The topochemical antennal sense also furnishes splendid proofs of memory in ants, bees, etc. An ant may perform an arduous journey of thirty meters from her ruined nest, there find a place suitable for building another nest, return, orienting herself by means of her antenna, seize a companion who forthwith rolls herself about her abductrix, and is carried to the newly selected spot. The latter then also finds her way to the original nest, and both each carry back another companion, etc. The memory of the suitable nature of the locality for establishing a new nest must exist in the brain of the first ant or she would not return, laden with a companion, to this very spot. The slave-making ants (_Polyergus_) undertake predatory expeditions, led by a few workers, who for days and weeks previously have been searching the neighborhood for nests of _Formica fusca_. The ants often lose their way, remain standing and hunt about for a long time till one or the other finds the topochemical trail and indicates to the others the direction to be followed by rapidly pushing ahead. Then the pupæ of the _Formica fusca_ nest, which they have found, are brought up from the depths of the galleries, appropriated and dragged home, often a distance of forty meters or more. If the plundered nest still contains pupæ, the robbers return on the same or following days and carry off the remainder, but if there are no pupæ left they do not return. How do the Polyergus know whether there are pupæ remaining? It can be demonstrated that smell could not attract them from such a distance, and this is even less possible for sight or any other sense. Memory alone, i. e., the recollection that many pupæ still remain behind in the plundered nest can induce them to return. I have carefully followed a great number of these predatory expeditions. While Formica species follow their topochemical trail with great difficulty over new roads, they nevertheless know the immediate surroundings of their nest so well that even shovelling away the earth can scarcely disconcert them, and they find their way at once, as Wasmann emphatically states and as I myself have often observed. That this cannot be due to smelling at long range can be demonstrated in another manner, for the olfactory powers of the genus Formica, like those of honey-bees, are not sufficiently acute for this purpose, as has been shown in innumerable experiments by all connoisseurs of these animals. Certain ants can recognise friends even after the expiration of months. In ants and bees there are very complex combinations and mixtures of odors, which Von Buttel has very aptly distinguished as nest-odor, colony- (family-) odor, and individual odor. In ants we have in addition a species-odor, while the queen-odor does not play the same rôle as among bees.