But she does not always do so without hindrance. Let me recount one of the misadventures of this work of storage. There is in the neighbourhood of the burrows a plant which catches insects with glue. It is the Oporto silene (S. portensis), a curious growth, a lover of the sea-side dunes, which, though of Portuguese origin, as its name would seem to indicate, ventures inland, even as far as my part of the country, where it represents perhaps a survivor of the coastal flora of what was once a Pliocene sea. The sea has disappeared; a few plants of its shores have remained behind. This Silene carries in most of its internodes, in those both of the branches and of the main stalk, a viscous ring, two- to four-fifths of an inch wide, sharply delimited above and below. The coating of glue is of a pale brown. Its stickiness is so great that the least touch is enough to hold the object. I find Midges, Plant-lice and Ants caught in it, as well as tufted seeds which have blown from the capitula of the Cichoriaceae. A Gad-fly, as big as a Blue bottle, falls into the trap before my eyes. She has barely alighted on the perilous perch when lo, she is held by the hinder tarsi! The Fly makes violent efforts to take wing; she shakes the slender plant from top to bottom. If she frees her hinder tarsi she remains snared by the front tarsi and has to begin all over again. I was doubting the possibility of her escape when, after a good quarter of an hour's struggle, she succeeded in extricating herself.

But, where the Gad-fly has got off, the Midge remains. The winged Aphis also remains, the Ant, the Mosquito and many another of the smaller insects. What does the plant do with its captures? Of what use are these trophies of corpses hanging by a leg or a wing? Does the vegetable bird-limer, with its sticky rings, derive advantage from these death-struggles? A Darwinian, remembering the carnivorous plants, would say yes. As for me, I don't believe a word of it. The Oporto silene is ringed with bands of gum. Why? I don't know. Insects are caught in these snares. Of what use are they to the plant? Why, none at all; and that's all about it. I leave to others, bolder than myself, the fantastic idea of taking these annular exudations for a digestive fluid which will reduce the captured Midges to soup and make them serve to feed the Silene. Only I warn them that the insects sticking to the plant do not dissolve into broth, but shrivel, quite uselessly, in the sun.

Let us return to the Tachytes, who is also a victim of the vegetable snare. With a sudden flight, a huntress arrives, carrying her drooping prey. She grazes the Silene's lime-twigs too closely. Behold the Mantis caught by the abdomen. For twenty minutes at least the Wasp, still on the wing, tugs at her, tugging again and again, to overcome the cause of the hitch and release the spoil. The hauling-method, a continuation of the flight, comes to nothing; and no other is attempted. At last the insect wearies and leaves the Mantis hanging to the Silene.

Now or never was the moment for the intervention of that tiny glimmer of reason which Darwin so generously grants to animals. Do not, if you please, confound reason with intelligence, as people are too prone to do. I deny the one; and the other is incontestable, within very modest limits. It was, I said, the moment to reason a little, to discover the cause of the hitch and to attack the difficulty at its source. For the Tachytes the matter was of the simplest. She had but to grab the body by the skin of the abdomen immediately above the spot caught by the glue and to pull it towards her, instead of persevering in her flight without releasing the neck. Simple though this mechanical problem was, the insect was unable to solve it, because she was not able to trace the effect back to the cause, because she did not even suspect that the stoppage had a cause.

Ants doting on sugar and accustomed to cross a foot-bridge in order to reach the warehouse are absolutely prevented from doing so when the bridge is interrupted by a slight gap. They would only need a few grains of sand to fill the void and restore the causeway. They do not for a moment dream of it, plucky navvies though they be, capable of raising miniature mountains of excavated soil. We can get them to give us an enormous cone of earth, an instinctive piece of work, but we shall never obtain the juxtaposition of three grains of sand, a reasoned piece of work. The Ant does not reason, any more than the Tachytes.

If you bring up a tame Fox and set his platter of food before him, this creature of a thousand tricks confines himself to tugging with all his might at the leash which keeps him a step or two from his dinner. He pulls as the Tachytes pulls, exhausts himself in futile efforts and then lies down, with his little eyes leering fixedly at the dish. Why does he not turn round? This would increase his radius; and he could reach then the food with his hind-foot and pull it towards him. The idea never occurs to him. Yet another animal deprived of reason.

Friend Bull, my Dog, is no better-endowed, despite his quality as a candidate for humanity. In our excursions through the woods, he happens to get caught by the paw in a wire snare set for rabbits. Like the Tachytes, he tugs at it obstinately and only pulls the noose tighter. I have to release him when he does not himself succeed in snapping the wire by his hard pulling. When he tries to leave the room, if the two leaves of the door are just ajar, he contents himself with pushing his muzzle, like a wedge, into the too narrow aperture. He moves forward, pushing in the direction which he wishes to take. His simple, dog-like method has one unfailing result: the two leaves of the door, when pushed, merely shut still closer. It would be easy for him to pull one of them towards him with his paw, which would make the passage wider; but this would be a movement backward, contrary to his natural impulse; and so he does not think of it. Yet another creature that does not reason.

The Tachytes, who stubbornly persists in tugging at her limed Mantis and refuses to acknowledge any other method of wresting her from the Silene's snare, shows us the Wasp in an unflattering light. What a very poor intellect! The insect becomes only the more wonderful, therefore, when we consider its supreme talent as an anatomist. Many a time I have insisted upon the incomprehensible wisdom of instinct; I do so again at the risk of repeating myself. An idea is like a nail: it is not to be driven in save by repeated blows. By hitting it again and again, I hope to make it enter the most rebellious brains. This time I shall attack the problem from the other end, that is, I shall first allow human knowledge to have its say and shall then interrogate the insect's knowledge.

The outward structure of the Praying Mantis would of itself be enough to teach us the arrangement of the nerve-centres which the Tachytes has to injure in order to paralyse its victim, which is destined to be devoured alive but harmless. A narrow and very long prothorax divides the front pair of legs from the two hinder pairs. There must therefore be an isolated ganglion in front and two ganglia, close to each other, about two-fifths of an inch back. Dissection confirms this forecast completely. It shows us three fairly bulky thoracic ganglia, arranged in the same manner as the legs. The first which actuates the fore-legs, is placed opposite their roots. It is the largest of the three. It is also the most important, for it presides over the insect's weapons, over the two powerful arms, toothed like saws and ending in harpoons. The other two, divided from the first by the whole length of the prothorax, each face the origin of the corresponding legs; consequently they are very near each other. Beyond them are the abdominal ganglia, which I pass over in silence, as the operating insect does not have to trouble about them. The movements of the belly are mere pulsations and are in no way dangerous.

Now let us do a little reasoning on behalf of our non-reasoning insect. The sacrificer is weak; the victim is comparatively powerful. Three strokes of the lancet must abolish all offensive movement. Where will the first stroke be delivered? In front is a real engine of warfare, a pair of powerful shears with toothed jaws. Let the fore-arm close upon the upper arm; and the imprudent insect, crushed between the two saw-blades, will be torn to pieces; wounded by the terminal hook, it will be eviscerated. This ferocious mechanism is the great danger; it is this that must be mastered at the outset, at the risk of life; the rest is less urgent. The first blow of the stylet, cautiously directed, is therefore aimed at the lethal fore-legs, which imperil the vivisector's own existence. Above all, there must be no hesitation. The blow must be accurate then and there, or the sacrificer will be caught in the vice and perish. The two other pairs of legs present no danger to the operator, who might neglect them if she had only her own security to think of; but the surgeon is operating with a view to the egg, which demands complete immobility in the provisions. Their centres of innervation will therefore be stabbed as well, with the leisure which the Mantis, now put out of action, permits. These legs, as well as their nervous centres, are situated very far behind the first point attacked. There is a long neutral interval, that of the prothorax, into which it is quite useless to drive the sting. This interval has to be crossed; by a backward movement conforming with the secrets of the victim's internal anatomy, the second ganglion must be reached and then its neighbour, the third. In short, the surgical operation may be formulated thus: a first stab of the lancet in front; a considerable movement to the rear, measuring about two-fifths of an inch; lastly, two lancet-thrusts at two points very close together. Thus speaks the science of man; thus counsels reason, guided by anatomical structure. Having said this much let us observe the insect's practice.

There is no difficulty about seeing the Tachytes operate in our presence; we have only to resort to the method of substitution, which has already done me so much service, that is, to deprive the huntress of her prey and at once to give her, in exchange, a living Mantis of about the same size. This substitution is impracticable with the majority of the Tachytes, who reach the threshold of their dwelling in a single flight and at once vanish underground with their game. A few of them, from time to time, harassed perhaps by their burden, chance to alight at a short distance from their burrow, or even drop their prey. I profit by these rare occasions to witness the tragedy.

The dispossessed Wasp recognizes instantly, from the proud bearing of the substituted Mantis, that she is no longer embracing and carrying off an inoffensive carcase. Her hovering, hitherto silent, develops a buzz, perhaps to overawe the victim; her flight becomes an extremely rapid oscillation, always behind the quarry. It is as who should say the quick movement of a pendulum swinging without a wire to hang from. The Mantis, however, lifts herself boldly upon her four hind-legs; she raises the fore-part of her body, opens, closes and again opens her shears and presents them threateningly at the enemy; using a privilege which no other insect shares, she turns her head this way and that, as we do when we look over our shoulders; she faces her assailant, ready to strike a return blow wheresoever the attack may come. It is the first time that I have witnessed such defensive daring. What will be the outcome of it all?

