138 C. H. TURNER. 



straight sections so articulated as to form an arm with three 

 adjustable sections. The straight arms and the sections of the 

 longer arms are all of the same length. The adjustments of the 

 arms make it possible to set each and every one at any desired 

 angle to the pull of gravitation. The lower pointed end of the 

 post was erected in loose soil in such a manner that, by rotating 

 the post on its axis, it was possible to vary the position of any arm 

 with reference to either the direction of the rays of the light or 

 the path of the wind. 



The copper horizontal maze (Fig. 3), which was supported on 

 slender glass pillars, had one of its central alleys connected with 

 the supporting table by means of a cardboard incline two inches 

 wide and thirty inches long. It was so located that the strongest 

 light came from one side. 



The point of view expressed in this communication was ob- 

 tained by a critical study of 1,000 caterpillars of the apple-tent 

 moth, Malocosoma americana Fab., 300 of the cabbage butterfly, 

 Pier-is rapce. Linn., fifty of the white tussock moth, Memerocampa 

 leucostigma A. & S., and a few individuals of the young of each 

 of a large number of species of butterflies and moths collected 

 from the surfaces of a large variety of food plants. Throughout 

 a summer practically every kind of surface-feeding caterpillar 

 that a diligent search of the vicinity of St. Louis yielded w r as 

 put to the test. Among those tested were the flat slug-like 

 Lyncenidae, the elongated Papilionidae and Pyralidse. and the 

 fusiform Hesperiidae; the naked Noctuidse, the hairy Arctiidae, 

 and Liparidse, and the bristly and spiny Ceratocampidae and 

 Nymphalidae; the Notodontidae with their erect anal somites 

 and the Sphingidse with their oblique stripes and caudal horns; 

 the grotesque Saturniidae and the looping Geometridae. 



EXPERIMENTS WITH THE VERTICAL MAZE. 



To facilitate description, the parts of the vertical maze were 

 named as follows: The upright was called the post; each straight 

 arm and each section of a jointed arm was given a distinctive 

 letter. Thus (Fig. i) the first arm above the ground was called 

 abc, the second, def, the third j, the fourth g and the fifth h. 



In the first few hundred experiments with the tent-caterpillars 



