To sum up, when working in the natural state, the Labyrinth Spider builds around the eggs, between two sheets of satin, a wall composed of a great deal of sand and a little silk.  To stop the Ichneumon’s probe and the teeth of the other ravagers, the best thing that occurred to her was this hoarding which combines the hardness of flint with the softness of muslin.

This means of defence seems to be pretty frequent among Spiders.  Our own big House Spider, Tegenaria domestica, encloses her eggs in a globule strengthened with a rind of silk and of crumbly wreckage from the mortar of the walls.  Other species, living in the open under stones, work in the same way.  They wrap their eggs in a mineral shell held together with silk.  The same fears have inspired the same protective methods.

Then how comes it that, of the five mothers reared in my cages, not one has had recourse to the clay rampart?  After all, sand abounded:  the pans in which the wire-gauze covers stood were full of it.  On the other hand, under normal conditions, I have often come across nests without any mineral casing.  These incomplete nests were placed at some height from the ground, in the thick of the brushwood; the others, on the contrary, those supplied with a coating of sand, lay on the ground.

The method of the work explains these differences.  The concrete of our buildings is obtained by the simultaneous manipulation of gravel and mortar.  In the same way, the Spider mixes the cement of the silk with the grains of sand; the spinnerets never cease working, while the legs fling under the adhesive spray the solid materials collected in the immediate neighbourhood.  The operation would be impossible if, after cementing each grain of sand, it were necessary to stop the work of the spinnerets and go to a distance to fetch further stony elements.  Those materials have to be right under her legs; otherwise the Spider does without and continues her work just the same.

In my cages, the sand is too far off.  To obtain it, the Spider would have to leave the top of the dome, where the nest is being built on its trellis-work support; she would have to come down some nine inches.  The worker refuses to take this trouble, which, if repeated in the case of each grain, would make the action of the spinnerets too irksome.  She also refuses to do so when, for reasons which I have not fathomed, the site chosen is some way up in the tuft of rosemary.  But, when the nest touches the ground, the clay rampart is never missing.

Are we to see in this fact proof of an instinct capable of modification, either making for decadence and gradually neglecting what was the ancestors’ safeguard, or making for progress and advancing, hesitatingly, towards perfection in the mason’s art?  No inference is permissible in either direction.  The Labyrinth Spider has simply taught us that instinct possesses resources which are employed or left latent