Popular Science Monthly/Volume 75/December 1909/An Arraignment of the Theories of Mimicry and Warning Colors



There is every reason to believe that all animals' eyes see upon one principle, an eye being a machine for receiving what we call light vibrations, so that to receive from any object more of these vibrations is to have it look lighter, and to receive from it less of them is to have it look darker.

IN the last few years, naturalists have received from outside their ranks, the first scientific analysis of the use of animal's colors that has ever been made.

They have been shown the effacing power of the universal counter-shading in animals' costumes, and later, they have seen with their own eyes the equally perfect effacing power of the patterns which up to that moment they had believed to be factors of conspicuousness.[1] They have thus been forced to perceive that all their own theories prove to have been built in ignorance. These were made before the world had perceived the universal importance of employing specialists, and even Darwin and Wallace failed to realize that in view of nature's infinity, one study like their own was all that they could hope to be faithful to. The laws of visibility reach, like all others, into infinity, and could not constitute part of the zoologist's field, while in the science of the painter, these laws are the very pith of his study.

The following demonstration of the fallacy of the badge and warning-color theories is not, in the same sense, an attack upon mimicry, although it inevitably calls attention to the fact that the latter can not survive the demise of these other theories. It does not imply that there is no case possible of protective resemblance of one animal by another, but contents itself with bringing forward conclusive evidence that the great mass of what is now called mimicry is nothing of the kind, but is, in every respect, the same common concealing coloration everywhere to be found where there are common habits and environment. This fact escaped naturalists, simply because it lay out of their special field, i. e., in optics rather than in zoology, and once off the track, they have been driven step by step into the erection of a wholly fictitious fabric, where no fabric at all was required.

The universal tendency of common habits and environment to be accompanied by common form and appearance has long been a familiar fact, and no one would have conceived of a vast gap in this tendency but that naturalists thought themselves forced to accept the evidence that such a gap existed, and set themselves to work to fill it as best they could. Now, however, the obstacle to their discovering the wholly concealing character of the whole array of costumes that had puzzled them vanishes, and no power can withhold this array from taking its place in the ranks of universal procryptic coloration.

The following pages demonstrate that all diversification of the colors of animals' costumes tends wholly and unmixedly to conceal them. This should set the believers in conspicuous species reflecting that while they are making their records of cases of momentary conspicuousness of individuals of one species or another, they are making no investigation whatever of the possibility that all the while a large number of individuals of this same species are, through some magic of their costume, escaping their sight and making no impression on their minds. It is precisely to such an investigation as this that I here invite the reader.

Any out-of-door naturalist knows that if he walk through a sunny field of fairly profuse vegetation, after first studying it, say, from an upper window, he will flush an immensely greater amount of so-called conspicuous aerial life than he had detected from the window, and he will believe that much of this life was all the time within the field of his vision.

These plates have been prepared with the especial purpose of exposing the weakness of the optical hypotheses upon which the theories of "warning-colors," "recognition," "mimicry," etc., so largely rest. They show that these hypotheses would never have lived a day had their originators begun by testing them. Darwin's erroneous supposition that a conspicuous mark on an object makes the object itself conspicuous has been built on and rebuilt on by the leaders of zoological research, even down to the present day. Entomologists, especially, make much of the supposed power of sharp and strong patterns to render conspicuous that particular part of the insect which they occupy. We now discover that the effect of these patterns is the very opposite. In the illustrations of this article we see the actual effect of such marks in several typical situations. Fig. 1 shows two butterflies and several letters, all of one color, and against one background. On each butterfly and on several of the letters bright spots or patterns have been painted. As the spectator recedes, those parts of the butterflies nearest the bright patterns fade, until, at a short distance, they are invisible, while the rest of the insect is clearly distinguishable up to a much greater distance.[2] The same thing is true of the letters, the unmarked ones being legible much farther than the others. Fig. 2 shows exactly the same effect with reversed colors. That part of the gray butterfly next the black ocellus fades, as one recedes, until it becomes pure white like the background, leaving the rest of the butterfly to continue visible at a much greater distance. In both these cases the effect of conspicuous pattern proves to be the exact reverse of the old hypothesis on which the Bates and Wallace theories so largely rest. Figs. 3 and 4 will, I think, still more surprise the many writers who, from Darwin and Wallace down to the present time, are accustomed to say of one or another brilliantly pied species, that its patterns are so conspicuous that one can see it a hundred yards off. It is here shown that it can not be the brilliancy or conspicuousness of the animal's patterns that enables them thus to see him from afar, since these very characters here produce the opposite effect. The reader will discover, in looking from a greater and greater distance, that it is the normally colored, strongly pied butterfly and skunk, respectively, that fade first, and that all of the remaining six figures can be seen further. (These can be tested not only by distance, but by decreased illumination, and, especially for the latter means, a still more satisfactory test of the skunk can be made by using life-size figures and turning down the lights in a hall, or studying them out of doors as night comes on.) He will discover that the supposed white blazon actually serves to efface the black animal on a nearer view (especially if seen through the leaves). He can not fail, either, to perceive that an all-white skunk, being exempt from the risk of giving an impression of two different things, a black one, and a white one, would in the long run, be, also, the more recognizable when seen against any ground, except snow.[3] It is not yet generally perceived that in the scenery about us every spot means to a casual observer one thing, and it follows that two different color-patches, as of the skunk, amidst the million color-patches in sight, tend, especially when more or less eclipsed by vegetation, to mean, not one pied animal, but two different elements of the scene. It must be remembered that the skunk's scene is a night scene, commonly abounding, in wild places, in black shadow masses relieved here and there by light spots made by bleached twigs, fragments of fallen birches, shining wet spots, etc., and, what is by far the most essential fact, with all visibility whatsoever at the minimum. In fact the whole "warning-color" theory in the case of these nocturnal species smacks of the laboratory. For instance, although skunks abound all
PSM V75 D557 Mimicry and camouflage in nature 1.png
Fig. 1.
PSM V75 D557 Mimicry and camouflage in nature 2.png
Fig. 2.