The Wasp continues to oscillate behind the Mantis, in order to avoid the formidable grappling-engine; then, suddenly, when she judges that the other is baffled by the rapidity of her manoeuvres, she hurls herself upon the insect's back, seizes its neck with her mandibles, winds her legs round its thorax and hastily delivers a first thrust of the sting, to the front, at the root of the lethal legs. Complete success! The deadly shears fall powerless. The operator then lets herself slip as she might slide down a pole, retreats along the Mantis' back and, going a trifle lower, less than a finger's breadth, she stops and paralyses, this time without hurrying herself, the two pairs of hind-legs. It is done: the patient lies motionless; only the tarsi quiver, twitching in their last convulsions. The sacrificer brushes her wings for a moment and polishes her antennae by passing them through her mouth, an habitual sign of tranquillity returning after the emotions of the conflict; she seizes the game by the neck, takes it in her legs and flies away with it.

What do you say to it all? Do not the scientist's theory and the insect's practice agree most admirably? Has not the animal accomplished to perfection what anatomy and physiology enabled us to foretell? Instinct, a gratuitous attribute, an unconscious inspiration, rivals knowledge, that most costly acquisition. What strikes me most is the sudden recoil after the first thrust of the sting. The Hairy Ammophila, operating on her caterpillar, likewise recoils, but progressively, from one segment to the next. Her deliberate surgery might receive a quasi-explanation if we ascribe it to a certain uniformity. With the Tachytes and the Mantis this paltry argument escapes us. Here are no lancet-pricks regularly distributed; on the contrary, the operating-method betrays a lack of symmetry which would be inconceivable, if the organization of the patient did not serve as a guide. The Tachytes therefore knows where her prey's nerve-centres lie; or, to speak more correctly, she behaves as though she knew.

This science which is unconscious of itself has not been acquired, by her and by her race, through experiments perfected from age to age and habits transmitted from one generation to the next. It is impossible, I am prepared to declare a hundred times, a thousand times over, it is absolutely impossible to experiment and to learn an art when you are lost if you do not succeed at the first attempt. Don't talk to me of atavism, of small successes increasing by inheritance, when the novice, if he misdirected his weapon, would be crushed in the trap of the two saws and fall a prey to the savage Mantis! The peaceable Locust, if missed, protests against the attack with a few kicks; the carnivorous Mantis, who is in the habit of feasting on Wasps far more powerful than the Tachytes, would protest by eating the bungler; the game would devour the hunter, an excellent catch. Mantis-paralysing is a most perilous trade and admits of no half-successes; you have to excel in it from the first, under pain of death. No, the surgical art of the Tachytes is not an acquired art. Whence then does it come, if not from the universal knowledge in which all things move and have their being!

What would happen if, in exchange for her Praying Mantis, I were to give the Tachytes a young Grasshopper? In rearing insects at home, I have already noted that the larvae put up very well with this diet; and I am surprised that the mother does not follow the example of the Tarsal Tachytes and provide her family with a skewerful of Locusts instead of the risky prey which she selects. The diet would be practically the same; and the terrible shears would no longer be a danger. With such a patient would her operating-method remain the same; should we again see a sudden recoil after the first stab under the neck; or would the vivisector modify her art in conformity with the unfamiliar nervous organization?

This second alternative is highly improbable. It would be nonsense to expect to see the paralyser vary the number and the distribution of the wounds according to the genus of the victim. Supremely skilled in the task that has fallen to its lot, the insect knows nothing further.

The first alternative seems to offer a certain chance and deserves a test. I offer the Tachytes, deprived of her Mantis, a small Grasshopper, whose hind-legs I amputate to prevent his leaping. The disabled Acridian jogs along the sand. The Wasp flies round him for a moment, casts a contemptuous glance upon the cripple and withdraws without attempting action. Let the prey offered be large or small, green or grey, short or long, rather like the Mantis or quite different, all my efforts miscarry. The Tachytes recognizes in an instant that this is no business of hers; this is not her family game; she goes off without even honouring my Grasshoppers with a peck of her mandibles.

This stubborn refusal is not due to gastronomical causes. I have stated that the larvae reared by my own hands feed on young Grasshoppers as readily as on young Mantes; they do not seem to perceive any difference between the two dishes; they thrive equally on the game chosen by me and that selected by the mother. If the mother sets no value on the Grasshopper, what then can be the reason of her refusal? I can see only one: this quarry, which is not hers, perhaps inspires her with fear, as any unknown thing might do; the ferocious Mantis does not alarm her, but the peaceable Grasshopper terrifies her. And then, if she were to overcome her apprehensions, she does not know how to master the Acridian and, above all, how to operate upon him. To every man his trade, to every Wasp her own way of wielding her sting. Modify the conditions ever so slightly; and these skilful paralysers are at an utter loss.

To every insect also its own art of fashioning the cocoon, an art which varies greatly, an art in which the larva displays all the resources of its instincts. The Tachytes, the Bembeces, the Stizi, the Palari and other burrowers build composite cocoons, hard as fruit-stones, formed of an encrustation of sand in a network of silk. We are already acquainted with the work of the Bembex. I will recall the fact that their larva first weaves a conical, horizontal bag of pure white silk, with wide meshes, held in place by interlaced threads which fix it to the walls of the cell. I have compared this bag, because of its shape, with a fishtrap. Without leaving this hammock, stretching its neck through the orifice, the worker gathers from without a little heap of sand, which it stores inside its workshop. Then, selecting the grains one by one, it encrusts them all around itself in the fabric of the bag and cements them with the fluid from its spinnerets, which hardens at once. When this task is finished, the house has still to be closed, for it has been wide open all this time to permit of the renewal of the store of sand as the heap inside becomes exhausted. For this purpose a cap of silk is woven across the opening and finally encrusted with the materials which the larva has retained at its disposal.

The Tachytes builds in quite another fashion, although its work, once finished, does not differ from that of the Bembex. The larva surrounds itself, to begin with, about the middle of its body with a silken girdle which a number of threads, very irregularly distributed, hold in place and connect with the walls of the cell. Sand is collected, within reach of the worker, on this general scaffolding. Then begins the work of minor masonry, with grains of sand for rubble and the secretion of the spinnerets for cement. The first course is laid upon the fore-edge of the suspensory ring. When the circle is completed, a second course of grains of sand, stuck together by the fluid silk, is raised upon the hardened edge of what has just been done. Thus the work proceeds, by ring-shaped courses, laid edge to edge, until the cocoon, having acquired half of its proper length, is rounded into a cap and finally is closed. The building-methods of the Tachytes-larva remind me of a mason constructing a round chimney, a narrow tower of which he occupies the centre. Turning on his own axis and using the materials placed to his hand, he encloses himself little by little in his sheath of masonry. In the same way the worker encloses itself in its mosaic. To build the second half of the cocoon, the larva turns round and builds in the same way on the other edge of the original ring. In about thirty-six hours the solid shell is completed.

I am rather interested to see the Bembex and the Tachytes, two workers in the same guild, employ such different methods to achieve the same result. The first begins by weaving an eel-trap of pure silk and next encrusts the grains of sand inside; the second, a bolder architect, is economical of the silk envelope, confines itself to a hanging girdle and builds course by course. The building-materials are the same: sand and silk; the surroundings amid which the two artisans work are the same: a cell in a soil of sandy gravel; yet each of the builders possesses its individual art, its own plan, its one method.

The nature of the food has no more effect upon the larva's talents than the environment in which it lives or the materials employed. The proof of this is furnished by Stiza ruficornis, another builder of cocoons in grains of sand cemented with silk. This sturdy Wasp digs her burrows in soft sandstone. Like the Mantis-killing Tachytes, she hunts the various Mantides of the countryside, consisting mainly of the Praying Mantis; only her large size requires them to be more fully developed, without however having attained the form and the dimensions of the adult. She places three to five of them in each cell.

In solidity and volume her cocoon rivals that of the largest Bembex; but it differs from it, at first sight, by a singular feature of which I know no other example. From the side of the shell, which is uniformly smoothed on every side, a rough knob protrudes, a little clod of sand stuck on to the rest. The work of Stizus ruficornis can at once be recognized, among all the other cocoons of a similar nature, by this protuberance.

Its origin will be explained by the method which the larva follows in constructing its strong-box. At the beginning, a conical bag is woven of pure white silk; you might take it for the initial eel-trap of the Bembeces, only this bag has two openings, a very wide one in front and another, very narrow one at the side. Through the front opening the Stizus provides itself with sand as and when it spends this material on encrusting the interior. This strengthens the cocoon; and the cap which closes it is made next. So far it is exactly like the work of the Bembex. We now have the worker enclosed, engaged in perfecting the inner wall. For these final touches a little more sand is needed. It obtains it from outside by means of the aperture which it has taken the precaution of contriving in the side of its building, a narrow dormer-window just large enough to allow its slender neck to pass. When the store has been taken in, this accessory orifice, which is used only during the last few moments, is closed with a mouthful of mortar, thrust outward from within. This forms the irregular nipple which projects from the side of the shell.

