398 Dr. T. A. Chapman on the 



I had not previously visited, which materially assisted me 

 in obtaining eggs. 



The eggs are laid in the little rosettes of leaves of the food- 

 plant, right down between the leaves, so as rarely to be 

 visible without separating them, and often as near the 

 centre as affords leaves large enough to have openings 

 between them, but also further out and even on quite 

 outside leaves. 



On my last day at Gavarnie (August 1st) I obtained some 

 more butterflies in order to secure a further supply of eggs, 

 but on the 2nd I left Gavarnie and was a week on the way 

 home. By good luck it happened that none of the eggs I 

 had obtained hatched until just as I got home on the after- 

 noon of August 9th, or so immediately before that they 

 managed to find food in the plants on which the eggs had 

 been laid. 



The egg of pyrenaica is 0*625 mm. in diameter, the mar- 

 gins well rounded from the upper to the under surface, 

 the top itself being somewhat dome-shaped, not flat as in 

 many Agriades. The micropylar area is about 0'125 mm. 

 across, and has about seven cells that would be crossed by 

 a line from centre to circumference, but they are hardly 

 regular enough to sa}^ there are seven circles. The cells 

 of the network are about 0'03 mm. in diameter, the lines 

 of netting have hardly any eminences at their junctions on 

 top, but these are present though small on the sides. 



Comparing this egg with that of orhitulus (see Trans. Ent. 

 Soc, 1911, PI. XII), we find that the size and form are in 

 very close agreement. But in other respects the differences 

 are considerable. In orbihdus, the micropylar area is much 

 smaller, only about half the diameter, 0"064 mm. instead of 

 0*125 mm., and the number of cells make only four or five 

 rings, if we count them in the circles they don't quite 

 arrange themselves in. Even though it be inaccurate to talk 

 of circles, there are nevertheless seven cells from centre to 

 margin counted in the same way as only give four or some- 

 times five in orhitulus. Similarly, there are about seven 

 cells from the micropylar area to the margin in 'pyrenaica, 

 where similar counting gives quite nine in orbitidvs. The 

 cells are smaller in orbit ukis, look much more numerous, and 

 have much more fully-developed prominences or pillars 

 where the lines of network meet. 



The following notes as made refer to the rearing of the 

 larvae, etc. : — 



ll 



