Every one has heard of a 'bee-line.' It would be no less correct to speak of a wasp-line. On August 6 I marked a wasp, the nest of which was round the corner of the house, so that her direct way home was not out at the window by which she had entered, but in the opposite direction, across the room to a window which was closed. I watched her for some hours, during which time she constantly went to the wrong window, and lost much time in buzzing about at it. For ten consecutive days this wasp paid numerous visits, coming in at the open window, and always trying, though always unsuccessfully, to return to her nest in the 'wasp-line' of the closed window--buzzing about that window for hours at a time, though eventually on finding it closed she returned and went round through the open window by which she always entered. This observation shows how strong must be the instinct in a wasp to take the shortest way home, and how much the insect depends upon its sense of direction in so doing. It also shows how long a time it requires to learn by individual experience the properties of a previously unknown substance such as glass. But to this latter point we shall presently have occasion to return. Next we must adduce evidence to show that in way-finding the 'sense of direction' in bees appears to be largely supplemented by observation of particular objects. Sir John Lubbock observes: 'I never found bees to return if brought any considerable distance at once. By taking them, however, some twenty yards each time they came to the honey, I at length _trained_ them to come to my room;' that is to say, bees require to _learn_ their way little by little before they can return to a store of honey which they may have been fortunate enough to find; their general sense of direction is not in itself a sufficient guide. This, at least, is the case where, as in the experiments in question, the bees are _carried_ from the hive to the store of honey (here a distance of less than 200 yards): possibly if they had found the honey by themselves flying towards it, and so probably taking note of objects by the way, one journey might have proved sufficient to teach them the way. But, whether or not this would have been the case, the fact that when carried they required also to be taught the way piece by piece, is conclusive proof that their sense of direction _alone_ is not sufficient to enable them to traverse a route of 200 yards a second time. The same result is brought out by other experiments conducted on a different plan, though not apparently with this object. 'My room is square, with two windows on the south-west side, where the hive was placed, and one on the south-east.' Besides the ordinary entrance from outside, the hive had a small postern door opening into the room.