294 British Dragonflies. 



lavender tints in reaching maturity as does A. puclla or 

 E. cycithigcntni. 



Variation. 



There is considerable difference in size, especially 

 among the males, two that I took in the New Forest in 

 August, 1898, being particularly small. In addition to 

 this the shape of the various spots, and the amount of 

 the segment which an}' particular one covers, is very 

 inconstant. The distinctive spot on the second segment 

 is sometimes lightly, and at others heavily, delineated ; 

 its m.ethod of attachnicnt to the circlet posterior to it, 

 or the absence of the attachment, might cause the insect 

 to be confused with A. pnlchellum on the one hand, or 

 A. pnella on the other, if the examination of the insect 

 were too cursory. 



Oviposition. 



Attention has been called b}' Mr. McLachlan to an 

 interesting fact in connection with this insect. In Savoy, 

 in 1884, the season being very dry, he saw the females of 

 certain Agrions, chiefly A. uicrcuriale, ovipositing in wet 

 mud, which proceeding he was led to discover by 

 noticing that certain of them " when fl>'ing were 

 conspicuous on account of the whitish colour of the 

 whole, or a portion, of the abdomen. On examination 

 these proved to be always females, and the whitish colour 

 due to an incrustation of mud ; in some it was only 

 at the tip of the abdomen, in others for its whole length 

 (nearly an inch). The explanation was obvious. These 

 females had been engaged in oviposition, and some 

 instinct had prompted them to sink their eggs as deeply 

 as possible in the mud, so as to afford some chance of 