Figs. 1 and 2 show that bright sharp pattern obliterates instead of rendering conspicuous the form on which it is painted: and that its details are left to be seen to be part of the similar details of the background. Test these fact by receding from the picture.

PSM V75 D558 Mimicry and camouflage in nature 1.png
Fig. 3.
PSM V75 D558 Mimicry and camouflage in nature 2.png
Fig. 4.

Figs. 3. and 4. Here the spectator will find, as he recedes or turns down the light, that all the monochrome figures, even the dimmest, can be seen further, or in a less illumination, than the two normally and brightly patterned ones. These latter fade first. They show how contrasted juxtaposed color-notes destroy each other, so that, contrary to the current theories. monochrome is far better both for revealing the wearer, and also for proclaiming his identity amidst the innumerable details of wild places.

over the premises of American country folk, it is very rare to see one of them, except by encountering him in the hen-house. This is the more significant in view of their well-known temerity and disinclination to get out of one's way. Their white pattern, if seen at all, and even when observed to move, is easily mistaken for some inanimate detail of the scene, some shifting shine on a wet leaf, or other of the above-mentioned light-colored details of the place. There seldom pass many minutes without some breeze to set in motion the many more or less white details of the shrubbery, so that the disembodied light patch of this little hunter's coat may move, even within the short visibility-range that we have discovered it to possess, yet by no means commonly attract our attention. It is certainly the universal experience of American country dwellers that although in their domestic duties they must frequently pass within a few feet of skunks, yet the whole family together scarcely see one a year. Under shrubbery, moonlight, which might be expected to reveal these animals, adds, on the contrary, to the scene hundreds of white patterns, and these often all in motion, so that what was before a comparatively negative form of concealment becomes a most brilliant and positive illusion. The skunk's pattern so absolutely reproduces the hundred surrounding ones, often all shifting in a thousand directions, that even when the animal is near enough to be seen, he is almost sure to escape detection. Dr. Merriam first called my attention to this, and also to a very clear statement of it by Verrill. Merriam says that spilogales amidst cactus shadows in moonlight are practically impossible to follow with the eye. It is easy to see that this must apply to the appearance of all the other top-patterned species under similar circumstances. This is one more instance of increased illumination producing increased procryptic effect, such as we also see in the operation of counter-shading.

These simple diagrams, then, prove that it is not the diversification into brilliantly contrasted pattern that makes its wearer conspicuous, or that is the most efficacious way to make him recognizable. It is the very juxtaposition of the skunk's black that makes his white fade out of sight at a short distance, and in a nearer view, amidst the many light spots likely to be in sight, one more has often no significance to a spectator; but if this light color took the full shape of a skunk, then, of course, it would make him recognizable. In the next plates we shall see what is the main cause of the frequent conspicuousness, during motion, of all aerial species, or of any that by virtue of being taller than others, are, to these lower-level observers, practically the same as aerial, because of looming against their sky.

But let us turn aside to notice the circumstances of the species already recognized by naturalists as procryptically colored. These are merely such species as live, or rest by day, in actual contact with their background, and, as now proves to be the case with practically the whole animal kingdom, wear its colors (or some of them).

Bark moths, terrestrial animals in general, terrestrial habited birds, etc., all such species, when thus in actual contact with their backgrounds, share its varying illumination from minute to minute. The patch of sunlight that falls on the squatting woodcock illumines also

PSM V75 D560 Mimicry and camouflage in nature.png

Fig. 5 shows the revealing-effect of being seen against a contrasting background, and illustrates the fact that no aerial creature, can go about at all without constantly passing across both revealing and concealing backgrounds.

the surrounding ground, so that the bird continues to seem a part of it; and the same is of course true of the rest of this great class of animal. But a very different fate attends the life of aerial species destined constantly to appear against more or less distant backgrounds which do not share their particular momentary illumination, and which constantly show, now light, when the flying bird or butterfly is dark, and the next instant dark when he is light. This fate causes all aerial species, in a very vital sense, to be conspicuous. Fig. 5 illustrates this. On the white of the sky a white butterfly has been pasted, which, of course, does not show. In the same way, a black one has been placed upon the darkest part of the tree's shadow, and a ground-colored one on the ground. Both of these, like the white one, are, of course, practically invisible, and, could they be always seen against these same backgrounds, they might be classed as cryptic. But their habits preclude the possibility of this, their own changes of position, not to speak of those of the spectator, bringing them, as they fly about, across, often in a single second, the whole gamut of backgrounds, from brightest sky to deepest tree-shadow, and back. And against every one, except the single one which they match, they are clearly visible. The black one shows against the sky, the ground, etc., the white one against the various darks, and so forth. Nor is it only their varying background that dooms them to visibility. Their flight carries them with equal speed through a series of metamorphoses of their own aspect. Now, for an instant of their passage, they are themselves practically black, because of being in deep shadow, and are perhaps seen against a bright sky-space. The next has brought them out into full sunlight, and they blaze bright against a new background of perhaps inky darkness. The principle of the inevitable visibility produced by this swift succession of visible moments, though alternating with repeated vanishings, is well illustrated by the complete visibility of landscape through the cracks in