For the present I shall not expatiate further upon Stizus ruficornis, whose complete biography would be out of place in this chapter. I will limit myself to mentioning its method of constructing strong-boxes in order to compare it with that of the Bembex and above all with that of the Tachytes, a consumer, like itself, of Praying Mantes. From this parallel it seems to me to follow that the conditions of life in which men see to-day the origin of instincts—the type of food, the surroundings amid which the larval life is passed, the materials available for a defensive wrapper and other factors which the evolutionists are accustomed to invoke—have no actual influence upon the larva's industry. My three architects in glued sand, even when all the conditions, down to the nature of the provisions, are the same, adopt different means to execute an identical task. They are engineers who have not graduated from the same school, who have not been educated on the same principles, though the lesson of things is almost the same for all of them. The workshop, the work, the provisions have not determined the instinct. The instinct comes first; it lays down laws instead of being subject to them.



CHAPTER 7. CHANGE OF DIET.

Brillat-Savarin, when pronouncing his famous maxim, "Tell me what you eat and I will tell you what you are," certainly never suspected the signal confirmation which the entomological world would bestow upon his saying. Our gastrosopher was speaking only of the culinary caprices of man rendered fastidious by the sweets of life; but he might, in a more serious department of thought, have given his formula a wider and more general bearing and applied it to the dishes which vary so greatly according to latitude, climate and customs; he might above all have taken into his reckoning the harsh realities suffered by the common people, when perhaps his ideal of moral worth would have been found in a platter of chick-peas oftener than in a pot of pate de foie gras. No matter: his aphorism, the mere whimsical sally of an epicure, becomes an imperious truth if we forget the luxury of the table and look into what is eaten by the little world which swarms around us.

To each its mess. The cabbage Pieris consumes the pungent leaves of the Cruciferae as the food of her infancy; the Silkworm disdains any foliage other than that of the mulberry-tree. The Spurge Hawk-moth requires the caustic milk-sap of the tithymals: the Corn-weevil the grain of wheat; the Pea-weevil, the seeds of the Leguminosae; the Balaninus (A genus of Beetles including the Acorn-weevil, the Nut-weevil and others.—Translator's Note.) the hazel-nut, the chestnut, the acorn; the Brachycera (A division of Flies including the Gad-flies and Robber-flies.—Translator's Note.) the clove of garlic. Each has its diet, each its plant; and each plant has its customary guests. Their relations are so precise that in many cases one might determine the insect by the vegetable which supports it, or the vegetable by the insect.

If you know the lily, you may name as a Crioceris the tiny scarlet Scarabaeid that inhabits it and peoples its leaves with larvae which keep themselves cool beneath an overcoat of ordure. (For the Lily-beetle, or Crioceris merdigera, cf. "The Glow-worm and Other Beetles," by J. Henri Fabre, translated by Alexander Teixeira de Mattos: chapters 16 and 17.—Translator's Note.) If you know the Crioceris, you may name as a lily the plant which she devastates. It will not perhaps be the common or white lily, but some other representative of the same family—Turk's cap lily, orange lily, scarlet Martagon, lancifoliate lily, tiger-spotted lily, golden lily—hailing from the Alps or the Pyrenees, or brought from China or Japan. Relying on the Crioceris, who is an expert judge of exotic as well as of native Liliaceae, you may name as a lily the plant with which you are unacquainted and trust the word of this singular botanical master. Whether the flower be red, yellow, ruddy-brown or sown with crimson spots, characteristics so unlike the immaculate whiteness of the familiar flower, do not hesitate, adopt the name dictated by the Beetle. Where man is liable to mistake the insect is never mistaken.

This insect botany, a cause of such grievous tribulations, has always impressed the worker in the fields, who for all that, is a very indifferent observer. The man who was the first to see his cabbage-plot devastated by caterpillars made the acquaintance of the Pieris. Science completed the process, in its desire to serve a useful purpose or merely to seek truth for truth's sake; and to-day the relations between the insect and the plant form a collection of records as important from the philosophical as from the practical, agricultural point of view. What is much less familiar to us, because it touches us less nearly, is the zoology of the insect, that is to say, the selection which it makes, to feed its larva, of this or that animal species, to the exclusion of others. The subject is so vast that a volume were not sufficient to exhaust it; besides, data are lacking in the vast majority of cases. It is reserved for a still very distant future to raise this point of biology to the level already reached by the question of vegetable diet. It will be enough if I contribute a few observations scattered through my writings or my notes.

What does the Wasp addicted to a predatory life eat, of course in the larval state? Now, to begin with, we see natural sections which adopt as their prey different species of one and the same order, in one and the same group. Thus the Ammophilae hunt exclusively the larvae of the night-flying Moths. This taste is shared by the Eumenes, a very different genus. (Cf. "The Mason-wasps" by J. Henri Fabre, translated by Alexander Teixeira de Mattos: chapter 1.—Translator's Note.) The Spheges and Tachytes are addicted to Orthoptera; the Cerceres, apart from a few exceptions, are faithful to the Weevil; both the Philanthi and the Palari capture only Hymenoptera; the Pompili specialize in hunting the Spider; the Astata revels in the flavour of Bugs; the Bembeces want Flies and nothing else; the Scoliae enjoy the monopoly of the Lamellicorn-grubs; the Pelopaei favour the young Epeirae (Or Garden Spiders. Cf. "The Life of the Spider": chapters 9 to 14 and appendix.—Translator's Note.), the Stizi vary in opinion: of the two in my neighbourhood, one, S. ruficornis, fills her larder with Mantes and the other, S. tridentatus, fills it with Cicadellae (Cf. "The Life of the Grasshopper": chapter 20.—Translator's Note.); lastly, the Crabronidae (Any Flies akin to the House-fly.—Translator's Note.). levy tribute upon the rabble of the Muscidae. (Hornets.—Translator's Note.)

Already you see what a magnificent classification of these game-hunters might be made with a faithfully listed bill of fare. Natural groups stand out, characterized merely by the identity of their victuals. I trust that the methodical science of the future will take account of these gastronomic laws, to the great relief of the entomological novice, who is too often hampered by the snares of the mouth-parts, the antennae and the nervures of the wings. I call for a classification in which the insect's aptitudes, its diet, its industry and its habits shall take precedence of the shape of a joint in its antennae. It will come; but when?

If from generalities we descend to details, we shall see that the very species may, in many instances, be determined from the nature of its victuals. The number of burrows of Philanthus apivorus which I have inspected since I have been rummaging the hot roadside embankments, to enquire into their population, would seem hyperbolical were I able to state the figures. (For the Bee-eating Philanthus cf. Chapter 10 of the present volume.—Translator's Note.) They must amount, it seems to me, to thousands. Well, in this multitude of food-stores, whether recent or ancient, uncovered for a purpose or encountered by chance, I have not once, not as often as once, discovered other remains than those of the Hive-bee: the imperishable wings, still connected in pairs, the cranium and thorax enveloped in a violet shroud, the winding-sheet which time throws over these relics. To-day as when I was a beginner, ever so long ago; in the north as in the south of the country which I explored; in mountainous regions as on the plains, the Philanthus follows an unvarying diet: she must have the Hive-bee, always the Bee and never any other, however closely various other kinds of game resemble the Bee in quality. If, therefore, when exploring sunny banks, you find beneath the soil a small parcel of mutilated Bees, that will be enough to point to the existence of a local colony of Philanthus apivorus. She alone knows the recipe for making potted Bee-meat. The Crioceris was but now teaching us all about the lily family; and here the mildewed body of the Bee tells us of the Philanthus and her lair.

Similarly the female Ephippiger helps us to identify the Languedocian Sphex: her relics, the cymbals and the long sabre, are the unmistakable sign of the cocoon to which they adhere. The black Cricket, with his red-braided thighs, is the infallible label of the Yellow-winged Sphex; the larva of Oryctes nasicornis tells us of the Garden Scolia as certainly as the best description; the Cetonia-grub proclaims the Two-banded Scolia and the larva of the Anoxia announces the Interrupted Scolia.

