Addison, Dr. Watts, Dr. Johnson, all refer to the provident habits of the ant in storing up food for future use; and it is quite clear that such a habit was considered a usual one amongst ants in general, and not one as occasionally occurring in a few species. Latreille, Kirby and Spence, Huber, Frederick Smith, and others, emphatically deny such a habit to the species found in Europe. The late Colonel Sykes, indeed, speaks of a species of Indian ant (Atta providens), and Dr. Jerdon of three species that harvest seeds on a large scale, collecting grain and stealing seeds, which they put away in their granaries. There can be no doubt of the fact; the question is what is the motive? The most recent English writer who has studied this debated subject, and has himself examined many ants' nests in the South of Europe, is the late Mr. J. Traherne Moggridge, F.L.S., whose very interesting work is before us as we write. Now Mr. Moggridge mentions four bonâ fide harvesting ants of the Riviera—namely, Atta barbara, two varieties; A. structor, an ant very similar to barbaru, and a minute yellow ant, the large workers of which have gigantic heads named Pheidole (or Atta) migacephala. In the nests of all these ants were found masses of seeds of various plants 'carefully stored in chambers.' The plants of which the seeds have been found in ants' nests by Mr. Moggridge belonged to eighteen distinct families; seeds of furmitory, medick, mallow, wild lentil, spiny broom, amaranth, pellitory, wild sarsaparilla, spirally twisted links of crane's bill, capsules of chickweed, shepherds' purse, orange pips, haricot beans, wheat, oats, &c. &c., are enumerated with those of other plants. Are these substances intended for food or not? if not, for what are they intended? Of course, a preliminary question suggests itself: What is the structure of an ant's mouth, and is it capable of gnawing hard substances such as grains of wheat? An ant's mouth consists of a pair of pincer-like mandibles, jaws, maxillary palpi or feelers, a labial palpi, and a tongue, upper and lower lips. Let us hear what the highest authority on the natural history of the ant, M. Forel, says on this point.

'Ants are for the most part omnivorous; that is certain, but they are unable to chew. Their mandibles never serve them for eating purposes; this fact, demonstrated by Huber, is perfectly certain; the most assiduous observation has confirmed it. The disproportion, moreover, between them (the mandibles) and the jaws would at once render this evident; they always keep fixed and immovable whilst the ant is eating. Ordinarily the mouth is closed by the upper lip, which falls upon it below and behind, completely covering the forepart of the jaws and of the lower lip. When an ant wishes to eat it makes a very complex movement of the pharynx, which thrusts forward

the tongue and all the surrounding parts, whilst raising the upper lip like a lid. But the jaws are much too short, too weak, too membranous to grind any solid substance whatever; they can only take into their mouth, by a backward and forward movement, a liquid, or at the most a pappy substance. Observation shows that it is the tongue especially which subserves the ants when they eat; they employ it precisely like dogs when they lap or lick the bottom of a plate; I made this comparison before I knew that it had already been made by Lespès, and I could not express myself more clearly. When the ants have to deal with a solid body which contains liquid, as an insect, for example, they first of all tear it with their mandibles and afterwards lap its contents. These facts have already been known to and well described by Huber, and subsequently confirmed by Lespès; but Léon Dafour thinks that ants are capable of chewing, and d'Esterne accuses them of devouring. I cannot insist too much upon this point, for it is incredible to see how many people persist in remaining in error on this subject.' (P. 108-109.)

Mr. Moggridge made some experiments in feeding ants. He cut out from the centre of a grain of millet, which had begun to sprout, a minute ball of flour; the ants (Atta structor) immediately seized it and set to work upon it; a similar ball from a grain which had sprouted, was also partially eaten, but the hard dry flour taken from a grain in its natural state not moistened, was at once rejected and thrown on the rubbish 'heap.' He tells us that the fat oily seeds of the hemp were eagerly taken, though not softened by water, their peculiar structure allowing the ants to scrape off particles, as in the case of the ball of flour of the sprouted millet. Now all this

confirms the assertions of Huber, Lespès, and Forel; ants cannot chew, but they can lap and cause to disappear food already reduced to a kind of pulp; so that it would seem that ants do occasionally convey into their nests seeds, which, when they begin to sprout, assume in parts a pulpy consistency, and are available for food; but this does not prove that the introduction of seeds into the nests has always for a motive, on the part of the ant, a desire to feed upon them, for Forel assures us with regard to this very species, Atta (aphænogaster) structor, that not only are grains of corn found in its nest, but also little round stones, and small shells of molluscs, which no one will ever suppose the ant could use as food. And here, again, a remarkable fact presents itself. Mr. Moggridge tells us that it is extremely rare to find other than round and intact seeds in the granaries,' and he concludes that the ants exercise some mysterious power over them which checks the tendency to germinate. This retardation of the germinating process, if really a fact, is most extraordinary. The ants cannot use the

6

grains as food before germination; their motive, therefore, must rest in the fact that they are not yet ready for them, having sufficient meat already in the larder ready for consumption; or, as Mr. Moggridge says, if simultaneous germination took place in all the seeds in the granaries after the lapse of a fixed interval, the provisions would have to be consumed at stated. periods and to be frequently renewed; but this is not the 'case.' These granaries are placed from an inch and a half 'to six inches below the surface, and are all horizontal; they are of various sizes and shapes, the average granary being 'about as large as a gentleman's gold watch.' (Harvesting Ants, p. 23.) These storing ants, Mr. Moggridge tells us, never look at the aphides and cocci so eagerly sought after by other kinds of ants; so we suppose their food consists only of germinating seeds of various kinds. The name of the 'provident one,' Mr. Moggridge allows, is only fully deserved by a limited number of ants; but why some ants should require food for winter's use, whilst others should lie dormant and require no food at all, at present must remain a problem yet to be solved. The evidence which Mr. Moggridge brings forward satisfactorily establishes the fact that ants do occasionally store up vegetable food; but it shows also that such habits are by no means prevalent amongst the whole. family, but that, on the contrary, they are rare and exceptional.

