The following very singular account is given by M. Erber[75] of the habits of Cteniza ariana, which he watched in the island of Tinos. I quote from the abstract given in the Zoological Record cited below:—"At night these spiders come out of their nests, fasten the open trap-door to neighbouring objects, and spin a net, about six inches long by scarcely half an inch in height. In the morning the nets were removed, and Erber believes that the net of each night is added to the trap-door. He found eggs at the bottom of the tubes, attached singly to threads, to the number of about sixty. The young seem to form dwellings very early."

It would be very interesting to know whether these nocturnal habits are also found in our spiders on the Riviera.

I have been favoured[76] with a sight of an unpublished manuscript by Mr. Hansard giving an account of his observations on Cteniza fodiens, made in Corfu. This gentleman states that some of these spiders which he kept in captivity, used to come out at night, and might sometimes be surprised roaming about the room at a very early hour in the morning. He, however, relates that he had received from a friend an account of a trap-door spider inhabiting the island of Formosa, in the China seas, which constructed nests similar to those of Cteniza fodiens, but which were habitually to be seen outside their nests in the daytime, attracting attention by "staring at" any one who might approach, and then hurrying back to their nests and closing their doors after them.

Lady Parker has also told me of some black trap-door spiders which were so common about Paramatta, near Sydney, in Australia, that scarcely any one paid attention to them, and which might habitually be seen out on the garden paths in the daytime near their holes, to which they would run in all haste when alarmed. The eye of the passer-by was attracted by the open doors, which were about the size of a sixpence, and fall over backwards when the spider makes her exit, but when closed, on her return, they fit so neatly that it is extremely difficult to detect them.

It will, perhaps, have been observed that I have throughout spoken of the female spider only, scarcely any allusion having been made to the male. The truth is that, though I have carefully searched for them, I have never been able to secure more than a single male spider.[77] During the winter, spring, and late autumn (October) the female appears to live solitary, in the daytime at least, and the male probably hides in the crevices of old walls and in similar places. I have diligently turned over piles of stone, greatly to the annoyance of many little scorpions, but have never secured, or even seen, another male spider. This is the more to be regretted as the species of trap-door spider are much better characterized in the male than in the female sex, the bulb-like enlargement which is found at the end of the palpi in the former taking on a great variety of forms, each of which is distinctive.

M. de Walckenaer[78] says:—"C'est toujours pendant la nuit que ces aranéides travaillent à leurs habitations et courent après leur proie. C'est en Août que la Mygale maçonne (Nemesia—or Mygale—cæmentaria) atteint toute sa grosseur.... En Septembre elle devient mère et méchante en même temps ... les mouches, les moucherons, les petits vers lui servent de pâture; elle les prend dans les filets qu'elle étend et attache sur les inégalités des terres voisines de sa demeure. Elle vit après la ponte en société avec son mâle. Dorthès a vu plusieurs fois, dans la même habitation, le mâle et la femelle avec une trentaine de petits."

Any one, therefore, who has an opportunity of examining the nests during the early autumn, might perhaps, discover the happy families spoken of by M. Dorthès, but which it has never been my good fortune to see. It is not known positively whether the male spider ever assists in the construction of the nest, but, as we know that the female is able to make it without his aid, there seems no reason to suppose that he does.

I have seen the female Nemesia meridionalis construct a trap-door in captivity, after having been placed on a flower-pot full of earth in which I had made a cylindrical hole.[79] She quickly disappeared into this hole, and, during the night following the day of her capture, she made a thin web over the aperture, into which she wove any materials which came to hand. The trap-door at this stage resembled a rudely constructed, horizontal, geometrical web, attached by two or three threads to the earth at the mouth of the hole, while in this web were caught the bits of earth, roots, moss, leaves &c. which the spider had thrown into it from above. After the second night the door appeared nearly of the normal texture and thickness, but in no case would it open completely, and it seemed that the spider was too much disgusted with her quarters to think it worth while to make a perfect door. I believe that when a door is finished the few threads which served as supports and connected it with the earth on either side of the hinge are severed, and this is borne out by the following instance. While I was at work one evening drawing the spider's nest concealed in the plant of ceterach fern (Plate XI., fig. A, p. 105) which I had dug out for the purpose, I detected something moving at the mouth of a tiny hole [just large enough to admit a crowquill pen] in the mass of earth on the opposite side of the fern, to that in which the large trap-door lay.