PSM V75 D561 Mimicry and camouflage in nature.png

Fig. 6. In this figure the two inconspicuous butterflies in the middle show the effacing-power of pattern when it repeats the background. At the left a butterfly of the same costume is represented passing through a moment of illumination too great to admit of its patterns' still cutting it apart into notes of the background. At the right a butterfly of the same pattern Is going through the reverse experience, being for an instant too much in shadow for its pattern to save It from appearing as one single dark form against the light space beyond. This illustration reminds one how perpetually such vicissitudes must succeed each other in the life of such species.

a board fence, to the eyes of any one passing swiftly by. The view recurs again and again to the retina, in time to keep up the image. This is why the average observer thinks he sees these butterflies through all their course. This plate only goes so far as to show how fatal to invisibility it is to have the wrong background. Fig. 6 illustrates the above explanation of that perpetual disharmony between flying species and their background which plays fast and loose, part of every second, with all their patterns' power to cut their forms into deceptive shapes, by making them, so constantly, first so bright against dark that all parts, even the blackest, fall into one light silhouette, and then so dark in shadow, against bright light, that even their white parts join the rest in one dark silhouette. These two climaxes of visibility, first one and then the other, occur in the flight of a bird or butterfly often at the rate of several a second, while, during the rest of the second, the creature is effaced by passing a background that his costume matches, and by being, as in the case of the butterflies A and B, midway between the extremes, favored by the momentary illumination's being neither too great nor too small. Even in the climaxes of conspicuousness his patterns still perpetually lessen his visibility in direct ratio to their strength. And when he is chased by an enemy every instant of confusion as to where he leaves off and the background begins must often save him, so that the brighter his light marks and the deeper his dark ones, the greater the range of background" he can meet without silhouetting as an entirety, and being for the instant conspicuous. One advantage which patterns do certainly sacrifice in purchasing the above advantages is that, although their Wearer is never seen entire until the background is too dark for his black, or too light for his white, yet it is true that, on the other hand, some note of a pied costume is always to be detected moving when the wearer moves. In this respect, monotone, whenever at rare moments it exactly matches its background, has the advantage. This fact additionally condemns the aerial animal to detection when he moves, yet it is often rather his motion than his form that becomes noticeable, because each of his patterns still has the chance of passing for something beyond him.

Fig. 7 shows one of the cardinal effects of patterns. C is a bird patterned in white, black and gray. Seen against the sky he loses his white part—against the dark he loses his dark part—and against the gray, his gray part.

Now when we find that pattern works always for concealment in direct ratio to its own conspicuousness and elaboration, there remains no vestige of evidence that the specific recognizability of the of course constant pattern of each species has had, even to the slightest degree, a hand in the evolution of such pattern. And those who would still claim for conspicuous patterns any other reason for being than concealment of their wearer must first show what patterns could in the slightest degree better serve procryptic ends, under the circumstances, than the very ones now in use; and also what ones would less aid identification.[4]

There are two groups of supposed warningly-colored animals that seem particularly to lend themselves to the exposure of the weakness of these theories. These are the carnivora which bear anal stink-glands, and, among insects, those armed with stings at their rear end. In the first place, very few of these species wear the so-called badge conspicuously

PSM V75 D563 Blending in and contrasting the background.png
Fig. 7 shows the fundamental optically disruptive effect of pattern. Against each background the bird loses that part of its form which matches it.

near their weapon, or anywhere that would call attention to the situation of such weapon. (Entomologists have been keen to cite any supposed cases of the contrary arrangement, any apparent association of "badge" and armament.) In the second place, the coloration of all the members of these groups proves to be the most perfect imaginable concealing coloration, picturing the details in the most exquisitely true colors of the very background against which it is most dangerous for their wearer to be detected (commonly that of their feeding ground). While, on the other hand, apparently no brilliant colors at all are found in any branch of the world of above-ground animal life, either in air or water, where no such colors are typical of any of the animal's backgrounds, no matter how much these animals may need advertising.

The famous black and gold, the supposed blazon of offensiveness, so characteristic of the wasp and bee family, is the utmost picture of the sunlit vegetation which they haunt, with the golden stamens of flowers, the yellow of fruit, and the dark interstices. Were this coloration really such a blazon, why does nature deny it to the hordes of stinging ants that often swarm within a few feet of the wasps, but wear only the comparatively dull tones that match the bark and earth surfaces to which their general lack of wings condemns their lives? (Such red as is found in ants' costumes is a universal detail of the forest débris.)

So ingrained is the time-honored conception that such a creature as a golden-patterned wasp, as he bustles among the flowers, owes his conspicuousness at least in part to his costume, that only actual personal experiment with the laws herein shown can dispel it. Not till naturalists give up collecting records of cases of conspicuousness, and begin to inquire by experiment whether any more procryptic coloration could, under the animal's circumstances, be devised, will they have begun at all the study of this subject. One day's investigation of this kind would greatly astonish them, and they would end by discovering that it is unequivocally the wasp's actions that condemn it to so much visibility, and this in spite of its wearing, as far as they can discover, every available form of concealment-coloration.

As to the supposed warningly colored carnivores, the light-colored marks that are considered as badges are often prominently concentrated upon the animal's face and front top, and in no case equally prominently arranged near his rear. Being always on the creature's sky-lit surfaces, they obliterate him to the eyes of beholders from a lower level, such as the seeing portion of his small terrestrial victims. In doing this they fall into the universal class of concealing coloration. Fig. 12 illustrates this function, and the previous illustrations have shown that this same white, so perfect an auxiliary of the animal's feeding operations, is not, in other views, unfavorable to its concealment.

