticethat after alighting it would always turn about until the hind end of its body pointed in the direction of the sun, so that the sun's rays struck its wings and body nearly at right angles. I repeatedly observed this habit of getting into the position in which the most benefit from the sunshine was received, and it is of interest as showing the extreme delicacy of perception toward the warmth of sunshine which these creatures possess."
A little later, some very elaborate observations were made upon this habit of the Mourning Cloak by Prof. G. H. Parker of Harvard University. Professor Parker noticed that during the warm spells in winter the butterflies came out of their hiding places and after alighting, always placed themselves with their heads away from the direction of the sun and their bodies lying nearly at right angles to the sun's rays. By experiment, he found that they adjusted themselves to this position as soon as they were fully exposed to direct sunshine, even if at the time of alighting they were in a shadow. He found that this movement was a reflex action through the eyes, for when the eyes were blinded no such adjustment took place. He called itnegative heliotropism.
Dr. Longstaff uses the termorientationfor this adjustment of the butterfly to the sun's rays and he finds it is a very general habit, especially with the Angle-wings, for the butterfly thus to orient itself after alighting, in such a way that the hind end of the body points toward the sun. This occurs not only with those species which keep their wings spread open when they alight but also with those in which the wings are closed together and held in a vertical position on alighting.
Various explanations of this phenomenon have beenoffered but apparently none of them are yet generally accepted. Were the habit confined to butterflies like the Mourning Cloak, it would seem easy to prove that a main advantage was found in the benefit derived from the heat rays of the sun. Were it confined to those species which always fold their wings on alighting, it would seem easy to believe that it was a device for reducing the shadow cast by the insect to its lowest terms. It has also been suggested that the habit is for the purpose of revealing to the fullest extent the markings of the butterfly. Evidently there is here an ample field for further investigation before definite conclusions are reached.