The lamp-light fell full upon it, and I soon saw that the moving object was a very small spider, not bigger than that drawn at B 2 in Plate IX., which was at work in the mouth of its tube. Whether I had, in removing this mass of earth, destroyed the door I cannot say, but it is certain that the opening of the tube was completely uncovered, and it soon became apparent that the little spider was intent upon remedying this deficiency. After a few threads had been spun from side to side of the tube I watched the spider make one or two hasty sorties, apparently spinning all the while, and finally I saw her gather up an armful, as it were, of earth and lay this on the web. After this the occupant of the tube was concealed, but I could see from the movement of the particles of earth that they were being consolidated, and that the weaving of the under surface of the door was being completed. Next morning I was able to lift up the door, which had the form of a small cup of silk, in which the earth lay. It was then soft and pliant, but in ten days time it had hardened and become a very fair specimen of a minute cork door (see figs. A 1, A 2, of Plate XI.).

On one occasion a captive Nemesia meridionalis employed some pieces of scarlet braid which I had purposely strewed, along with bits of moss and fragments of leaves, in a circle round the opening of, and about two inches away from, the hole.

It is probable that these spiders have in times past learned by experience that they cannot do better than take such materials as come to hand, as these will ordinarily serve for the concealment of their door.

However, these trap-door spiders do seem to exercise some discrimination in the choice of materials; for I have observed several instances in which, when the door of a cork nest has been removed, if the door was originally covered with moss, moss will again be used in its reconstruction, even though the mouth of the tube be then surrounded by bare earth.

Thus, for example, in one case where I had cut out a little clod of mossy earth, about two inches thick and three square on the surface, containing the top of the tube and the moss-covered cork door of N. cæmentaria, I found, on revisiting the place six days later, that a new door had been made, and that the spider had mounted up to fetch moss from the undisturbed bank above, planting it in the earth which formed the crown of the door.[80] Here the moss actually called the eye to the trap, which lay in the little plain of brown earth made by my digging.

I have seen the same thing happen when the door of N. Eleanora has been removed and replaced, moss being again used in the work of reconstruction. Trap-door spiders in warm weather very quickly replace their trap-doors; and if you pass by a wall where several nests have been robbed of their doors only a week before, they will usually be found quite perfect again.

It has been stated[81] that, if the door of a cork nest be fastened down with a pin, a second door will be found next day by the side of the former one. No doubt spiders not unfrequently find their doors blocked up by a fall of earth, and are thus obliged either to make a new opening or to prolong the old tube.

I once fastened open the surface doors of three of the double-door nests by passing a thread through the silk of the door and tying it back to some twigs above. The doors were thus turned backwards, and the aperture of the tubes, which lay in a vertical terrace wall, exposed to view.

Next day, after a night of very heavy rain, I found the doors as I had left them, but in one nest the lip of the tube had been dragged inwards so as partially to close the tube; in the second nothing appeared to have been done, but in the third nest a new covering had been very cleverly extemporized out of three fallen olive-leaves, which were loosely spun together and attached by one or two threads to the margin of the tube. This formed an admirable concealment, but did not move freely as a door, the web being too imperfect. Two days later, however, it was completed and had become a perfect door, moving on a hinge just within and below that of the former door, which still remained as I had fastened it. The other nests remained in the same condition as before, only that a little moss had been dragged into the mouth of the tube of the nest, which had been partially closed with its own lip.

The extreme reluctance which these spiders show to abandon their dwellings is curiously exemplified by what follows.

Certain nests which were furnished with two doors of the cork type were observed by Mr. S. S. Saunders[82] in the Ionian Islands. The door at the surface of these nests was normal in position and structure, but the lower one was placed at the very bottom of the nest and inverted, so that, though apparently intended to open downwards, it was permanently closed by the surrounding earth. The presence of a carefully constructed door in a situation which forbade the possibility of its ever being opened seemed, indeed, something difficult to account for. However, it occurred to Mr. Saunders that, as these nests were found in the cultivated ground round the roots of olive-trees, they may occasionally have got turned topsy-turvy when the soil was broken up. The spider then, finding her door buried below in the ground and the bottom of the tube at the surface, would have either to seek new quarters or to adapt the nest to its altered position, and make an opening and door at the exposed end. In order to try whether one of these spiders would do this Mr. Saunders placed a nest, with its occupant inside, upside down in a flower-pot. After the lapse of ten days a new door was made, exactly as he had conjectured it would be, and the nest presented two doors like those which he had found at first.

There is a specimen of one of these inverted nests, with its two doors, in the British Museum, and this might easily be supposed, at first sight, to be an example of a new kind of double-door nest. On close inspection, however, it will be seen that one of the two doors is discoloured and partly decayed, this being, no doubt, the one which had been buried beneath in the earth and so rendered useless.

