From the facts which we have just recited we may conclude that insects are individually capable of modifying their habits, and the experiments cited in the preceding chapter prove that they can acquire new ones. This is what Bethe demands, and since the German author calls himself "a man of good intent, open to proof," I suppose that he will cease to see in insects simply reflex machines.

Observation has not established that in a state of nature, without the help of man, insects are capable of acquiring new habits, but we cannot doubt that this is the case. We have seen that certain females of Sphex flavipennis know how to overcome the artifice which consists in taking away their prey while they are making a visit to the burrow. After this surprise has been repeated two or three times they omit the visit and immediatelv store awav their captures. But the wind may play the same part as the experimenter, and then, the insect responds with a change of habit. In the beginning the new habit will be transitory, but, repeated at intervals, it may become definitive. It would be very interesting to know if the two Ammophilas observed by the Peckhams and by Williston would use a stone to perfect another burrow ; but these wasps rebel at captivity, and the experiment was not tried,- without doubt, because it was very difficult.

We know that new habits change very quickly to automatism with the articulates. If they pass on to progeny, if they are born in the young of the parents which have acquired them, they will have changed in character and will have become instincts. But can acquired characters become truly hereditary ? That is the question which we must consider.

Arnold Pictet, an. eminent Swiss biologist, in 1905 raised caterpillars of different species, offering them plants which differed as much as possible from those which they naturally eat. The caterpillars of the diurnal butterflies almost all rebelled at these attempts, but those of several night-flying moths took the new food, though not without some difficulty. With these last "the acquaintance with a new food was often transmitted by heredity, and individuals whose parents had become habituated to a strange diet ate the same leaves, after hatching, with much greater facility." Of all the rearings carried on by Pictet, the most significant were with the caterpillars of Ocneria dispar, popularly known in our country under the name of the zigzag and in America under that of the gipsy moth. The caterpillars of this insect have a bad reputation because they despoil trees of their foliage, especially oaks and birches. Pictet offered them the leaves of the walnut, which have a very special flavor, and they naturally disdained them. It was with the greatest difficulty that he brought the young caterpillars to feed upon them. When older they always were able to accustom themselves to them, but the moths which issued from this rearing were less highly colored and smaller, and the generation to which they gave birth and which was given the same food could not be raised to the egg-laying point. To render this second generation fertile Pictet nourished the caterpillars with oak leaves at first, after which the rearing was carried on with walnut leaves. In -a rearing conducted to the fourth generation (first walnut, second oak, third walnut, fourth walnut) they had become so accustomed to the new diet that the moths had resumed the form and the size of the insects which had normal food. The experiment was not carried further, but it is probable that the caterpillars produced by the fourth generation would have remained faithful to their new diet.

An experiment by Paul Marchai (1908) led to definite results of even more importance. This distinguished biologist induced a scale insect of the peach (Lacanium corni) to live on the honey-locust, or false acacia. This was not done without trouble, and only certain larvae issuing from the laying of the Leaanium were able to adapt themselves to the new food; but with these there was complete success, beyond all expectation. Becoming adult the following year, the scale insects reproduced upon the locust, and their young had so entirely accepted this host plant that it was impossible to rear to sexual maturity those which he transplanted to the original host (peach). Very quickly the new habit became hereditary and hereditary to the point where the insect could not return to its primitive habit.

In the two preceding examples there is the inheritance of becoming accustomed to a new diet. Here are others in which heredity takes a new turn. The larvae of the willow chrysomelid (Phratora vitellinœ) feed normally on the leaves of a willow (Salix fragilis) which has leaves whose lower surface is smooth. They attach themselves to this lower surface, attacking the epidermis to reach the parenchyma, and skeletonize the leaves. Some young larvae were established by Schroeder (1903) upon a willow with downy leaves,-Salico viminalis. They adapted themselves without difficulty, cutting and pushing aside with their heads the long pilose clothing of the lower face of the leaves and forming with it and with the skin of this surface a sort of blanket which covered them completely. These larvae devoured the tissue of the leaves, and one of them made a mine three centimeters long. The adults coming from these larvae deposited 127 layings on viminalis and 219 on fragilis. In the course of successive generations, all raised on viminalis, the number of layings placed upon this species increased progressively while those laid on fragilis diminished. In the fourth generation Schroeder obtained fifteen layings all upon viminalis, and the larvae coming from these layings adopted the mining type of life. It is evident, adds the author, that similar modifications are produced in nature, especially when the normal food plant is lacking.

Still more astonishing is the hereditary transformation of habits obtained recently by Arnold Pictet in 1911 with Lasiocampa quercus. The caterpillars of this moth attack deciduous trees and bushes, notably the oak, and eat the leaves at their edges. When placed upon pine they try to feed, but their mandibles cannot open widely enough. Many die, but certain individuals succeed in existing by attacking the ends of the pine needles, where they dig into the parenchyma. In the second generation the caterpillars had become adapted to the new diet, but they were adapted beyond return, for they perished when placed upon a tree with flat leaves, or they attacked the leaves at the tip and hollowed out the interior as the caterpillars of the preceding generation had learned to do in the pine needles.

As Darwin has noted, a striking example of the heredity of acquired habits is shown by the mulberry silkworm. It is not to be doubted that this insect formerly lived in a wild state and that it could then live under conditions which would be fatal to it now. What has happened to these insects under the influence of long domestication? Perez says :

Life on the trees has become impossible, at least in our climate ; they cannot resist the wind, which would blow them to the ground ; they cannot sustain the rays of the sun, which the wild caterpillars could avoid by going around to the lower face of the leaves. If high temperature and free air give silkworms a remarkable appetite, one notices on the other hand that whether they are sleeping or are in the state of repose preceding the molt, they are placed in the least favorable circumstances upon the upper surface of the leaves, exposed to the rain. A long domestication has evidently made them lose a host of instinctive habits whose loss [before domestication] would have led infallibly to their total destruction even before the end of the caterpillar stage.