Let us now find out what traits and habits in these groups do constantly go together. We find among the stench-bearing carnivores, just as among the above insects, that the bright patterns are only found on such species as have, in their background, colors that these patterns match, to the eyes of certain other animals whose sight they need to avoid. They are found on skunks, civets, badgers, teledus, ratels, for instance, and the animal life devoured by these carnivores is said to consist largely of worms, insects and mice, most of which are presumably either caught on the surface or dug out of the turf, i. e., procured on a lower level than the predator's head. Such of this list of

PSM V75 D565 Blending in and contrasting the background 1.png
Fig. 8. Fig. 9.
Fig. 8 shows that a monochrome figure will continue distinguishable, because of its continuity of color, even when largely eclipsed by interposed forms.
Fig. 9 shows how much less distinguishable an animal would be in the same situation if his head, feet and tail were light colored.
PSM V75 D565 Blending in and contrasting the background 2.png
Fig. 10. Fig. 11.
In Fig. 10 we see simply the skunk's reproduction of the other light-colored details which partly form the animal's background, and partly mask his form, so that both his darks and his lights tend, as It were, to dissolve their partnership and ally themselves to their counterparts in the surroundings.
Fig. 11. Here we see a skunk whose patterns are experiencing exactly the same over-darkening as that of the right-hand butterfly in Fig. 6. This sketch and that of the butterfly illustrate one of the greatest uses of pattern in forest species in combatting the silhouetting propensity universal to animals observed in a dim forest illumination against lighter regions beyond them.

victims as can see would certainly have much more chance to escape, were not what would be a dark-looming predator's head converted, by its white sky-counterfeiting, into a deceptive imitation of mere sky. Now let us see whether such stench-gland bearers as hunt in a bolder way wear the white "badges." We find at once that minks, martens and most weasels, though well-armed with the same glands, have no top white, and, instead of hunting along the earth's surface or putting merely their heads into holes for their prey, go boldly under ground and attack such prey as hares and marmots, or fasten upon fowls much larger than they themselves. From all such prey their foreheads have nothing to gain by being white, since in the hole all is dark, and in the case of these large victims attacked above ground, the attacker, if it be a weasel, is looked down at, not seen against the sky, while the martens,

PSM V75 D566 Blending in and contrasting the background.png
Fig. 12 shows the animal's white top performing its perhaps cardinal function, viz., that of effacing his top contour against the sky to the eyes of inhabitants of the turf.
This is the only function of the skunk's white top that is practically unceasing as long as the animal is above ground. We have already seen that his white and black cause each other, especially at night, to fade from sight at a short distance, and even at a near view, confuse themselves with forest details. But the obliterating power of night itself largely suffices to render all devices for concealment unnecessary. The great development of the ears of nocturnal animals attests their difficulty in seeing at this time. On the other hand, the night is scarcely ever so dark but that a solid form within a foot of one's eye would show dark against the sky or the light parts of the forest ceiling, and surely this must be the reason why skunks and the other grubbers of small surface life wear this wonderful counterfeit of sky on their foreheads. By its aid, they must constantly come close to many kinds of small surface-life on which they so largely feed, which would evade them if they could see them. This must be especially obvious to any one who has often tried in vain to creep within catching-distance of grass-hoppers. A single night's out-of-door experimenting will convince students of the importance of the white top to such an animal as a skunk.

arboreal, acrobatic, swift and bloodthirsty, catch doubtless much after the bold manner of the small weasels, and obviously would not seem to have so much use for concealment from any particular viewpoint. The pine marten, however, is enough light-foreheaded to save his head from too much silhouetting in his above-ground forest operations against small terrestrial life.

In tall grass, to catch small terrestrial prey like mice, cats creep low, and fling themselves high in air, dropping flat outspread upon the dazed victim. Foxes vary this by coming down head-first upon it. In neither case would top-white help them—and they haven't it.

The one thing which all these sting and stink bearers have constantly and in common, with perhaps no exception, even including the dogs and hyenas which have also anal glands (cats lack such glands and also lack the habit of digging) is this: In pursuit of food, or in storing it, they all either go bodily into holes, as bees and wasps into flowers and fruit cavities, or ants into their galleries, and as do the weasel family after burrowing mammals, or like the grubbing species above mentioned, and foxes, stick their heads into holes for similar purposes. In all these cases these rear-armed species have a common need to be so armed, being totally helpless to defend themselves while thus immersed. Of all animal adaptations this stink apparatus was the thing most to be expected in a part of the animal so entirely defenseless.

Picture a bee deep in a flower, or a badger with his head jammed deep in a mouse hole—what a chance for his enemy! But these hind ends have taken care of themselves. Now notice the thing that seems to bring final ridicule on the "badge" theory. Take, for instance, the grison, Patagonian weasel, bridled weasel, the badgers and the skunk, species whose white pattern is worn upon the head (the skunk's tail is normally a mixture of black and white hairs—like a gray cloud); the moment when these animals most need to advertise the offensiveness of their armament would be when they were most defenseless, and this is, of course, when their heads are in holes, and at such a moment their "badges," being on their heads, are concealed! The apparent reason for the white patterns' extending so often along the back nearly to the tail is very simple. The act of digging or of stepping down into a hole tends to bring the fore part of an animal lower than his rear, and this, to eyes upon the turf, brings the whole of his back against the sky, and an erect white tail (like the upturned plumes of the egret) additionally blends the wearer into the sky.

To realize how inevitable was the development of special rear protectors a man has only to conceive what an anxious sensation he himself would experience if in a jungle he had to spend much time with his head down a hole, and the rest of his body a tempting bait for tigers.