Questions have often been asked as to the manner in which trap-door nests are commenced in the first instance, and whether the weaving of the silk lining is begun at the top or the bottom of the tube.

The structure of the cork door also, which often appears so perfectly turned as to resemble the work of a potter's lathe, is another difficulty.

These questions have, as it seems to me, been needlessly complicated by taking it for granted that the perfect nest of the mature spider is made all at one time, that the tube, perhaps of a foot in length, is excavated, lined, and furnished with a door within some short period of time, such as ten days or a fortnight, perhaps.

On the contrary, I believe that the nests are, as a rule, the result of many successive enlargements, and that the nest of the infant, the tube of which is no bigger than a crowquill, is not abandoned, but becomes that of the full-grown spider. This must require time, but how long, whether months or years, we have yet to learn.

Very little is known at present as to the longevity of spiders, but Mr. Blackwall[83] says that some live only one year, while others, such as Tegenaria civilis and Segestria senoculata, have been known to live four.

Whether the trap-door spiders are very long lived or not I cannot positively say, but, from the appearance of the growth of moss and lichen on the doors of some nests which I have observed, I am inclined to think that they must have been inhabited for more than a twelvemonth.

Evidence of the enlargement of the door is not very rare to meet with, though, as a rule, the new piece is woven on to the old with such neatness as more or less to obscure this. In fig. B, Plate X., p. 100, the old and smaller surface-door of a nest of Nemesia meridionalis is seen partially attached to the larger new door, which has been constructed below it; while in fig. C of the same plate, three doors, or rather three enlargements of one door, may be traced. It is this, I believe, that gives rise to the tiled appearance which these trap-doors sometimes present, and which has caused them to be compared to oyster-shells. Something similar may also be occasionally seen in doors of the cork type, as, for example, in that figured at A and A 1 in Plate VIII., p. 94, where the old and smaller door is seen partially raised above the surface of the new one. This I imagine to be merely an example of rather clumsy workmanship, as, if I am right, a full-sized cork door usually incloses within itself several lesser doors, which formerly fitted the tube and have had to be enlarged.

This is borne out by the fact that such a door will, on examination, be found to consist of several layers of silk, with more or less earth between each, these layers decreasing in size from without inwards, and together forming a sort of saucer in which the small central mass of earth lies. Thus by moistening a series of the cork doors of Nemesia cæmentaria, I have been able to detach, in one of medium size, from six to fourteen circular patches of silk, of which the outermost, or that which forms the lower surface of the door, is the largest, and the innermost the smallest, the others being intermediate in size as in position. Perhaps if I had had larger doors at my disposal for examination I might have found more layers, as other authors[84] speak of a much greater number of layers in the cork doors of Cteniza fodiens. Be this as it may, I am confirmed in my opinion that the layers of silk mark the successive enlargements of the nest by the additional fact that in very small doors the layers of silk are few or single, and that a proportion is observable as a rule between the size of the door and the number of layers of which it is composed.[85]

Another proof that enlargement takes place, may at times be found in the nests of N. Eleanora, where one, or even two useless doors may be detected behind the lower door.

Now when there are three lower doors in this way the one which is in use is the largest, and the door lying nearest to this one the next in size, while the hindmost is the smallest of all. But though those abandoned doors are now too small to fit the existing tube, they did so, no doubt, in their day, for they are exact copies in miniature of the ordinary horse-shoe shaped lower doors. The lower door actually in use may sometimes be found to have two separable cases of thick silk enclosing the central mass of earth, and this also, very probably, represents enlargement. In the nests of N. meridionalis I have never found any of these abandoned doors behind the one in use, nor should I expect to find any, for if they were present they would permanently obstruct the entrance from the main tube to the branch.

It is clear that it is better economy on the part of the spider to enlarge its nest rather than build a new one each time. If we compare the infant spider and its nest (fig. B, Plate IX., p. 98) with the full-grown creature and its nest (fig. A, Plate IX.), it becomes evident that the growing spider must either construct many nests of intermediate size, or frequently enlarge the original domicile. And we do in fact find nests of all sizes between the two extremes.

I cannot help thinking that these very small nests, built as they are by minute spiders probably not very long hatched from the egg, must rank among the most marvellous structures of the kind with which we are acquainted. That so young and weak a creature should be able to excavate a tube in the earth many times its own length, and know how to make a perfect miniature of the nest of its parents, seems to be a fact which has scarcely a parallel in nature.

When we remember how difficult a thing it is for even a trained draughtsman to reduce by eye a complicated drawing or model to a greatly diminished scale, we must own that the performance of this feat by a baby spider is so surprising as almost to exceed belief.