After these exclusive ones, who disdain to vary their meals, let us mention the eclectics, who, in a group which is generally well-defined, are able to select among different kinds of game appropriate to their bulk. The Great Cerceris (Cerceris tuberculata. Cf. "The Hunting Wasps": chapters 2 and 3.—Translator's Note.) favours above all Cleonus ophthalmicus, one of the largest of our Weevils; but at need she accepts the other Cleoni, as well as the kindred genera, provided that the capture be of an imposing size. Cerceris arenaria (Cf. idem: chapter 1.—Translator's Note.) extends her hunting-grounds farther afield: any Weevil of average dimensions is to her a welcome capture. The Buprestis-hunting Cerceris adopts all the Buprestes indiscriminately, so long as they are not beyond her strength. The Crowned Philanthus (P. coronatus, FAB.) fills her underground warehouses with Halicti chosen among the biggest. (Cf. "Bramble-bees and Others" by J. Henri Fabre, translated by Alexander Teixeira de Mattos: chapters 12 to 14.—Translator's Note.) Much smaller than her kinswoman, Philanthus raptor, LEP., stores away Halicti chosen among the less large species. Any adult Acridian approaching an inch in length suits the White-banded Sphex. The various tidae of the neighbourhood are admitted to the larder of Stizus ruficornis and of the Mantis-hunting Tachytes on the sole condition of being young and tender. The largest of our Bembeces (B. rostrata, FAB., and B. bidentata, VAN DER LIND (For the Rostrate Bembex and the Two-pronged Bembex, cf. "The Hunting Wasps": chapter 14.—Translator's Note.)) are eager consumers of Gad-flies. With these chief dishes they associate relishes levied indifferently from the rest of the Fly clan. The Sandy Ammophila (A. sabulosa, VAN DER LIND (Cf. idem: chapter 13.—Translator's Note.)) and the Hairy Ammophila (A. hirsuta, KIRB.) cram into each burrow a single but corpulent caterpillar, always of the Moth tribe and varying greatly in coloration, which denotes distinct species. The Silky Ammophila (A. holosericea, VAN DER LIND. (Cf. idem: chapter 14.—Translator's Note.)) has a better assorted diet. She requires for each banqueter three or four items, which include the Measuring-worms, or Loopers, and the caterpillars of ordinary Moths, all of which are equally appreciated. The Brown-winged Solenius (S. fascipennis, LEP.), who elects to dwell in the soft dead wood of old willow-trees, has a marked preference for Virgil's Bee, Eristalis tenax (Actually the Common Drone-fly and somewhat resembling a Bee in appearance. Cf. "The Hunting Wasps": chapter 14.—Translator's Note.), willingly adding, sometimes as a side-dish, sometimes as the principal game, Helophilus pendulus, whose costume is very different. On the faith of indistinguishable remains, we must no doubt enter a number of other Flies in her game-book. The Golden-mouthed Hornet (Crabro chrysostomus, LEP.) another burrower in old willow-trees, prefers the Syrphi, without distinction of species. (The Syrphi, like the Eristales, resemble Bees through having the abdomen transversely banded with yellow.—Translator's Note.) The Wandering Solenius (S. vagus, LEP. (For this Fly-hunting insect cf. "Bramble-bees and Others": chapters 1 and 3.—Translator's Note.)), an inmate of the dry bramble-stems and of the dwarf-elder, lays under contribution for her larder the genera Syritta, Sphaerophoria, Sarcophaga, Syrphus, Melanophora, Paragus and apparently many others. The species which recurs most frequently in my notes is Syritta pipiens.

Without pursuing this tedious list any farther, we plainly perceive the general result. Each huntress has her characteristic tastes, so much so that, when we know the bill of fare, we can tell the genus and very often the species of the guest, thus proving the proud truth of the maxim, "Tell me what you eat and I will tell you what you are."

There are some which always need the same prey. The offspring of the Languedocian Sphex religiously consume the Ephippiger, that family dish so dear to their ancestors and no less dear to their descendants; no innovation in the ancient usages can tempt them. Others are better suited by variety, for reasons connected with flavour or with facility of supply; but then the selection of the game is kept within fixed limits. A natural group, a genus, a family, more rarely almost a whole order: this is the hunting-ground beyond which poaching is strictly forbidden. The law is absolute; and one and all scrupulously refrain from transgressing it.

In the place of the Praying Mantis, offer the Mantis-hunting Tachytes an equivalent in the shape of a Locust. She will scorn the morsel, though it would seem to be of excellent flavour, seeing that Panzer's Tachytes prefers it to any other form of game. Offer her a young Empusa, who differs so widely from the Mantis in shape and colour: she will accept without hesitation and operate before your eyes. Despite its fantastic appearance, the Devilkin is instantly recognized by the Tachytes as a Mantid and therefore as game falling within her scope.

In exchange for her Cleonus, give to the Great Cerceris a Buprestis, the delight of one of her near kinsfolk. She will have nothing to say to the sumptuous dish. Accept that! She, a Weevil-eater! Never in this world! Present her with a Cleonus of a different species, or any other large Weevil, of a sort which she has most probably never seen before, since it does not figure on the inventory of the provisions in her burrows. This time there is no show of disdain: the victim is seized and stabbed in the regulation manner and forthwith stored away.

Try to persuade the Hairy Ammophila that Spiders have a nutty flavour, as Lalande asserts; and you will see how coldly your hints are received. (Joseph Jerome Le Francois de Lalande (1732-1807), the astronomer. Even after he had achieved his reputation, he sought means, outside the domain of science, to make himself talked about and found these in the display partly of odd tastes, such as that for eating Spiders and caterpillars, and partly of atheistical opinions.—Translator's Note.) Try merely to convince her that the caterpillar of a Butterfly is as good to eat as the caterpillar of a Moth. You will not succeed. But, if you substitute for her underground larva, which I suppose to be grey, another underground larva striped with black, yellow, rusty-red or any other tint, this change of coloration will not prevent her from recognizing, in the substituted dish, a victim to her liking, an equivalent of her Grey Worm.

So with the rest, so far as I have been able to experiment with them. Each obstinately refuses what is alien to her hunting-preserves, each accepts whatever belongs to them, always provided that the game substituted is much the same in size and development as that whereof the owner has been deprived. Thus the Tarsal Tachytes, an appreciative epicure of tender flesh, would not consent to replace her pinch of young Acridian-grubs with the one big Locust that forms the food of Panzer's Tachytes; and the latter, in her turn, would never exchange her adult Acridian for the other's menu of small fry. The genus and the species are the same, but the age differs; and this is enough to decide the question of acceptance or refusal.

When its depredations cover a somewhat extensive group, how does the insect manage to recognize the genera, the species composing her allotted portion and to distinguish them from the rest with an assured vision which the inventory of her burrows proves never to be at fault? Is it the general appearance that guides her? No, for in some Bembex-burrows we shall find Sphaerophoriae, those slender, thong-like creatures, and Bombylii, looking like velvet pincushions; no again, for in the pits of the Silky Ammophila we shall see, side by side, the caterpillar of the ordinary shape and the Measuring-worm, a living pair of compasses which progresses by alternately opening out and closing; no, once more, for in the storerooms of Stizus ruficornis and the Mantis-hunting Tachytes we see stacked beside the Mantis the Empusa, her unrecognizable caricature.

Is it the colouring? Not at all. There is no lack of instances. What a variety of hues and metallic reflections, distributed in a host of different fashions, appear in the Buprestes that are hunted by the Cerceris celebrated by Leon Dufour. (Jean Marie Leon Dufour (1780-1865) was an army surgeon who served with distinction in several campaigns and subsequently practised as a doctor in the Landes. He attained great eminence as a naturalist. Cf. "The Hunting Wasps": chapter 1; also "The Life of the Spider": chapter 1.—Translator's Note.) A painter's palette, containing crushed gold, bronze, ruby and amethyst, would find it difficult to rival these sumptuous colours. Nevertheless the Cerceris makes no mistake: all this nation of insects, so indifferently attired, represents to her, as to the entomologist, the nation of the Buprestes. The inventory of the Hornet's larder will include Diptera clad in grey or russet frieze; others are girdled with yellow, flecked with white, adorned with crimson lines; others are steel-blue, ebony black, or coppery green; and underneath this variety of dissimilar costumes we find the invariable Fly.

Let us take a concrete example. Ferrero's Cerceris (C. Ferreri, VAN DER LIND) consumes Weevils. Her burrows are usually lined with Phynotomi and Sitones both an indeterminate grey, and Otiorhynchi, black or tan-coloured. Now I have sometimes happened to unearth from her cells a collection of veritable jewels which, thanks to their bright metallic lustre, made a most striking contrast with the sombre Otiorhynchus. These were the Rhynchites (R. betuleti), who roll the vine-leaves into cigars. Equally magnificent, some of them were azure blue, others copper gilt, for the cigar-roller has a twofold colouring. How did the Cerceris manage to recognize in these jewels the Weevil, the near relative of the vulgar Phynotomus? Any such encounters probably found her lacking in expert knowledge; her race cannot have handed down to her other than very indeterminate propensities, for she does not appear to make frequent use of the Rhynchites, as is proved by my infrequent discovery of them amid the mass of my numerous excavations. For the first time, perhaps, passing through a vineyard, she saw the rich Beetle gleaming on a leaf; it was not for her a dish in current consumption, consecrated by the ancient usages of the family. It was something novel, exceptional, extraordinary. Well, this extraordinary creature is recognized with certainty as a Weevil and stored away as such. The glittering cuirass of the Rhynchites goes to take its place beside the grey cloak of the Phynotomus. No, it is not the colour that guides the choice.

Neither is it the shape. Cerceris arenaria hunts any medium-sized Weevil. I should be putting the reader's patience to too great a test if I attempted to give in this place a complete inventory of the specimens identified in her larder. I will mention only two, which my latest searches around my village have revealed. The Wasp goes hunting on the holm-oaks of the neighbouring hills the Pubescent Brachyderes (B. pubescens) and the Acorn-weevil (Balaninus glandium). What have these two Beetles in common as regards shape? I mean by shape not the structural details which the classifier examines through his magnifying-glass, not the delicate features which a Latreille would quote when drawing up a technical description, but the general picture, the general outline that impresses itself upon the vision even of an untrained eye and makes the man who knows nothing of science and above all the child, a most perspicacious observer, connect certain animals together.

In this respect, what have the Brachyderes and the Balaninus in common in the eyes of the townsman, the peasant, the child or the Cerceris? Absolutely nothing. The first has an almost cylindrical figure; the second, squat, short and thickset, is conical in front and elliptical, or rather shaped like the ace of hearts, behind. The first is black, strewn with cloudy, mouse-grey spots; the second is yellow ochre. The head of the first ends in a sort of snout; the head of the second tapers into a curved beak, slender as a horse-hair and as long as the rest of the body. The Brachyderes has a massive proboscis, cut off short; the Balaninus seems to be smoking an insanely long cigarette-holder.

