J'lF.Rin.-E. 27 



having two or three dark grey cloudy spots and streaks along 

 the costal margin. Under side of fore wings with markings 

 as above, with the addition of two sub-marginal blackish spots, 

 and having the apical patch dusky green ; hind wings dusky 

 green, spotted and streaked with white. Female similar, but 

 the dark markings more prominent, an additional patch near 

 the dorsal margin of the fore wings, and the hind wings 

 mottled with blackish. 



Larva, greyish-blue, covered with small black granu- 

 lations, with four longitudinal white stripes, and with a 

 yellow spot on each segment. Legs and ventral surface 

 wiiite. Feeds on Crucifenv and Rcscdacccv. 



Pupa grey, speckled with Ijlack, and with reddish stripes 

 (Lang) ; but other authors say that the stripes in the pupa 

 are yellow. 



The butterfly appears in May and August on the Continent, 

 the larva in June and September. 



This is one of our rarest butterflies, but occurs now and then 

 on the South Coast. A specimen of the May brood has been 

 recorded in Devon, but this is a ciicumstance of extreme 

 rarity, the captures made being usually in August. Its 

 flight is said to be slow, from flower to flower, but this doubt- 

 less arises from languor caused by a climate too chilly for its 

 constitution. In hotter regions its flight is sufficiently bold 

 and swift. A large propoitiou of the British specimens have 

 been captured in East Kent, but isolated specimens have 

 been taken in most of the southern coast counties and also in 

 Essex. The Devil's Dyke near Xewmarket, and Gamliugay, 

 Cambs, have been recorded as favourite localities for it from an 

 early period, and its repeated occurrence in a locality so far 

 inland as Newmarket seems to indicate the probability of its 

 being native, but no proof of the capture of a larva or pupa in 

 this country is obtainable, and the evidence so far is rather 

 in favour of the view that the specimens taken are casual 

 wanderers from abroad. There seems to be no doubt that 