And yet even the most complicated form of nest—namely, that of the branched double-door type—is perfectly reproduced in miniature by these tiny architects, with the upper door, lower door, main tube, and branch (fig. B, Plate IX., p. 98).

In order to test whether the doors are enlarged or not I measured the surface doors of seven double-door nests and one minute cork door on April 30th, making a careful plan of the terrace wall in which they lay, in order to make sure of finding them again on my return to Mentone in October.

The following table will show that all were enlarged, the average rate of increase being 1⁷/₁₀ lines in the five and a half months which had elapsed:—

We can scarcely venture from such limited premises to draw any precise conclusions, but if we suppose that during the entire course of the year the nests increase on an average by about four lines in diameter, and assume that the rate of growth continue the same, the nest of the infant spider, whose surface door measures scarcely a line across, would still require four years to attain the dimensions of some of the largest double-door nests, whose surface doors measure sixteen lines across.

It seems to be the rule with spiders generally that the offspring should leave the nest and construct dwellings for themselves when very young.

Mr. Blackwall,[86] speaking of British spiders, says:—"Complicated as the processes are by which these symmetrical nets are produced, nevertheless young spiders, acting under the influence of instinctive impulse, display, even in their first attempts to fabricate them, as consummate skill as the most experienced individuals."

Again, Mr. F. Pollock[87] relates of the young of Epeira aurelia, which he observed in Madeira, that when seven weeks old they made a web the size of a penny, and that these nets have the same beautiful symmetry as those of the full-grown spider. Those of the latter are vertical, circular, made of about 250 feet of thread, having about 35 radial lines and 38 concentric circles, the outermost of which is some 20 inches in diameter. After the lapse of a day or two the web loses its adhesive property and a new one is made. In about six months the female Epeira has completed her ten changes of skin, one of which takes place in the cocoon, and "at the end of eight months the spider is 2700 times as heavy as at its birth." This Epeira lives, we are told, for about eighteen months.

One can scarcely contemplate the work of these architects and weavers, and especially of the trap-door makers, without being carried away into the whirlpool of discussion which has so long raged round the word instinct.

Do the young spiders build their first nest by instinct—that is to say, independently of all teaching or personal experience—or do they copy the nests in which they were hatched?

What is wanting, however, is not discussion, of which we have had enough, but demonstration, and demonstration is hard to come by, depending as it must upon careful and repeated experiment.

If it were practicable, and I have no reason to know that it is not, to rear spiders from the egg away from the nest, and then to cause them to build in places where they should be perfectly at home and yet cut off from all communication with their kind, we might hope to learn whether they can construct the characteristic nests of their species without ever having seen one.

Mr. Wallace[88] shows that there is some reason to doubt whether birds, which are so frequently said to build by instinct, would, under parallel circumstances, construct the nest proper to their kind; and he states that birds brought up from the egg in cages do not do so, nor do they even sing their parents' song without being taught.

Of course we can scarcely compare birds and spiders together, but we should hesitate, in view of Mr. Wallace's expressed opinion as to the nest-building habits of the former, to assume that the latter are independent of teaching and personal experience. It may very possibly be so, but it has never been proved.

I have endeavoured to gather together all the published records of the nests of spiders belonging to the sub-order Territelariæ, with a view, if possible, to trace out the geographical range of the several types of structure. I have, however, met with but a small amount of success, and even among the limited number of tolerably complete accounts of nests which I have been able to discover, several made no mention of the spider to which the nest belongs.

Prof. Ausserer[89] has enumerated 215 species of Territelariæ as having been found in the world at large, but of this large number ten only, as far as I have been able to learn, have been described in connexion with their nests, and eight of these belong to the Mediterranean region.[90] To these we may now add two more, namely, Nemesia meridionalis, with its branched double-door nest, and N. Eleanora the builder of the unbranched double-door nest, thus making twelve in all.

Three of the twelve, however, Atypus piceus, A. Blackwallii, and Nemesia cellicola,[91] do not appear to build true trap-doors, but only a simple silk tube without any covering at the mouth.

The following tabular view will show to which of the four types of trap-door nest those of the remaining nine spiders belong, and their geographical distribution:—

TRAP-DOOR SPIDERS WHICH BUILD

Nests of the cork type.

Nests of the single-door wafer type.

Nests of the double-door branched type.

Nests of the double-door unbranched type.

As far, therefore, as I know at present, the cork type of nest is the only one which is widely spread, and which is constructed by spiders of more than one species. For, while the single-door wafer, and the branched and unbranched double-door nests are each the work of one particular spider, we see that nests of the cork type are made by spiders of six distinct species, belonging to at least three genera.

