Cerceris deserta was one of the wasps that objected strongly to our presence, and she also made a great deal of fuss about leaving her nest. Nearly all the species circle before leaving a spot to which they intend to return, but deserta begins her flight with a series of short zigzags in the form of a half circle on one side of the nest. C. nigrescens, too, begins with semicircles, while C. clypeata flies entirely around and around the opening. The contrast between the deliberate movements of Astata and the rapid flight of Cerceris is very striking. We have now given a sufficient number of instances, from widely separated genera, to show the care that is taken by wasps to acquaint themselves with the surroundings of their nests. It has also been shown that in spite of all this care they frequently have trouble in finding their way about. All these facts have led us to conclude that wasps are guided in their movements by their memory of localities. They go from place to place quite readily because they are familiar with the details of the landscape in the district they inhabit. Fair eyesight and a moderately good memory on their part are all that need be assumed in this simple explanation of the problem. Chapter XIII INSTINCT AND INTELLIGENCE OUR study of the activities of wasps has satisfied us that it is impracticable to classify them in any simple way. The old notion that the acts of bees, wasps, and ants were all varying forms of instinct is no longer tenable, and must give way to a more philosophical view. It would appear to be quite certain that there are not only instinctive acts but acts of intelligence as well, and a third variety also—acts that are probably due to imitation, although whether much or little intelligence accompanies this imitation is admittedly difficult to determine. Again, acts that are instinctive in one species may be intelligent in another, and we may even assert that there is a considerable variation in the amount of intelligence displayed by different individuals of the same species. We have met with such difficulty in our attempts to arrange the activities of wasps in different groups that we are forced to the conclusion that any scheme of classification is merely a convenience, useful for purposes of study or generalization, but not to be taken for an absolutely true expression of all the facts. This kind of perplexity is well understood and allowed for in all morphological work, but it has never been fully realized in the study of habits. The explanation is not far to seek. The habits of but few animals have been studied in sufficient detail to bring out the evidence that there is as much variation on the psychological as on the morphological side, although this field seems fresh and inviting when compared with the researches of the laboratory.