502 Special Protection of Appendages among Insects. 



to replace a limb detached by self-mutilation sometimes shows 

 itself, but in an indistinct manner, coiled up under the pro- 

 tective cuticle, which has some degree of transparency in 

 certain species. 



So far as the Mantidge and Blattidae are concerned, I 

 noticed that the amount of withdrawal of the muscles divided 

 by self- mutilation within the chitinous covering was propor- 

 tional to the violence of the efforts made by the insects in 

 getting rid of the limb. When self-mutilation took place 

 easily the contraction was practically nil. 



It now remains for us to examine the particular case pre- 

 sented by the Phasmidje. 



When artificial separation takes place in the region com- 

 prising the femur and the upper two thirds of the tibia, con- 

 traction of the severed muscles is most marked. When, again, 

 similar cuts are made in the upper part of the region formed 

 by the lower third of the tibia, contraction is still produced, 

 and as it is exactly there that the power of regeneration begins 

 to show itself, the part in process of growth remains hidden 

 until the following moult. Then in proportion as the cuts 

 approach the tarsus, contraction becomes less and less evident, 

 until in the neighbourhood of the tibio-tarsal articulation the 

 divided muscles practically remain in position ; so that the 

 part in process of regeneration may be visible before a moult 

 takes place. The result is the same when the cuts are made 

 on any of the first three joints of the tarsus. 



On examining the internal structure of the limb we find 

 that it is precisely in the spot under discussion — upon the 

 lower portion of the tibia and the first joints of the tarsus — 

 that the muscles are inserted whicli move the tarsus as a 

 whole and its various parts. Cuts made in this region pass 

 through the surfaces of the chitinous covering to which these 

 muscles are attached and from which they run towards the 

 joints below which they have to move. Under tliese circum- 

 stances one can understand why the contraction will be slight 

 or not produced at all. It is only possible to produce it when 

 the cut is more or less remote from the upper attachment- 

 surface of the divided muscles, which is not the case. In 

 other insects there is complete withdrawal of the cut muscles 

 when the section passes through the tarsal region. It is 

 evident that these particular features point to differences in 

 the number and position of the attachment-surfaces of the 

 muscles in question, differences which are revealed upon 

 careful dissection. 



It follows, then, that among the Phasmidge parts in process 

 of regeneration in the region we are considering are more or 



