176 Dr. T. A. Chapman on 



These " eminences " I take to be what Mr. Morice calls 

 *' the apical plates of the saw sheath," they remind me 

 in position and structure of the 10th abdominal segment 

 of many female Lepidoptera, a pair of rounded plates, 

 well clothed with hairs, not densely, but spread and especi- 

 ally marginal, and impressed one that they were for pre- 

 cisely the same office as the very important one in the 

 Lepidopterous structure, viz. to verify that the situation 

 selected, and in fact to select it, is exactly what is required, 

 in the one case for laying the egg, in the case of the sawfly, 

 for cutting the pocket. 



The insect is resting on the leaf with legs on both sides 

 of it, and with this fulcrum is able to press the abdomen 

 sufficiently against the leaf. The two apical plates do not 

 grip the leaf, but still, with its margin actually or apparently 

 between them, give the selected place the required steadi- 

 ness. The one facing the observer covers the actual point 

 of entry of the saws, but the movements noted above are 

 really the beginning of their operation and their actual 

 entry into the leaf. They do so " on the upper surface, 

 but so close to the margin that I am not quite positive 

 that the actual margin is not the real or intended place. 

 When the operation is finished it is practically impossible 

 to verify any actual opening, but one guessed it to be some 

 0*1 mm. from the margin on the upper surface. The body 

 of the insect is so placed that the plane of the leaf coincides 

 with the median antero-posterior plane of the insect." 

 Fuller observations on more specimens of the work, shows 

 that the incision is on the upper surface, parallel with the 

 margin and nearer 0'3 mm. from it than the distance 

 noted above. " When once the saw enters it works very 

 rapidly and the full extent of cutting is done in from 

 about a minute to a minute and a half. This is guessed, 

 as I was too intent on observing the process to note the 

 watch at the same time ; but in one instance, I found that 

 there was a rest of 10 seconds at the end of the cutting, 

 and that the laying of the eggs and extraction of the saw 

 took place in 15 seconds. In observing the action of the 

 saw, two circumstances made this much more satisfactory 

 than I had anticipated. One was the absolute trans- 

 parency of the upper cuticle of the leaf, beneath which 

 the saw penetrated, the parenchyma of the leaf being below 

 the saw ; the other was that during part of the operation, 

 a portion of the saw between the leaf and the body of the 



