TWO INSECTS OF THE ORCHARD SNODGRASS. 503 



arc near one end of the carpet, working from left to right (fig. 13, 

 1-6). Then she turned about and located the seventh (/) at the 

 opposite end, half again her own length from the arc of the first 

 six. This she followed with six others placed in a regular curve 

 to the right, ending with the thirteenth {13). Again changing po- 

 sition she erected the fourteenth (14) to the left of the seventh and 

 proceeded to the left till the nineteenth (19) was in place. Now 

 she went back and inspected the first six. Evidently deciding that 

 these were too far away, she erected palisade 20 nearer the carpet 

 at this same end, and then filled the gap between 20 and 19 with six 

 pickets. The last three of these were a little crowded and 26 was 

 out of line. Finally, 27 and 28 were placed with good spacing be- 

 tween 20 and IS, when the stockade was completed. Eighteen min- 

 utes had elapsed since the start, during which time the caterpillar 

 set up 28 palisades, 



and 6 of these were • . i3 jg 



superfluous. # 



A completed pal- * z? • 



isade looks like a .20 \ 



tapering thread of 

 silk standing up- 

 right on a spreading 

 base, and, before 

 seeing one made, it 

 is rather puzzling 9 « 



to imagine how the • • • 2S 



Caterpillar Can Spin FlG 13> _ Diagram of plan followed by a Bucculatrix cater- 

 S U C h a filament. pillar in constructing its stockade. The numbers indicate 



The palisade, how- the order in which the P alisades were set u p- 

 ever, is not made of a single thread nor of a strand of threads, but 

 of many short loops drawn up one over the other till finally a long 

 apical loop is whipped out, when the caterpillar works rapidly 

 down again, but only to bind the fibers together here and there. 

 The whole is completed in 30 seconds. While working the cater- 

 pillar rears up on her abdominal legs (pi. 3, (J) but uses only 

 the first pair of thoracic legs to support herself against the pali- 

 sade. When the last picket was in place the caterpillar under 

 observation made no further inspection of her fence, appearing 

 to know by instinct that the job was done, or should be. In fact, 

 from now on she ignored the stockade. 



Having gone through with the preliminary formality of fencing 

 herself in, the caterpillar stretched herself out on her carpet and 

 indulged in a short rest. After seven minutes repose she became 

 active again, turned to the end of the carpet first made, and there 