Who would think of connecting two creatures so unlike, of calling them by the same name? Outside the professional classifiers, no one would dare to. The Cerceris, more perspicacious, knows each of them for a Weevil, a quarry with a concentrated nervous system, lending itself to the surgical feat of her single stroke of the lancet. After obtaining an abundant booty at the cost of the blunt-mouthed insect, with which she sometimes stuffs her cellars to the exclusion of any other fare, according to the hazards of the chase, she now suddenly sees before her the creature with the extravagant proboscis. Accustomed to the first, will she fail to know the second? By no means: at the first glance she recognizes it as her own; and the cell already furnished with a few Brachyderes receives its complement of Balanini. If these two species are to seek, if the burrows are far from the holm-oaks, the Cerceris will attack Weevils displaying the greatest variety of genus, species, form and coloration, levying tribute indifferently on Sitones, Cneorhini, Geonemi, Otiorhynchi, Strophosomi and many others.

In vain do I rack my brains merely to guess at the signs upon which the huntress relies as a guide, without going outside one and the same group, in the midst of such a variety of game; above all by what characteristics she recognizes as a Weevil the strange Acorn Balaninus, the only one among her victims that wears a long pipe-stem. I leave to evolutionism, atavism and other transcendental "isms" the honour and also the risk of explaining what I humbly recognize as being too far beyond my grasp. Because the son of the bird-catcher who imitates the call of his victims has been fed on roast Robins, Linnets and Chaffinches, shall we hastily conclude that this education through the stomach will enable him later, without other initiation than that of the spit, to know his way about the ornithological groups and to avoid confusing them when his turn comes to set his limed twigs? Will the digesting of a ragout of little birds, however often repeated by him or his ascendants, suffice to make him a finished bird-catcher? The Cerceris has eaten Weevil; her ancestors have all eaten Weevil, religiously. If you see in this the reason that makes the Wasp a Weevil-expert endowed with a perspicacity unrivalled save by that of a professional entomologist, why should you refuse to admit that the same consequences would follow in the bird-catcher's family?

I hasten to abandon these insoluble problems in order to attack the question of provisions from another point of view. Every Hunting Wasp is confined to a certain genus of game, which is usually strictly limited. She pursues her appointed quarry and regards anything outside it with suspicion and distaste. The tricks of the experimenter, who drags her prey from under her and flings her another in exchange, the emotions of the possessor deprived of her property and immediately recovering it, but under another form, are powerless to put her on the wrong scent. Obstinately she refuses whatever is alien to her portion; instantly she accepts whatever forms part of it. Whence arises this insuperable repugnance for provisions to which the family is unaccustomed? Here we may appeal to experiment. Let us do so: its dictum is the only one that can be trusted.

The first idea that presents itself and the only one, I think, that can present itself is that the larva, the carnivorous nurseling, has its preferences, or we had better say its exclusive tastes. This kind of game suits it; that does not; and the mother provides it with food in conformity with its appetites, which are unchangeable in each species. Here the family dish is the Gad-fly; elsewhere it is the Weevil; elsewhere again it is the Cricket, the Locust and the Praying Mantis. Good in themselves, in a general way, these several victuals may be noxious to a consumer who is not used to them. The larva which dotes on Locust may find caterpillar a detestable fare; and that which revels in caterpillar may hold Locust in horror. It would be hard for us to discover in what manner Cricket-flesh and Ephippiger-flesh differ as juicy, nourishing foodstuffs; but it does not follow that the two Sphex-wasps addicted to this diet have not very decided opinions on the matter, or that each of them is not filled with the highest esteem for its traditional dish and a profound dislike for the other. There is no discussing tastes.

Moreover, the question of health may well be involved. There is nothing to tell us that the Spider, that treat for the Pompilus, is not poison, or at least unwholesome food, to the Bembex, the lover of Gad-flies; that the Ammophila's succulent caterpillar is not repugnant to the stomach of the Sphex fed upon the dry Acridian. The mother's esteem for one kind of game and her distrust of another would in that case be due to the likes and dislikes of her larvae; the victualler would regulate the bill of fare by the gastronomic demands of the victualled.

This exclusiveness of the carnivorous larva seems all the more probable inasmuch as the larva reared on vegetable food refuses in any way to lend itself to a change of diet. However pressed by hunger, the caterpillar of the Spurge Hawk-moth, which browses on the tithymals, will allow itself to starve in front of a cabbage leaf which makes a peerless meal for the Pieris. Its stomach, burned by pungent spices, will find the Crucifera insipid and uneatable, though its piquancy is enhanced by essence of sulphur. The Pieris, on its part, takes good care not to touch the tithymals: they would endanger its life. The caterpillar of the Death's-head Hawk-moth requires the solanaceous narcotics, principally the potato, and will have nothing else. All that is not seasoned with solanin it abhors. And it is not only larvae whose food is strongly spiced with alkaloids and other poisonous substances that refuse any innovation in their food; the others, even those whose diet is least juicy, are invincibly uncompromising. Each has its plant or its group of plants, beyond which nothing is acceptable.

I remember a late frost which had nipped the buds of the mulberry-trees during the night, just when the first leaves were out. Next day there was great excitement among my neighbours: the Silk-worms had hatched and the food had suddenly failed. The farmers had to wait for the sun to repair the disaster; but how were they to keep the famishing new-born grubs alive for a few days? They knew me for an expert in plants; by collecting them as I walked through the fields I had earned the name of a medical herbalist. With poppy-flowers I prepared an elixir which cleared the sight; with borage I obtained a syrup which was a sovran remedy for whooping-cough; I distilled camomile; I extracted the essential oil from the wintergreen. In short, botany had won for me the reputation of a quack doctor. After all, that was something.

The housewives came in search of me from every point of the compass and with tears in their eyes explained the situation. What could they give their Silk-worms while waiting for the mulberry to sprout afresh? It was a serious matter, well worthy of commiseration. One was counting on her batch to buy a length of cloth for her daughter, who was on the point of getting married; another told me of her plans for a Pig to be fattened against the coming winter; all deplored the handful of crown-pieces which, hoarded in the hiding-place in the cupboard, would have afforded help in difficult times. And, full of their troubles, they unfolded, before my eyes, a scrap of flannel on which the vermin were swarming:

"Regardas, moussu! Venoun d'espeli; et ren per lour douna! Ah, pecaire!" "Look, sir! The frost has come and we've nothing to give them! Oh, what a misfortune!"

Poor people! What a harsh trade is yours: respectable above all others, but of all the most uncertain! You work yourselves to death; and, when you have almost reached your goal, a few hours of a cold night, which comes upon you suddenly, destroys your harvest. To help these afflicted ones seemed to me a very difficult thing. I tried, however, taking botany as my guide; it suggested to me, as substitutes for the mulberry, the members of closely-related families: the elm, the nettle-tree, the nettle, the pellitory. Their nascent leaves, chopped small, were offered to the Silk-worms. Other and far less logical attempts were made, in accordance with the inspiration of the individuals. Nothing came of them. To the last specimen, the new-born Silk-worms died of hunger. My renown as a quack must have suffered somewhat from this check. Was it really my fault? No, it was the fault of the Silk-worm, which remained faithful to its mulberry leaf.

It was therefore in nearly the certainty of non-fulfilment that I made my first attempts at rearing carnivorous larvae with a quarry which did not conform with the customary regimen. For conscience' sake, more or less perfunctorily, I endeavoured to achieve something that seemed to me bound to end in pitiful failure. Only the Bembex-wasps, which are plentiful in the sand of the neighbouring hills, might still afford me, without too prolonged a search, a few subjects on which to experiment. The Tarsal Bembex furnished me with what I wanted: larvae young enough to have still before them a long period of feeding and yet sufficiently developed to endure the trials of a removal.

These larva are exhumed with all the consideration which their delicate skin demands; a number of head of game are likewise unearthed intact, having been recently brought by the mother. They consist of various Diptera, including some Anthrax-flies. (Cf. "The Life of the Fly": chapters 2 and 4.—Translator's Note.) An old sardine-box, containing a layer of sifted sand and divided into compartments by paper partitions, receives my charges, who are isolated one from another. These Fly-eaters I propose to turn into Grasshopper-eaters; for their Bembex-diet I intend to substitute the diet of a Sphex or a Tachytes. To save myself tedious errands devoted to provisioning the refectory, I accept what good fortune offers me at the very threshold of my door. A green Locustid, with a short sabre bent into a reaping-hook, Phaneroptera falcata, is ravaging the corollae of my petunias. Now is the time to indemnify myself for the damage which she has caused me. I pick her young, half to three-quarters of an inch in length; and I deprive her of movement, without more ado, by crushing her head. In this condition she is served up to the Bembex-larvae in place of their Flies.

If the reader has shared my convictions of failure, convictions based on very logical motives, he will now share my profound surprise. The impossible becomes possible, the senseless becomes reasonable and the expected becomes the opposite of the real. The dish served on the Bembeces' table for the first time since Bembeces came into the world is accepted without any repugnance and consumed with every mark of satisfaction. I will here set down the detailed diary of one of my guests; that of the others would only be a repetition, save for a few variations.