In fine, we find upon certain species of carnivora that we know to be more or less scavengers and catchers of small fry such as require rather to be picked up than stalked or chased, and on others that we suspect of having the same habit, the same sky-picturing patterns that, from the eye-level of their prey, efface the top contours of most other slow walking feeders-on-small-life, in all branches of the animal kingdom. We find, on the other hand, a large number of Mustelidæ, as well as a number of other carnivora, well armed with stink-glands, but, as if because of not feeding in the same manner, entirely without top-white patterns.

Among the genera that move erect without crouching in the vicinity of their prey, and catch it upon surfaces below their level, and which are commonly effaced as to their top contours, by white in their costume, are herons, cranes, pelicans and the omnivorous swans, geese and ducks. And a similar use of white effaces the rear view of an immense number of widely separated members of the animal kingdom which tend to be seen by their pursuers against the sky. Among these are hares, deer and antelopes, ground-nesting birds that commonly spring on wing from the nest on being flushed, such as waders, ducks and geese, many passerines, and such hawks as nest on the ground. These all wear some white rump or tail pattern, which obliterates them in the most magical way, against the sky, exactly while, in getting under way, they are for a fateful instant within springing reach of the cougar, lynx or fox that, with head close to the ground, has crept up to them. White patterns abound on aerial passerines in general, but even white wing bars are lacking from such as keep so close to the ground that no enemy sees them against the sky. And among small rodents none has top-white unless, like the jerboa, he jumps high enough to be seen by his pursuer against the sky.

Now, while these white top patterns seem to be universally employed wherever they can be of the above service, on the other hand, they are conspicuously lacking from the rears of such ground-nesters as habitually run, instead of flying from the nest, and thereby avoid showing against the sky, to terrestrial eyes. Such are gallinaceous birds, tinamou, rails and many other sedge-haunting species. (Gallinules whose upturned tails present, as they fly, a white sky-picture, probably keep their tails down when they slink from their nest.)

The white patterns on the breasts of several kinds of bear, which Mr. Pocock has classed as warning colors, serve perfectly the same obliterative purpose that we find in all the rest of upward-facing white patterns. (These patterns face upward when the bear stands erect, i. e., face upward to the degree necessary for catching top light.) Now, we find that for recognition a monochrome silhouette, especially in a thicket, is far superior to a patterned one (see Fig. 8), and also we have small grounds for thinking that such bears, when standing erect in the jungle, have need to be afraid of being attacked by mistake by any of their neighbors that would avoid them if they recognized them. On the other hand, be the beards object, in thus assuming man's attitude, either aggression or defense, he gets the same general advantage out of escaping detection, and the chance of enjoying this boon is, as we now realize, greatly increased by the chance of this light patch being, at the right moment, just sufficiently illuminated to pass for a sky-hole through the dark mass of shrubbery of which the erect bear forms the dark center. All woodsmen know that when one has followed with the eye some bird or arboreal beast, and lost it in the trees, the eye searches every mass of foliage silhouetted against the sky, scarcely counting on discerning the creature's outline, so much as on noting some mass of unbroken dark—dark without any sky-hole through it—sufficiently extensive to contain the animal itself. Into this mass, if one wish to kill the animal, one shoots, at a venture, and very often with success. In such cases, a single white mark on the concealed animal has a great chance of showing through the foliage, and saving the creature's life by passing for sky, making the hunter think he can see through the clump as if there were no opaque animal in it.

This wonderful universal function of top-white will only begin to have vitality in the minds of students of natural history, when they begin to take the trouble to spend hours, lying flat on the ground, studying terrestrial life from the true point of view. They will at last realize that the terms "cryptic" and "conspicuous" can refer only to the relation of objects to their background, and that a hare is as conspicuous, dark outlined against the sky, to the little mouse at his side, as the white heron looked at on the ground by man; while to the mouse, this same heron, now seen against the sky, is the perfection of procryptic coloration, just as is the hare seen from the level of a man's or hawk's eyes.

There is one more point that particularly bears on the skunk matter. There is not, through all his range, any mammal, unless one counts the porcupine (an animal to resemble which would be a protection) with which any other creature could possibly confuse him, except when he is very dimly seen, either by virtue of darkness or of interposed forms, and in either of these situations, as this article has shown, the light pattern only diminishes his recognizability. Do we not, in fact, forget the evidence that the wild animals know each other by far more subtile means than we might suppose? The dog, even after all these centuries of domestication, still keeps the power to recognize his master, not merely by his scent, but by his foot-fall, when he can not see him. And the kind of faculty implied by this is even strong in the wild races of man.

Many naturalists think that such circumstances in the life of a race as are of only occasional occurrence have no part in its evolution. History seems to demonstrate the opposite. It is the stresses that are formative, and are the weeders-out of weak elements. The men of a village, regardless, for instance, of whether all of them could swim, might go yearly all summer to their meadows, and all come home at night, till once when some sudden deluge swept their valley, only the strong swimmers would escape, and thenceforth that village would comprise only strong swimmers. How often do we hear some one tell of a small and half forgotten faculty having saved his life. This seems equally to apply to the evolution of all animals. Their races are, in the long run, subject to great fluctuation of prosperity, many of them coming, occasionally, near to extermination in some part of their range. This, according to universal belief, is oftenest through famine, and in that case, plainly, those individuals best able to accommodate themselves to new food, and to new methods of procuring it, would be most apt to survive. For this reason it does not signify whether badgers, etc., eat a larger or a smaller proportion of seeing food, since those individuals best fitted to catch it will ultimately constitute the race, because, while a white-topped animal would be no worse than a plain one at eating turnips, he would excel him at catching mice and crickets when turnips chanced to fail; and, as this article shows, his white does not in any way increase his conspicuousness.[5]

Patterns of animals are like scars of ordeals, recording what their wearers have been through. Those hares and antelopes and deer which, by virtue of a white sky-imitation on their rears, were not too fatally good a target against the night sky for the stalking feline that flushed them, have survived to propagate their race. The same record of how they escaped the eyes of prey or enemy is found on the costumes of most of the animal kingdom.

