262 THK entomologist's becord. 



the side of another caterpillar, and was able to see how the circular 

 form of the hole in the larva's skin was brought aboiit. The grub 

 makes only a ver}' small opening, through which it thrusts its head, 

 and then works its body backwards and forwards, gradually enlarging 

 the hole by distension, the skin at first moving in and out eversibly and 

 retroversibly, until some two-thirds of the grub's body is visible. 

 Those which make their cocoons from this position must emerge before 

 the cocoon is finished, as one never finds any part of the cocoon em- 

 bedded in the body of the caterpillar. 



Notes on the Distribution and Variation of Agriades coridon 

 in tlie Cambridge District. 



By G. L. KEYNES. 



In England, Aiiriade'^ coridon appears to be almost entirely con- 

 fined to chalky districts, and, in the neighbourhood of Cambridge, this 

 is certainly the case, but, even on the chalk, it is not generally 

 distributed, since it appears only at certain isolated points along the 

 range of chalky hills, which stretch across the southern part of the 

 county, from Newmarket, in the east, to Baldock, beyond the south- 

 west boundary. This localisation is probably only secondary, since 

 the species seems to occur in most of the locahties suited to it, that is, 

 in the places, where, for various reasons, the land has escaped culti- 

 vation.. That there is no communication between the various colonies 

 of this species seems to be indicated by facts to be mentioned later in 

 connexion with its variation. 



At the eastern end of the chalky range, the species occurs in some 

 numbers on Newmarket Heath, though here it is to be found chiefly 

 along the great prehistoric earthwork, known as the Devil's Dyke. 

 The deep ditch on the western aspect of this earthwork afi'ords the 

 most complete protection for a species such as A. coridon, whereas 

 the greater part of the heath itself is in constant use for horse-racing 

 and exercising. Passing from here south-west along the Icknield 

 Way, one finds the next stronghold of A. coridon to be another great 

 earthwork. Fleam Dyke, which runs parallel to, and, in most ways, 

 resembles the Devil's Dyke. The ground on either side of this dyke is 

 under cultivation, but in the ditch and on the sides of the rampart A. 

 coridon is to be found in great numbers. About a mile further on, A. 

 coridon is to be found again along the course of the now disused 

 Koman road, the Via Devana. The next point at which the species 

 appears is some way further to the south-west, namely, on Koyston 

 Heath ; this heath is of great extent, and may be considered to be- the 

 most important locality in the district for A. coridon, which occurs 

 abundantly all over it. Beyond this region the country is all culti- 

 vated, and my investigations have extended no further. 



Taking into account the type of country, the distribution is onlyj 

 what might be expected, and it is, in itself, of no particular interest ;i 

 when, however, it is correlated with the variation of the species it at-] 

 once becomes interesting and affords a striking instance of the, 

 peculiarities of local races. 



The (? is of fairly constant form throughout the district, and! 

 presents only the usual amount of variation in size, in the breadth and! 

 intensity of the dark border upperside forewing, and in the shade (from! 