2 AUGUST, 1883.—The larva of the Bembex, as I extract it from its burrow, is about half-developed. Around it I find only some scanty relics of its meals, consisting chiefly of Anthrax-wings, half-diaphanous and half-clouded. The mother would appear to have completed the victualling by fresh contributions, added day by day. I give the nurseling, which is an Anthrax-eater, a young Phaneroptera. The Locustid is attacked without hesitation. This profound change in the character of its victuals does not seem in the least to disturb the larva, which bites straight into the rich morsel with its mandibles and does not let go until it has exhausted it. Towards evening the drained carcase is replaced by another, quite fresh, of the same species but bulkier, measuring over three-quarters of an inch.

3 AUGUST.—Next day I find the Phaneroptera devoured. Nothing remains but the dry integuments, which are not dismembered. The entire contents have disappeared; the game has been emptied through a large opening made in the belly. A regular Grasshopper-eater could not have operated more skilfully. I replace the worthless carcase by two small Locustidae. At first the larva does not touch them, being amply sated with the copious meal of the day before. In the afternoon, however, one of the items is resolutely attacked.

4 AUGUST.—I renew the victuals, although those of the day before are not finished. For the rest, I do the same daily, so that my charge may constantly have fresh food at hand. High game might upset its stomach. My Locustidae are not victims at the same time living and inert, operated upon according to the delicate method of the insects that paralyse their prey; they are corpses, procured by a brutal crushing of the head. With the temperature now prevailing, flesh soon becomes tainted; and this compels me frequently to renew the provisions in my sardine-box refectory. Two specimens are served up. One is attacked soon afterwards; and the larva clings to it assiduously.

5 AUGUST.—The ravenous appetite of the start is becoming assuaged. My supplies may well be too generous; and it might be prudent to try a little dieting after this Gargantuan good cheer. The mother certainly is more parsimonious. If all the family were to eat at the same rate as my guest, she would never be able to keep pace with their demands. Therefore, for reasons of health, this is a day of fasting and vigil.

6 AUGUST.—Supplies are renewed with two Phaneropterae. One is consumed entirely; the other is bitten into.

7 August.—To-day's ration is tasted and then abandoned. The larva seems uneasy. With its pointed mouth it explores the walls of its chamber. This sign denotes the approach of the time for making the cocoon.

8 AUGUST.—During the night the larva has spun its silken eel-trap. It is now encrusting it with grains of sand. Then follow, in due time, the normal phases of the metamorphosis. Fed on Locustidae, a diet unknown to its race, the larva passes through its several stages without any more difficulty than its brothers and sisters fed on Flies.

I obtained the same success in offering young Mantes for food. One of the larvae thus served would even incline me to believe that it preferred the new dish to the traditional diet of its race. Two Eristales, or Drone-flies, and a Praying Mantis an inch long composed its daily allowance. The Drone-flies are disdained from the first mouthful; and the Mantis, already tasted and apparently found excellent, causes the Fly to be completely forgotten. Is this an epicure's preference, due to the greater juiciness of the flesh? I am not in a position to say. At all events, the Bembex is not so infatuated with Fly as to refuse to abandon it for other game.

The failure which I foresaw has proved a magnificent success. It is fairly convincing, is it not? Without the evidence of experiment, what can we rely upon? Beneath the ruins of so many theories which appeared to be most solidly erected I should hesitate to admit that two and two make four if the facts were not before me. My argument had the most tempting probability on its side, but it had not the truth. As it is always possible to find reasons after the event in support of an opinion which one would not at first admit, I should now argue as follows:

The plant is the great factory in which are elaborated, with mineral materials, the organic principles which are the materials of life. Certain products are common to the whole vegetable series, but others, far less numerous, are prepared in special laboratories. Each genus, each species has its trade-mark. Here essential oils are manufactured; here alkaloids; here starches, fatty substances, resins, sugars, acids. Hence result special energies, which do not suit every herbivorous animal. It assuredly requires a stomach made expressly for the purpose to digest aconite, colchicum, hemlock or henbane; those who have not such a stomach could never endure a diet of that sort. Besides, the Mithridates fed on poison resist only a single toxin. (Mithridates VI. King of Pontus (d. B.C. 63) is said to have secured immunity from poison by taking increased doses of it.—Translator's Note.) The caterpillar of the Death's-head Hawk-moth, which delights in the solanin of the potato, would be killed by the acrid principle of the tithymals that form the food of the Spurge-caterpillar. The herbivorous larvae are therefore perforce exclusive in their tastes, because different genera of vegetables possess very different properties.

With this variety in the products of the plant, the animal, a consumer far more than a producer, contrasts the uniformity in its own products. The albumen in the egg of the Ostrich or the Chaffinch, the casein in the milk of the Cow or the Ass, the muscular flesh of the Wolf or the Sheep, the Screech-owl or the Field-mouse, the Frog or the Earth-worm: these remain albumen, casein or fibrin, edible if not eaten. Here are no excruciating condiments, no special acridities, no alkaloids fatal to any stomach other than that of the appointed consumer; so that animal food is not confined to one and the same eater. What does not man eat, from that delicacy of the arctic regions, soup made of Seal's blood and a scrap of Whale-blubber wrapped in a willow-leaf for a vegetable, to the Chinaman's fried Silk-worm or the Arab's dried Locust? What would he not eat, if he had not to overcome the repugnance dictated by habit rather than by actual necessity? The prey being uniform in its nutritive principles, the carnivorous larva ought to accommodate itself to any sort of game, above all if the new dish be not too great a departure from consecrated usage. Thus should I argue, with no less probability on my side, had I to begin all over again. But, as all our arguments have not the value of a single fact, I should be forced in the end to resort to experiment.

I did so the next year, on a larger scale and with a greater variety of subjects. I shrink from a continuous narrative of my experiments and of my personal education in this new art, where the failure of one day taught me the way to succeed on the morrow. It would be long and tedious. Enough if I briefly state my results and the conditions which must be fulfilled in order to run the delicate refectory as it should be run.

And, first, we must not dream of detaching the egg from its natural prey to lay it on another. The egg adheres pretty firmly, by its cephalic pole, to the quarry. To remove it from its place would inevitably jeopardize its future. I therefore let the larva hatch and acquire sufficient strength to bear the removal without peril. For that matter, my excavations most often provide me with my subjects in the form of larvae. I adopt for rearing-purposes the larvae that are a quarter to a half developed. The others are too young and risky to handle, or too old and limited to a short period of artificial feeding.

Secondly, I avoid bulky heads of game, a single one of which would suffice for the whole growing-stage. I have already said and I here repeat how nice a matter it is to consume a victim which has to keep fresh for a couple of weeks and not to finish dying until it is almost entirely devoured. Death here leaves no corpse; when life is extinct, the body has disappeared, leaving only a shred of skin. Larvae with only one large prey have a special art of eating, a dangerous art, in which a clumsy bite would prove fatal. If bitten before the proper time at such a point, the victim becomes putrid, which promptly causes death by poisoning in the consumer. When diverted from its plan of attack, deprived of its clue, the larva is not always able to rediscover the lawful morsels in good time and is killed by the decomposition of its badly dissected prey. What will happen if the experimenter gives it a game to which it is not accustomed? Not knowing how to eat it according to rule, the larva will kill it; and by next day the victuals will have become so much toxic putrescence. I have already told how I found it impossible to rear the Two-banded Scolia on Oryctes-larvae, fastened down to deprive them of movement, or even on Ephippigers, paralysed by the Languedocian Sphex. In both cases the new diet was accepted without hesitation, a proof that it suited the nurseling; but in a day or two putrescence supervened and the Scolia perished on the fetid morsel. The method of preserving the Ephippiger, so well known to the Sphex, was unknown to my boarder; in this was enough to convert a delicious food into poison.

Even so did my other attempts miscarry wretchedly, attempts at feeding with the single dish consisting of one big head of game to replace the normal ration. Only one success is recorded in my notebooks, but that was so difficult that I would not undertake to obtain it a second time. I succeeded in feeding the larva of the Hairy Ammophila with an adult black Cricket, who was accepted as readily as the natural game, the caterpillar.

To avoid putrefaction of victuals which last overlong and are not consumed according to the method indispensable to their preservation, I employ small game, each piece of which can be finished by the larva at a single sitting, or at most in a single day. It matters little then that the victim is slashed and dismembered at random; decomposition has no time to seize upon its still quivering tissues. This is the procedure of those larvae which gulp down their food, snapping at random without distinguishing one part from another, such as the Bembex-larvae, which finish the Fly into which they have bitten before beginning another in the heap, or the Cerceris-larvae, which drain their Weevils methodically one after another. With the first strokes of the mandibles the victim broached may be mortally wounded. This is no disadvantage: a brief spell suffices to make use of the corpse, which is saved from putrefaction by being promptly consumed. Close beside it, the other victims, quite alive though motionless, await their respective turns and supply reserves of victuals which are always fresh.