Let us try to get a vivid view of the whole field of the world's animals; over the whole earth, all species, of all orders (that ever prey or are preyed on), wear, regardless of all possible needs of badge or mimicry, such colors, and nothing but such colors, as are to be found in certain of their backgrounds. Nothing but failure to perceive this broad fact has made it possible for all these rootless theories to gain a foothold. The two most recent theories, Professor Gadow's, and that of several experimenters, that humidity is the cause of patterns, both these are invalidated by the same general arguments. Dr. Gadow, who believes that it is shadows flickering over a lizard's back that cause his patterns, ignores the unmistakable fact that lizards, like all other terrestrial species, are colored and patterned to match the ground on which they live, no matter whether there be vegetation over head to cast shadows, or, as on sea-beaches and bare rocks, nothing but air and sunlight. The humidity theory has the same defect. It believes that the increased richness in the colors of a species as one traces it from the arid part of its habitat to such a region as the moist-aired gulf-state forests, arises from the increased humidity, not noticing that with the increase of humidity goes always a corresponding enriching of the vegetation which forms the species' background. Let these investigators push through the mangroves that border this sultry aired forest against the bare sands of the gulf, and they will find, two steps out upon the beach, in a saturated ocean atmosphere, a beach and ocean fauna of the purest beach and ocean colors, palest gray and pearl.

Black-and-gold is as truly the background color of the flower haunting black-and-gold wasp, as is stone-weed-and-sand color of the stone-and-weed-and-sand colored sandpiper. Scarlet and yellow fruit colors, sky-blue and green leaf colors, on the macaw, are as absolutely the picture of this bird's background while he is dangerously absorbed in feeding in a tropical fruit tree, as is the little terrestrial mammal's brown the picture of the universal earth-brown on which he lives. The thousands of species of open ocean fish, the bare sand-dwellers and the ocean-air-fliers, all wear only the colors that characterize their backgrounds, often adding for the breeding season bits of the scenery of their nesting place, as in the case of puffins, whose gaudy breeding-season-bill on guard at the mouth of the burrow, obliterates the dark hole itself, and at the same time substitutes a semblance of flowers to complete the deception. The moment these domestic duties are over, and the puffin back in the open sea, we behold him dressed again in the universal ocean-and-rock colors of his habitat. (To show that no physiological difficulty prohibits fish, for instance, from wearing gaudy colors, we find such colors upon them wherever they live amidst brilliant corals and brilliant water-plants.)

To complete the above argument, notice that, as my illustrations show, it is in the midst of vegetation, or other confusing and more or less eclipsing surroundings, that monochrome is far the best costume for identification, while out in the open spaces, the air, the beach, and the sea, there, where no twigs or other forest details threaten to confuse the identity of pattern, striking devices of all kinds would have their fullest chance to effect the identification for which they have been supposed to exist. And what do we find? We find nature foregoing, from end to end of the world, every chance to make use of this obvious opportunity.

Furthermore, to show that it is not a matter of regions, notice, as I have pointed out, the gilded wasps living within a few feet of earth-colored ants, and little earth-colored rodents swarming on the brown forest floor, two feet below bright dressed inhabitants of the bright dressed overhanging foliage. Why do not these rodents, forever preyed upon, in fact the stand-by diet of carnivora in every order, why do they not develop unpalatability and badges? All attainable unpalatability they must possess, after their immeasurable period of being picked from, but why not the badges? The truth appears to be that all advantageous attributes have, in every animal, grown side by side, and that the culminations, for instance, of concealing coloration, such as the transparency of a group of the supposed mimetics, have gone on, in this group, hand-in-hand with that of unpalatability. Now that we see that all procryptic coloration (except, of course, the facsimile kind, such as that of geometers) produces its effect by making the observer seem to see through the place where the colors are, it follows that actual transparency, as in these "mimetics," must, in ever so many situations, be wonderfully potent for obliteration. It is, of course, the only scheme for succeeding equally against both the light and the shadow, tending both to escape showing light against dark backgrounds and dark against light ones. Here are Bates's own remarks about the degree of conspicuonsness of a transparent butterfly. In "A Naturalist on the Amazons," on page 39, he writes:

Some have wings transparent as glass; one of these clear-wings is especially beautiful, namely, the Hetaera Esmeralda; it has one spot only of opaque coloring on its wings, which is of a violet and rose hue; this is the only part visible when the insect is flying low over dead leaves, in the gloomy shades where alone it is found, and it there looks like the wandering petal of a flower.[6]

As a few hours' experimenting in obliteration by juxtaposition of patterns will prove to any student, the optical laws which govern it are so absolute that one is not surprised to find that the whole world's butterflies have scarcely three different schemes of pattern. The principle of pattern arrangement in these famous "mimetic" groups (shown in Fig. 6) is out and away the predominant one over the whole globe. If this is the case, is it strange that in each most swarmingly populated seat of butterfly life there prove to be a number of species which, living in the very same station, and with seemingly identical habits, have, in obedience to this great pattern-law, practically identical patterns and form? We see in the ocean, for instance, even mammals wearing the shape and color of fishes?

The question, now, is, at most, merely why they have the same station and habits.

