64 KECOED OF CUBKENT KESEAKCHES KELATING TO 



transformations, and the operation caused, apparently, little inconve- 

 nience, for they recommenced feeding almost immediately afterwards. 

 The effect on the moths produced from these larvae was as follows : — 

 One was deprived of three tarsal joints, but the claw was developed. 

 Three were deprived of three tarsal joints, and of the claw also. Three 

 had only the femur and tibia. One had the leg " amputated " in the 

 middle of the femur. The two others had only a stump scarcely a 

 millimetre in length. M. Melise adds that in not one of the moths 

 was the leg absolutely absent, and that the variation in the amount of 

 deformity probably resulted from the difficulty of performing the 

 amputation in the larvee at precisely the same place in each. In the 

 case of insects with incomplete metamorphoses parallel experiments 

 have often been made, and with similar results ; but, Mr. M'Lachlan 

 points out, with Lepidoptera they have been so few as to render 

 confirmatory evidence of the statements of other experimenters of 

 much value. 



Morphology and Ancestry of Insects.* — Mr. Wood-Mason makes 

 use of the following observations on certain of the Orthoptera in 

 support of the theory that insects have been derived from crustacean 

 ancestry : — 



The position, he considers, of the anfennce of the Thysanuran 

 Machilis, relatively to the eyes and mandibles, point to a homology 

 with the antennae proper of Crustacea ; but further, in Machilis 

 maritima the two-jointed peduncle of the antenna shows a small blunt 

 papilla on its inner side near the distal end ; an Indian species of the 

 allied form, Lepisma, has a minute but movable apjiendage in the same 

 position on the peduncle. Now, in the Myriapod Pauropus, Sir John 

 Lubbock has shown the antenna to be biramose, the two rami spring- 

 ing from the peduncle, which is two-jointed in an early stage ; this 

 condition is practically that of the crustacean antenna, with its 

 protopodite, endopodite and exopodite, with which the antenna of 

 Macliilis is thus connected through the medium of Lepisma and 

 Pauropus. The ripe embryos of Blatta (PanestJiia) javanica show a 

 similarly placed tubercle on their two-jointed antennary peduncle. 



Again, the mandible in Machilis has three joints, the second 

 representing the second and third of Myriapoda ; while in the embryo 

 of the above-mentioned Blatta, four are indicated by three folds by 

 which the mandible is constricted ; only one existing in the adult, 

 whose ancestor nevertheless appears to have had the four joints typical 

 in Myriapoda. The mandible in Machilis is bifid in front, being 

 divided into a molar and cutting segment by a deep fork containing 

 an integumental fold : in Lepisma the fork is a notch and in the 

 cockroaches it is still a notch ; the two divisions probably represent 

 an endopodite and exopodite, which seems confirmed — among other 

 proofs — by the presence in some Staphylinid beetles of a movable 

 appendage articulated to the jaw in the same position as the usual 

 molar process, which is absent in them, and probably represents the 

 endopodite of the crustacean mandible. 



On each of the four posterior legs of Machilis is to be found, as 

 * ' Trans. Eutomol. Soc, Lond.,' 1879, p. 145. 