I am too unskilful a butcher to imitate the Wasp and myself to resort to paralysis; moreover, the caustic liquid injected into the nerve-centres, ammonia in particular, would leave traces of smell or flavour which might put off my boarders. I am therefore compelled to deprive my insects of the power of movement by killing them outright. This makes it impracticable to provide a sufficiency of provisions beforehand in a single supply: while one item of the ration was being consumed the rest would spoil. One expedient alone remains to me, one which entails constant attendance: it is to renew the provisions each day. When all these conditions are fulfilled, the success of artificial feeding is still not without its difficulties; nevertheless, with a little care and above all plenty of patience, it is almost certain.

It was thus that I reared the Tarsal Bembex, which eats Anthrax-flies and other Diptera, on young Locustidae or Mantidae; the Silky Ammophila, whose diet consists chiefly of Measuring-worms, on small Spiders; the pot-making Pelopaeus, a Spider-eater, on tender Acridians; the Sand Cerceris, a passionate lover of Weevils, on Halicti; the Bee-eating Philanthus, which feeds exclusively on Hive-bees, on Eristales and other Flies. Without succeeding in my final aim, for reasons which I have just explained, I have seen the Two-banded Scolia feasting greedily on the grub of the Oryctes, which was substituted for that of the Cetonia, and putting up with an Ephippiger taken from the burrow of the Sphex; I have been present at the repast of three Hairy Ammophilae accepting with an excellent appetite the Cricket that replaced their caterpillar. One of them, as I have related, contrived to keep its ration fresh, which enabled it to reach its full development and to spin its cocoon.

These examples, the only ones to which my experiments have extended hitherto, seem to me sufficiently convincing to allow me to conclude that the carnivorous larva does not have exclusive tastes. The ration supplied to it by the mother, so monotonous, so limited in quality, might be replaced by others equally to its taste. Variety does not displease the larva; it does it as much good as uniformity; indeed, it would be of greater benefit to the race, as we shall see presently.



CHAPTER 8. A DIG AT THE EVOLUTIONISTS.

To rear a caterpillar-eater on a skewerful of Spiders is a very innocent thing, unlikely to compromise the security of the State; it is also a very childish thing, as I hasten to confess, and worthy of the schoolboy who, in the mysteries of his desk, seeks as best he may some diversion from the fascinations of his exercise in composition. And I should not have undertaken these investigations, still less should I have spoken them, not without some satisfaction, if I had not discerned, in the results obtained in my refectory, a certain philosophic import, involving, so it seemed to me, the evolutionary theory.

It is assuredly a majestic enterprise, commensurate with man's immense ambitions, to seek to pour the universe into the mould of a formula and submit every reality to the standard of reason. The geometrician proceeds in this manner: he defines the cone, an ideal conception; then he intersects it by a plane. The conic section is submitted to algebra, an obstetrical appliance which brings forth the equation; and behold, entreated now in one direction, now in another, the womb of the formula gives birth to the ellipse, the hyperbola, the parabola, their foci, their radius vectors, their tangents, their normals, their conjugate axes, their asymptotes and the rest. It is magnificent, so much so that you are overcome by enthusiasm, even when you are twenty years old, an age hardly adapted to the austerities of mathematics. It is superb. You feel as if you were witnessing the creation of a world.

As a matter of fact, you are merely observing the same idea from different points of view, which are illumined by the successive phases of the transformed formula. All that algebra unfolds for our benefit was contained in the definition of the cone, but it was contained as a germ, under latent forms which the magic of the calculus converts into explicit forms. The gross value which our mind confided to the equation it returns to us, without loss or gain, in coins stamped with every sort of effigy. And here precisely is that which constitutes the inflexible rigour of the calculus, the luminous certainty before which every cultivated mind is forced to bow. Algebra is the oracle of the absolute truth, because it reveals nothing but what the mind had hidden in it under an amalgam of symbols. We put 2 and 2 into the machine; the rollers work and show us 4. That is all.

But to this calculus, all-powerful so long as it does not leave the domain of the ideal, let us submit a very modest reality: the fall of a grain of sand, the pendular movement of a hanging body. The machine no longer works, or does so only by suppressing almost everything that is real. It must have an ideal material point, an ideal rigid thread, an ideal point of suspension; and then the pendular movement is translated by a formula. But the problem defies all the artifices of analysis if the oscillating body is a real body, endowed with volume and friction; if the suspensory thread is a real thread, endowed with weight and flexibility; if the point of support is a real point, endowed with resistance and capable of deflection. So with other problems, however simple. The exact reality escapes the formula.

Yes, it would be a fine thing to put the world into an equation, to assume as the first principle a cell filled with albumen and by transformation after transformation to discover life under its thousand aspects as the geometrician discovers the ellipse and the other curves by examining his conic section. Yes, it would be magnificent and enough to add a cubit to our stature. Alas, how greatly must we abate our pretensions! The reality is beyond our reach when it is only a matter of following a grain of dust in its fall; and we would undertake to ascend the river of life and trace it to its source! The problem is a more arduous one than that which algebra declines to solve. There are formidable unknown quantities here, more difficult to decipher than the resistances, the deflections and the frictions of the pendulum. Let us eliminate them, that we may more easily propound the theory.

Very well; but then my confidence in this natural history which repudiates nature and gives ideal conceptions precedence over real facts is shaken. So, without seeking the opportunity, which is not my business, I take it when it presents itself; I examine the theory of evolution from every side; and, as that which I have been assured is the majestic dome of a monument capable of defying the ages appears to me to be no more than a bladder, I irreverently dig my pin into it.

Here is the latest dig. Adaptability to a varied diet is an element of well-being in the animal, a factor of prime importance for the extension and predominance of its race in the bitter struggle for life. The most unfortunate species would be that which depended for its existence on a diet so exclusive that no other could replace it. What would become of the Swallow if he required, in order to live, one particular Gnat, a single Gnat, always the same? When once this Gnat had disappeared—and the life of the Mosquito is not a long one—the bird would die of starvation. Fortunately for himself and for the happiness of our homes, the Swallow gulps them all down indiscriminately, together with a host of other insects that perform aerial ballets. What would become of the Lark were his gizzard able to digest only one seed, invariably the same? When the season for this seed was over—and the season is always a short one—the haunter of the furrows would perish.

Is not man's complaisant stomach, adapted to the largest variety of nourishment, one of his great zoological privileges? He is thus rendered independent of climates, seasons and latitudes. And the Dog: how is it that of all the domestic animals he alone is able to accompany us everywhere, even on the most arduous expeditions? The Dog again is omnivorous and therefore a cosmopolitan.

The discovery of a new dish, said Brillat-Savarin, is of greater importance to humanity than the discovery of a new planet. The aphorism is nearer to the truth than it appears to be in its humorous form. Certainly the man who was the first to think of crushing wheat, kneading flour and cooking the paste between two hot stones was more deserving than the discoverer of the two-hundredth asteroid. The invention of the potato is certainly as valuable as that of Neptune, glorious as the latter was. All that increases our alimentary resources is a discovery of the first merit. And what is true of man cannot be other than true of animals. The world belongs to the stomach which is independent of specialities. This truth is of the kind that has only to be stated to be proved.

Let us now return to our insects. If I am to believe the evolutionists, the various game-hunting Wasps are descended from a small number of types, which are themselves derived, by an incalculable number of concatenations, from a few amoebae, a few monera and lastly from the first clot of protoplasm which was casually condensed. Let us not go back as far as that; let us not plunge into the fogs where illusion and error too easily find a lurking-place. Let us consider a subject with exact limits to it; this is the only way to understand one another.

The Sphegidae are descended from a single type, which itself was already a highly-developed descendant and, like its successors, fed its family on prey. The close similarity in form, in colouring and, above all, in habits seem to refer the Tachytes to the same origin. This is ample; let us be satisfied with it. And now please tell me, what did this prototype of the Sphegidae hunt? Was its diet varied or uniform? If we cannot decide, let us examine the two cases.

The diet was varied. I heartily congratulate the first born of the Sphex-wasps. She enjoyed the most favourable conditions for leaving a prosperous offspring. Accommodating herself to any kind of prey not disproportionate to her strength, she avoided the dearth of a given species of game at this or that time and in this or that place; she always found the wherewithal to endow her family magnificently, they being, for that matter, fairly indifferent to the nature of the victuals, provided that these consisted of fresh insect-flesh, as the tastes of their cousins many times removed prove to this day. This matriarch of the Sphex clan bore within herself the best chances of assuring victory to her offspring in that pitiless fight for existence which eliminates the weakly and incapable and allows none but the strong and industrious to survive; she possessed an aptitude of great value which atavism could not fail to hand down and which her descendants, who are greatly interested in preserving this magnificent inheritance, must have permanently adopted and even accentuated from one generation to the next, from one branch, one offshoot, to another.

Instead of this unscrupulously omnivorous race, levying booty upon every kind of game, to its very great advantage, what do we see to-day? Each Sphex is stupidly limited to an unvarying diet; she hunts only one kind of prey, though her larva accepts them all. One will have nothing but the Ephippiger and must have a female at that; another will have nothing but the Cricket. This one hunts the Locust and nothing else; that one the Mantis and the Empusa. Yet another is addicted to the Grey Worm and another to the Looper.

Fools! How great was your mistake in allowing the wise eclecticism of your ancestress, whose relics now repose in the hard mud of some lacustrian stratum, to become obsolete! How much better would things be for you and yours! Abundance is assured; painful and often fruitless searches are avoided; the larder is crammed without being subject to the accidents of time, place and climate. When Ephippigers run short, you fall back upon Crickets; when there are no Crickets, you capture Grasshoppers. But no, my beautiful Sphex-wasps, you were not such fools as that. If in our days you are each confined to a standing family-dish, it is because your ancestress of the lacustrian schists never taught you variety.