Let us dwell a moment upon the significance of this finding of the greatest cryptic coloration in the very midst of the so-called mimetics. First we must remember that all men agree that it is only persecution that can have engendered any form at all of protection. It is, as I have said, inconceivable that any forever preyed-on and picked-from race should not have acquired all possible unpalatability. And it is equally inconceivable that any race that either preys or is preyed on should not during the same periods have become, also, as nearly as possible either invisible, or at least unrecognizable as any form of animal life. Such a boon incomparably surpasses any advantage from passing for some other at the best not wholly inedible animal. Therefore one would have expected to find all species of the classes above referred to proving to be by one means or another at the minimum of recognizability as animals, and at the same time, at a corresponding minimum of palatability; and behold, that is just what we find! We find in the very ranks of the supposed mimetics (a term which asserts a protection involving conspicuousness of the protected individual) the actual climax of invisibility, as Bates practically testifies in the above extract, and as I too, and all others who have studied these insects in their homes, must testify. (In deep forest shades the actual illumination is faint, and objects show most when they come between the beholder and regions of more lighted foliage beyond or up nearer the forest's top. In these circumstances all rank patterns are potent to thwart the revealing-power of silhouette, and behold, here we find the very prince of silhouette thwarters—transparency itself!)

As to the impression that "flaunting flight" (i. e., slow or weak flight), gaudy costume and unpalatability keep together, they do not do this to any very impressive degree, as the accompanying table will remind the student. Entomologists will see that this table is sufficiently correct for my purpose.

Acme or Respective Traits

Of Strength
and Speed of
Of Gaudiness
of Costume.
Of Slowness. Of Dulness of
Of Alleged
Morpho. Morpho. Heliconius. The whole Heliconius.
Papilio. Papilio. "mimetic" group The whole
Heliconius. proper. "mimetic" group.

In fact, one finds, as one would have expected, that every butterfly has the gait best suited to the kind of place that he lives in. Heliconius, one of the very slowest genera on our continent, is particularly at home while flying through the densest copses. It is perfectly natural that such a butterfly as a grapta, matched to the colors of the ground, should hurry, in flying from one safe spot on the ground to another, but the case of Heliconius is very different. He lives in cover, the very kind of cover to which small birds fly from a hawk, and through this he sails and flits in the only conceivable manner, threading its minute alleys with short wing beats, and at times almost seeming to stop and crawl through the narrowest places. This is, at least, true of charitonius. sara and melpomene, in the West Indies and Trinidad, where I have seen them. As is characteristic of all nature, these insects overflow from the situations that most nurture them into less favorable places.

Yet it is almost a sufficient answer to the natural question why they are not there preyed upon, to point out afresh that on the American continent, at least, no kind of butterfly at all appears often to be attacked on the wing. In Trinidad, one of the keenest of that remarkable able family of born naturalists, the Carrs, told me that he had never seen a bird catch a butterfly, and this has almost or quite been my experience too. In Trinidad, for instance, one may see flycatchers catching slow fliers like beetles, by the hour, any day, but never see them pay the slightest attention to any butterfly whatever. I reiterate this here, merely for what it is worth, and am nowise averse to believing that Heliconius is more than ordinarily unpalatable. If it be true that feeding among red or orange flowers has now or formerly so predominated in the life of the red and yellow spotted species, as to make this dress do them more good than harm, it is equally logical that, as in the case of the digging and burrowing animals that I have referred to, with their corresponding rear armaments, butterflies particularly subject to dangerous absorption while feeding, should have been in the whole period of their existence bred to an excessive degree of inedibility. As to the flight of such butterflies as on the one hand, papilios and morphos, and on the other, the "mimetic" groups proper, the former two families comprise between them, the strongest and swiftest of American butterfly flight, and an unsurpassed brilliancy of costume, bright colors not proving, in their case, to be accompanied either by slow flight, or by equally notable unpalatability. On the other hand, the American so-called mimetic groups proper have a middle-class flight apparently well suited to the by no means open under-brush of the forest, where they go about much in the manner of the genus Hipparchia in the north.

Now to glance for a moment at the significance ascribed by entomologists to the injuries which are found along the borders of butterflies' wings.

Perhaps the most highly artificial and strained hypothesis that has been released from duty by the discovery of the use of patterns is the conception that after a million years' experience birds would not inevitably know what part of a butterfly is edible and instinctively seek it, rather than try to eat the tissue-paper pictures of background painted along its wing-borders. This is entirely contrary to the stern rectitude of nature. One might as well hope to fool a ship about her center of gravity, and induce her to float at an angle that did not defer to it, as induce a million-year-long race of eaters of butterflies' bodies to waste energy over these patterns.

A butterfly has, of course, a fairly tough body, and wings that begin tough next to the body, but become mere tissue-paper at the lateral borders. Now, every slightest contact is perilous to the entirety of these borders, and, at the same time every circumstance of the butterfly's life threatens contact to them. Even the wind may blow things against them, and when the butterfly is pursued by an aerial enemy, his own efforts to escape must often bring them into collision with vegetation. Again, if the pursuer be a bird, his swoops bring him into almost inevitable collision with these outstretched wing-borders. To lunge at a thing and miss it is inevitably to be carried on, the next instant, close past it. To put it from the insect's point of view, barely to dodge an onrushing foe, is, as we all know, to have him almost inevitably brush against us, to say the least, as his impetus carries him forward. It would be absurd to doubt the very great likelihood of mutilation to the butterfly's wing-borders at such a moment. Again; a bird struggling, against difficulties, to seize such a thing as a zigzagging butterfly, inevitably tries for the mass of the target, the most visible part. Now, although the wings do, certainly, more or less wag the body up and down, nevertheless the body is the axis of the mill-wheel of which the wing-borders are the floats, so that even if the bird tried for the body, unless the attack came exactly from behind, the flapping wings would tend to protect it by constantly getting in the way of the bird's beak, but this would be at the expense of these delicate fabrics, which would smash themselves against it. So much for the immensely greater risk of every sort to this delicate border than to the body itself.