On the interesting question as to mutual social affection, and the extraordinary powers of communication with which ants have long been credited by careful observers, we must now go to Sir John Lubbock, who has with much labour and assiduous application carried out some very original observations, not only amongst the ants, but also amongst wasps and bees. Of the power to communicate and receive information possessed by ants, Huber tells us that he has frequently seen the antennæ used on the field of battle to intimate approaching danger, and to ascertain their own party when mingled 'with the enemy; they are also employed in the exterior of the 'ant-hill to warn their companions of the presence of the sun, 'so favourable to the development of the larva; in their ex'cursions and emigrations to indicate their route; in their ' recruitings to determine the time of their departure.' (P. 206.) Other entomologists, besides Huber, state that the social Hymenoptera can communicate their ideas, and that this communication takes place by means of their antennæ. Sir John Lubbock, commenting upon the above quotation from Huber, whilst allowing the statements to be most interesting,

regrets that Huber has not given in detail the evidence on which those statements rest, and that he nowhere gives experiments he had himself conducted.

6

As regards the affection of ants and their behaviour to wounded comrades, instances of which are detailed by Huber, Latreille, M. de Saint Fargeau and others; as that an ant never meets with a wounded comrade without taking her up and placing her in the nest; or that, the antennæ of an ant having been cut off, a companion, 'pitying its sufferings, anointed the wounded part with a drop of transparent fluid from ' its mouth,' Sir John's experiences have been of the opposite character, and he states that he has often been surprised that in certain cases, ants render one another so little assistance.' Ants may not unfrequently be seen with the heads of others hanging on to their legs for a considerable time, and as this must certainly be very inconvenient, it seems remarkable that their friends should not relieve them of such an awkward 'encumbrance.' Having tried various experiments by immersing ants in water, in order to test the tenderness attributed to those insects, Sir John records in nearly all cases, none took any notice;' still he admits that individual differences may exist for in two cases an immersed ant was picked up and rescued and taken to the nest-and humorously remarks that there may be Priests and Levites and good Samaritans among 'them as among men.'

6

Our own experiments on this point-though they have been limited compared with those Sir John Lubbock has conducted -have convinced us that this compassion for either wounded or drowning companions has been considerably overestimated ; we have often immersed ants in water, and never observed that their companions take the slightest notice of them.

On the question as to ants being able to recognise friends after a separation of some months, Sir John's experiments bear out Huber's observations so far as this, that the friends are not killed while strangers are. After separating some ants for a period of four months, Huber brought them together again, when they immediately recognised one another and fell 6 to mutual caresses with their antennæ.' In Sir John's observations, a friend when restored to her old companions was generally left unattacked, but there were no signs of welcome, no greeting around a returned friend;' a stranger, on the contrary, was, as a rule, at once seized upon and sometimes killed, though occasionally, after due punishment, forgiven and received as a friend into the community.

As to those delicate, and doubtless important organs, the

antennæ of insects, while all entomologists regard them as organs of touch, there is considerable difference of opinion as to what other special function they may have. Some regard them as olfactory, others as auditory organs. Sir John suggests that in those insects in which the sense of hearing is highly developed, the antennæ may serve as ears, while in those which have a very delicate sense of smell, they may act as olfactory organs. The same instrument may serve for different purposes; the different senses according to some physiologists being only a modification of a similar organic instrument adapted to different purposes. From their position on the head and the constant use made of them, the antennæ are, no doubt, important organs of sense, and Sir John Lubbock considers and we think he has proved his case-that they are organs of smell. To all sounds, whether loud inharmonious noises or sounds produced by a complete set of tuning-forks, ants would seem to be almost as deaf as posts; they never took the slightest notice of any of these sounds: but Sir John cautiously and justly adds that the insects possibly if not probably may be deaf to sounds which we hear, and yet hear others to which we are deaf. The question as to the faculty of hearing possessed by insects is one of the most curious and puzzling subjects connected with their history. Sir John has secured the promise of the valuable assistance of Mr. Spottiswoode, with whom he hopes to make further experiments on this subject.

6

To experiments with various agents, as essence of cloves, lavender-water, peppermint-water, and other strong scents, to which the ants were subjected, the insects were acutely responsive; one of the antenna was touched with a feather dipped in essence of musk; it was slowly retracted and drawn quite back; the other antenna was touched- the ant started away, apparently smarting;' but when the antennæ were. softly touched with an unperfumed feather they did not move at all. No one,' he adds, who watched the behaviour of ants under these circumstances could have the slightest doubt as to"their power of smell.' We have not space to follow Sir Johnfurther in his interesting experiments, which certainly must, to some extent, modify our conception of certain high qualities which have been attributed to ants in general. At the same time it is necessary to bear in mind the existence amongst ants of individual variations of habits and character, even in the case of the same species; and to take care, in making experiments, to generalise with hesitation and great caution. Nor can we follow M. Forel further in his admirable monograph of

VOL. CXLV. NO. CCXCVII.

H