TRANSMISSION OF MODIFICATIONS 391 



Of many attempts to watch and record the process from 

 first to last, only three were successful in covering the entire 

 period. The rest were fragmentary, covering only portions of 

 the process. This was owing to the difficulty of keeping the 

 insect under focus for the fifteen to twenty-five minutes re- 

 quired for the complete operation without disturbing its work. 

 The record of the observer is as follows : 



In the first observation the female moved about the apple for several 

 seconds, keeping the end of her beak in contact with the surface, as if 

 seeking a favorable spot. When the exact spot was decided upon", the 

 minute jaws at the end of the snout began a rapid movement which quickly 

 made an opening through the skin. This opening was no larger than neces- 

 sary for admission of the tip of the beak. No skin was removed ; it was 

 simply torn and thrust aside to give access to the pulp below. Later, as 

 the excavation proceeded, the broken skin was seen as a sort of fringe 

 around the beak at the surface of the fruit. As soon as excavation in the 

 pulp was commenced, the beak was deflected backward so that the work 

 was carried on under the insect, just beneath the skin and nearly parallel 

 with the surface. As the work advanced, the opening through the skin 

 became slightly enlarged by lateral motions of the beak. The pulp was 

 all eaten as excavated. During the process the beak was not once with- 

 drawn, nor was there any cessation of motion. When the excavation of 

 the cavity was completed the beak was withdrawn by a quick motion, the 

 insect turned about, adjusted the tip of the abdomen to the opening and 

 deposited an egg, which was forced to the extremity of the excavation by 

 the ovipositor. The insect now rested without motion for two minutes ; 

 then, turning again, proceeded to cut the crescent in front of the egg. 

 This crescent puncture was not wholly a separate puncture, but, starting in 

 the original opening through the skin, was cut laterally in either direction, 

 partly by the jaws and partly by crowding the beak first one way and then 

 the other. The direction of the beak was but little deflected from the per- 

 pendicular, and the cut was made as deep as the length of the beak would 

 allow. The pulp torn away in making the crescent was eaten, just as was 

 done in excavating the egg cavity. The crescent completed, the insect 

 walked away, drew the legs closely under the body, and settled down, 

 apparently to sleep. The time occupied in the process described was dis- 

 tributed -as follows : 



Excavating egg cavity . . .' . . . . 9 minutes 



Deposition of egg . . . . . . . . i minute 



Rest 2 minutes 



Cutting the crescent 3} minutes 



Total 15'- minutes 