Could she have taught you uniformity? Let us suppose that the Sphex of antiquity, a novice in the gastronomic art, prepared her potted meats with a single kind of game, no matter what. It was then her descendants who, subdivided into groups and constituted into so many distinct species by the slow travail of the centuries, realized that in addition to the ancestral fare there existed a host of other foods. Tradition being abandoned, there was nothing to guide their choice. They therefore tried a bit of everything in the way of insect game, at hap-hazard; and each time the larva, whose tastes alone had to be consulted, was satisfied with the food supplied, as it is to-day in the refectory provisioned by my care.

Every attempt led to the invention of a new dish, an important event, according to the masters, an inestimable resource for the family, who were thereby delivered from the menace of death and enabled to thrive over large areas whence the absence or rarity of a uniform game would have excluded it. And, after making use of a host of different viands in order to attain the culinary variety which is to-day adopted by the whole of the Sphex nation, lo and behold, each species confines itself to a single sort of game, outside which every specimen is obstinately refused, not at table, of course, but in the hunting-field! By your experiments, from age to age, to have discovered variety in diet; to have practised it, to the great advantage of your race, and to end up with uniformity, the cause of decadence; to have known the excellent and to repudiate it for the middling: oh, my Sphex-wasps, it would be stupid if the theory of evolution were correct!

To avoid insulting you and also from respect for common sense, I prefer therefore to believe that, if in our days you confine your hunting to a single kind of game, it is because you have never known any other. I prefer to believe that your common ancestress, your precursor, whether her tastes were simple or complex, is a pure chimera, for, if they were any relationship between you, having tested everything in order to arrive at the actual food of each species, having eaten everything and found it grateful to the stomach, you would now, from first to last, be unprejudiced consumers, omnivorous progressives. I prefer to believe, in short, that the theory of evolution is powerless to explain your diet. This is the conclusion drawn from the dining-room installed in my old sardine-box.



CHAPTER 9. RATIONING ACCORDING TO SEX.

Considered in respect of quality, the food has just disclosed our profound ignorance of the origins of instinct. Success falls to the blusterers, to the imperturbable dogmatists, from whom anything is accepted if only they make a little noise. Let us discard this bad habit and admit that really, if we go to the bottom of things, we know nothing about anything. Scientifically speaking, nature is a riddle to which human curiosity finds no definite solution. Hypothesis follows hypothesis; the theoretical rubbish-heap grows bigger and bigger; and still truth escapes us. To know how to know nothing might well be the last word of wisdom.

Considered in respect of quantity, the food sets us other problems, no less obscure. Those of us who devote ourselves assiduously to studying the customs of the game-hunting Wasps soon find our attention arrested by a very remarkable fact, at the time when our mind, refusing to be satisfied with sweeping generalities, which our indolence too readily makes shift with, seeks to enter as far as possible into the secret of the details, so curious and sometimes so important, as and when they become better-known to us. This fact, which has preoccupied me for many a long year, is the variable quantity of the provisions packed into the burrow as food for the larva.

Each species is scrupulously faithful to the diet of its ancestors. For more than a quarter of a century I have been exploring my district; and I have never known the diet to vary. To-day, as thirty years ago, each huntress must have the game which I first saw her pursuing. But, though the nature of the victuals is constant, the quantity is not so. In this respect the difference is so great that he would need to be a very superficial observer who should fail to perceive it on his first examination of the burrows. In the beginning, this difference, involving two, three, four times the quantity and more, perplexed me extremely and led me to the conclusions which I reject to-day.

Here, among the instances most familiar to me, are some examples of these variations in the number of victims provided for the larva, victims, of course, very nearly identical in size. In the larder of the Yellow-winged Sphex, after the victualling is completed and the house shut up, two or three Crickets are sometimes found and sometimes four. Stizus ruficornis (Cf. "The Hunting Wasps": chapter 20; also "Bramble-bees and Others": chapter 9.—Translator's Note.), established in some vein of soft sandstone, places three Praying Mantes in one cell and five in another. Of the caskets fashioned by Amedeus' Eumenes (Cf. "The Mason-wasps": chapter 1.—Translator's Note.) out of clay and bits of stone, the more richly endowed contain ten small caterpillars, the more poorly furnished five. The Sand Cerceris (Cf. "The Hunting Wasps": chapter 2.—Translator's Note.) will sometimes provide a ration of eight Weevils and sometimes one of twelve or even more. My notes abound in abstracts of this kind. It is unnecessary for the purpose in hand to quote them all. It will serve our object better if I give the detailed inventory of the Bee-eating Philanthus and of the Mantis-hunting Tachytes, considered especially with regard to the quantity of the victuals.

The slayer of Hive-bees is frequently in my neighbourhood; and I can obtain from her with the least trouble the greatest number of data. In September I see the bold filibuster flying from clump to clump of the pink heather pillaged by the Bee. The bandit suddenly arrives, hovers, makes her choice and swoops down. The trick is done: the poor worker, with her tongue lolling from her mouth in the death-struggle, is carried through the air to the underground den, which is often a very long way from the spot of the capture. The trickling of earthy refuse, on the bare banks, or on the slopes of footpaths, instantly reveals the dwellings of the ravisher; and, as the Philanthus always works in fairly populous colonies, I am able, by noting the position of the communities, to make sure of fruitful excavations during the forced inactivity of winter.

The sapping is a laborious task, for the galleries run to a great depth. Favier wields the pick and spade; I break the clods which he brings down and open the cells, whose contents—cocoons and remnants of provisions—I at once pour into a little screw of paper. Sometimes, when the larva is not developed, the stack of Bees is intact; more often the victuals have been consumed; but it is always possible to tell the number of items provided. The heads, abdomens and thoraxes, emptied of their fleshy substance and reduced to the tough outer skin, are easily counted. If the larva has chewed these overmuch, the wings at least are left; these are sapless organs which the Philanthus absolutely scorns. They are likewise spared by moisture, putrefaction and time, so much so that it is no more difficult to take an inventory of a cell several years old than one of a recent cell. The essential thing is not to overlook any of these tiny relics while placing them in the paper bag, amid the thousand incidents of the excavation. The rest of the work will be done in the study, with the aid of the lens, taking the remains heap by heap; the wings will be separated from the surrounding refuse and counted in sets of four. The result will give the amount of the provisions. I do not recommend this task to any one who is not endowed with a good stock of patience, nor above all to any one who does not start with the conviction that results of great interest are compatible with very modest means.

My inspection covers a total of one hundred and thirty-six cells, which are divided as in the table below:

2 cells each containing 1 Bee 52 cells each containing 2 Bees 36 cells each containing 3 Bees 36 cells each containing 4 Bees 9 cells each containing 5 Bees 1 cell containing 6 Bees —- 136

The Mantis-hunting Tachytes consumes its heap of Mantes, the horny envelope included, without leaving any remains but scanty crumbs, quite insufficient to establish the number of items provided. After the meal is completed, any inventory of the rations becomes impossible. I therefore have recourse to the cells which still contain the egg or the very young larva and, above all, to those whose provisions have been invaded by a tiny parasitic Gnat, a Tachina (Cf. "The Hunting Wasps": chapters 4 and 16.—Translator's Note.), which drains the game without cutting it up and leaves the whole skin intact. Twenty-five larders, put to the count, give me the following result:

8 cells each containing 3 items 5 cells each containing 4 items 4 cells each containing 6 items 3 cells each containing 7 items 2 cells each containing 8 items 1 cell containing 9 items 1 cell containing 12 items 1 cell containing 16 items —- 25

The predominant game is the Praying Mantis, green; next comes the Grey Mantis, ash-coloured. A few Empusae make up the total. The specimens vary in dimensions within fairly elastic limits: I measure some which are a third to a half inch long, averaging two-thirds to one inch long, and some which are two-fifths, averaging three quarters. I see pretty plainly that their number increases in proportion as their size diminishes, as though the Tachytes were seeking to make up for the smallness of the game by increasing the amount; none the less I find it quite impossible to detect the least equivalence by combining the two factors of number and size. If the huntress really estimates the provisions, she does so very roughly; her household accounts are not at all well kept; each head of game, large or small, must always count as one in her eyes.

Put on my guard, I look to see whether the honey-gathering Bees have a double service, like the game-hunting Wasps'. I estimate the amount of honeyed paste; I gauge the cups intended to contain it. In many cases the result resembles the first obtained: the abundance of provisions varies from one cell to another. Certain Osmiae (O. cornuta and O. tricornis (Cf. "Bramble-bees and Others": passim; and, in particular, chapters 3 to 5.—Translator's Note.)) feed their larvae on a heap of pollen-dust moistened in the middle with a very little disgorged honey. One of these heaps may be three or four times the size of some other in the same group of cells. If I detach from its pebble the nest of the Mason-bee, the Chalicodoma of the Walls, I see cells of large capacity, sumptuously provisioned; close beside these I see others, of less capacity, with victuals parsimoniously allotted. The fact is general; and it is right that we should ask ourselves the reason for these marked differences in the relative quantity of foodstuffs and for these unequal rations.