Now as to the supposition that birds prefer to seize this border region, rather than the body. One simple fact suffices to show us that we have not the slightest evidence that they do so. It is this. A butterfly seized by his body can not escape (unless, of course, he chance to be cut nearly through by the beak that seized him) while one seized by the wing-border is no more detained by being thus seized than by receiving at this point any of the merely accidental injuries above referred to. Now if a butterfly seized by the body, is generally eaten, while on the other hand every butterfly injured as to its border, escapes, what possible significance has our finding, as we do, mainly border injuries?

Now, although it seems scarcely necessary to finish the argument, to consider for a moment the supposed selection by the bird, of special points along these borders, the reader has been sufficiently reminded how very far the bird is, in one of these chases, from being in a position to select a point of attack.

We find that the whole subject of animals' coloration has been handled with very loose thinking, as if the old time disrespect of natural history still haunted men's minds and dissuaded them from real study. This cloud that enveloped natural history in former centuries has been steadily thinning, but it is certainly accountable for many loosenesses even up to the present time.

For instance, it is perfectly plain to-day that nature would not ask a coral snake to get along with a costume which, while it often served to warn off his enemies, proved, at other times, a disadvantage to him by identifying him to the animals which he wished to eat. The writings upon these subjects, down to the present day, teem with just this kind of weakness. Also, being falsely based, they have needed props and dikes at one point after another, and these have naturally proved to be out of harmony with each other. Here it has been assumed that animals need badges for mutual recognition, and there, as in the "mimetic" groups, a theory has obtained which assumed that they need nothing of the kind. Individuals of each species of these groups have been expected to know each other amidst a crowd of close imitations (and doubtless they could do so).

The much insisted upon significance of the superficiality of the resemblances among the "mimetic" groups vanishes upon our discovery of a full blown use, of the most direct and primitive character, for all these colorations. From that moment, these resemblant costumes are seen to be, as I have pointed out, on one basis with the many other resemblances among species of widely different origin that have long enough had the same habits and environment. All these, and they are to be found in many orders of the animal kingdom, are only superficial resemblances, yet it is perfectly plain that they have been acquired for a use. The proof that they are only superficial is that the anatomist can discover the real pedigree of the disguised species by an examination of the elements of its structure. Good examples of this fact are the whales and seals, with their hind legs more or less arranged into a fishtail, yet perfectly recognizable by the zoologist. (I assume the truth of natural selection.)

In fine to imagine that the forest population, living side by side, in perpetual need of knowing each other, would be in any way helped by badges, is as if some person, newly arrived in a long-established community, supposed, because he could only distinguish its members by prominent superficial marks, the red hair of one, the pock marks of another, etc., that this was how the members themselves knew each other, after lifelong familiarity.

The truth, however, is, that were he to cite these distinguishing marks, in speaking of one member to another, he would find that the mutual familiarity of these members had become so subtile, had, so to speak, sunk in so deep, that they had almost forgotten the existence of such marks at all, except where men's names commemorated these.

Lifelong members of a community, all reacting upon each other in a hundred ways, know each other by innumerable means, all communicating with their subliminal consciousness. To this consciousness, the movements, for instance, of a mink in the bushes, probably announce his identity to all his neighbors, who hear him, just as plainly as if they saw him, and the least glimpse of him would, upon the same principle, be as good as a full view. Habitual woodsmen generally tend to believe this, because of finding that they themselves tend to this intuitive method of identifying their wild neighbors in the forest.

  1. I have shown, both to the naturalists at Woods Hole, Mass., and in London, the wonderful concealing power of various representative "conspicuous" costumes, from white patterned birds seen against the sky, to the bright red-black-and-yellow coral snake, supposed to be one of the most conspicuous animals in nature.
  2. In most butterflies, the body itself is wonderfully effaced by having its color blent off into the wing; yet it profits greatly by the effacing-power of the strong pattern to right and left.
  3. To learn the superiority of monochrome for identification, paint a uniform tone over a chiselled inscription that has become hard to read because of weather-stains, and instantly it is as legible as when it was first cut.
  4. Naturalists confound identification with mere detection. Our identification of familiar objects depends, fundamentally, upon unvaryingness of their appearance. We know the mink just as well by his slim form and sleek dark monochrome as the skunk by his fatter form and bushy black and white. The skunk, by the way, varies in appearance far more than the mink, ranging from nearly all black to half white, and this is another evidence that his pattern is not for identification. Naturalists' assertion that patterns serve equally both purposes is like two men claiming the same dog. The wise judge puts it to the test as to which man the dog will obey against the commands of the other. If we call animals' costumes the dog, we find that he always obeys Master Concealment, but obeys Master Warning-color only when Master Concealment has commanded the same thing.
  5. This applies to all such cases as the objection that seated butterflies are apt to have their wings closed, and therefore need no concealment-colors on their upper sides, and that flamingoes seldom prey on animal food that can see, and therefore have little need to match the sky against which they loom. The butterfly's fitness for opening his wings in safety, when he needs to do so, and the flamingo's, for eating seeing-food when he must, are distinct advantages.
  6. The deepest forest shades seem to be, everywhere, the typical home of these transparent species. In Trinidad they bear a popular name that alludes to this characteristic.