68 BRITISH LEPIDOPTERA. 



blood. The ultimate destination of these cells is unknown. Their 

 presence reveals a special activity and they are found in all places 

 where great changes are taking place. They are very numerous in 

 the legs at the commencement of the 4th stadium, disseminating them- 

 selves some days later in all the body-cavity. During the histolytic 

 period they attack the larval tissues and increase in size at their 

 expense, on the other hand, they serve for the nutrition of the imaginal 

 structures, on which they have no destructive action. Van Eees 

 agrees with Kowalewski in comparing the attacks of the embryonic 

 cells, now victorious, then powerless, to the struggle that the leucocytes 

 carry on against attenuated and virulent bacteria. 



Capillary tracheoles appear in the leg at the same time as in the 

 wing. They originate at the termination of a tracheal trunk near the 

 base of the limb on the dorsal and convex side. This point of origin 

 is analogous with the umbilicus of the wing. After the third moult, 

 the hypodermis thickens in its neighbourhood. It forms a pad or 

 cushion in a few days, then a large bud with circular invagination. 

 This bud elongates before and behind, pushing out the thin envelope ; 

 it soon extends beyond the limit of the leg but remains lodged in a 

 depression of the lower surface of the thorax. It then makes a pro- 

 jection, like the wing, in the body-cavity, and, on opening the larva, 

 it is seen to emerge from the interior of each of the true legs. Lyonet 

 has well described these six little masses as having " a very bright 

 iridescent white " appearance, and he surmised that they might be 

 " the germs of the imaginal legs." The nerves and a tracheal branch, 

 before distributing themselves over the rest of the limb, penetrate into 

 the bud and form there a little loop that marks the point of junction of 

 the femur with the tibia. This (which Gonin calls the femoro-tibial) bud 

 is none other than a combination of these two parts, intercalated, so 

 to say, between the larval leg and its base (or root). There is not yet 

 a separating membrane, and the body-cavity still remains in direct 

 communication with the extremity of the limb. This is indispensable 

 to the normal functioning of the muscles, which have not yet com- 

 pleted their duty in the larva. The tracheoles also follow this shortest 

 route without passing through the bud. 



In the caterpillar at the period of pupation the extremity only of 

 the imaginal leg germs are drawn from the larval legs. The other 

 parts are applied closely to either side of the thorax. Near the ventral 

 line is a little swelling representing the hip and the trochanter ; the 

 femur and the tibia are clearly recognisable, but united to each other, 

 and separated only by a slight furrow. They form, at their union, a 

 very sharp knee. The femur is movable on the swelling or pad of the 

 hip ; the tibia is continued without precise limits, with the extremity 

 bidden in the larval leg. The three divisions of the latter do not 

 appear to have any connection with the five joints of the imaginal leg. 

 Under the microscope, the rudiment or disc appears very strongly 

 folded at the level of the tarsus, much less in the other regions. A 

 large trachea penetrates into the femur with some capillaries ; arrived 

 at the knee, it is bent inwards to the tibia by a sharp curve, but only 

 becomes really sinuous as it approaches the extremity. It is then the 

 tarsus that is particularly susceptible of elongation, and it is likely in 

 being withdrawn to give the impression that the entire organ frees 

 itself from the larval leg. If, therefore, one cuts the larval leg at its 