It is almost certain that a much larger number of spiders of different kinds, though all probably members of the sub-order Territelariæ, construct nests of the cork type, for descriptions and specimens of trap-doors of this kind are brought from the most distant parts of the globe. It is true that these specimens and descriptions usually only show us the surface-door, but as far as our present knowledge goes, we are led to suppose that a door of the cork type is always associated with a simple tube, in which there is no trace of a second door or valve, so that, judging of the unknown by the known, we conclude that nests which possess the characteristic peculiarity of a true cork door are true cork nests in other respects also. Further research may possibly show that there are exceptions to this generalization, but I do not at present know of any.

I have seen Australian specimens of large trap-doors, of the cork type, measuring from one to two inches across. In some of these the doors were scarcely more than semicircular but very thick, and having their edges bevelled so as to correspond with the sloping margin of the tube;[92] in others, found at Paramatta, and described to me by Lady Parker as being tenanted by a black spider, the doors were said to be circular and much smaller, scarcely larger than a sixpence, and of the cork type.

The upper portion of a nest from New Granada has been figured and described by M. Victor Audouin,[93] which closely resembles that drawn at Fig. A in Plate VII., p. 88, but the door is about a third larger.

I have also been assured that nests of the cork type are found in many parts of India, and we have seen above that they are reported to be common in the island of Formosa.

Putting all this together, it will be seen that nests of this type are found all round the globe; in Formosa, India, Syria, the Grecian Archipelago, Italy, and the adjacent islands, Trieste, South France, Spain, Morocco, New Granada, and Australia; while the single-door wafer nest is only known at present in the West India islands;[94] the branched double-door nest at Mentone, Cannes, and Pegli near Genoa, and [doubtfully] near Naples and in Ischia; and the unbranched double-door type at Mentone and Cannes alone. It is quite probable that these three latter forms of nest will some day be found to have a much wider range than that assigned to them here, but I can scarcely think it likely that they will ever be shown to claim the world-wide distribution of the cork type. Supposing that these nests are eventually discovered in many widely distant localities, a very interesting question will arise as to the specific characters of the spiders which inhabit and construct them. Shall we then find, for example, that nests of the unbranched double-door type are not tenanted and fabricated by Nemesia Eleanora alone, as we have hitherto found to be the case, but by many other distinct species also, each in its peculiar district?

That is to say, will the type of nest remain the same while the occupants vary, as in the cork nests?

If, on the other hand, we learn that these three types, the single-door wafer, the branched and unbranched double-door nests, are very local, we shall be led to inquire into the probable causes of this limitation.

But we must study much more closely the habits of these trap-door spiders, and the difficulties and dangers to which they are exposed, if we wish to appreciate fully the true meaning and intention of the structure of their nests, and to find the clue to the difficult question why one type should be more frequently adopted than another. Above all, we must discover what are their enemies, and how and when they are most exposed to them. M. de Walckenaer gives an entertaining account[95] of the enemies to which spiders generally are exposed, and of this the following list is an abstract.

Many kinds of monkeys, squirrels, and several sorts of birds, as well as lizards, tortoises, frogs, and toads prey upon spiders. A species of black sheep, found in the steppes of Asiatic Russia, unearths the tarantulas (Lycosa), and eats them. ("Une espèce de brebis noire, dans les steppes de la Russie asiatique, déterre les tarentules et les mange"). In the East India Archipelago there is an entire genus of birds of the passerine order, which have been named "Arachnoptères" because the different species of which it is composed live exclusively on spiders. Besides these, the centipede (Scolopendra), and the following Hymenopterous insects, Philanthes, Sphex, Pompilus, Pimpla Ovivora, and P. Arachnitor [which last lay their eggs in the eggs of spiders], carry on perpetual hostilities against them.

I have seen it stated that ants are among the worst enemies of spiders, driving their galleries through the silk tubes of the latter and devouring their eggs. Of this I have never seen any trace, and, on the contrary, have on four occasions found the remains of ants' bodies at the bottom of the trap-door spiders' nests.

I have but seldom detected any refuse in these nests, and this accords with what M. Erber tells us[96] of the care with which Cteniza Ariana, which he watched by moonlight in the island of Tinos, carried away the empty bodies of the beetles, the juices of which had been sucked out, to a distance of some feet from its hole. In October, 1872, however, I found a black layer of débris at the bottom of five nests of Nemesia Eleanora, and this was composed principally of the remains of insects, and among others of some rather large beetles.