100 BRITISH BUTTERFLIES. 



white line and the eyes. I find no mention of this red colour in the 

 details of any description, it appears that there may be individual 

 variations in the colour of the larvae. On the underside, the body, with 

 16 feet, was also bluish-green. On the penultimate segment, one sees 

 two square, clear, white marks, also on the third ring from the end is 

 a white band in the form of two united halves. The last pair of feet 

 were larger and broader than any of the abdominal pairs (see fig. 3). 

 I had this larva figured on June 15th. It rested during the day with 

 the head retracted. I did not notice that it ate much at night, and, on 

 the 18th following, it had already pupated. [Snellen van Vollenhoven, 

 Sepp's Nederland. Insecten, 2nd series, iii., p. 86. Described from larva 

 found June, 1871, by Mr. H. W. de Graaf, feeding on a soft kind of 

 grass with free, broad leaves. Figured at fig. 1.] 



Foodplants. — Coarse grasses — Arrhenatherum dating, etc. (Frey), 

 Triticitm, etc. (Hawes). 



Puparium. — In nature the puparium appears to be made by the 

 spinning up of blades of coarse grass. Such an one was found near 

 Harwich, in July, 1891 (Mathew). In confinement, the larva, when full- 

 grown, spins the grass-stems together low down by means of a network of 

 white silk, changes to pupa therein, and in this state remains for about a 

 fortnight to three weeks, according to temperature (Hawes). Snellen 

 van Vollenhoven notes (Sepp's Ned. Insecten, 2nd series, iii., p. 86) that 

 a larva spun up in confinement on June 16th, 1871, fastening itself 

 firmly with a slight spinning to the paper with which the top of the 

 bottle, in which it was confined, had been closed. The fixture of the 

 pupa consisted only in some cross-threads over the anterior extremity, 

 and some confused spinning in which the anal extremity was barely 

 fixed. This species, he further observes, does not, therefore, appear, like 

 the larva of A. sylvanus, to roll up a green leaf into a sort of cocoon in 

 which to pupate. One suspects from the evidence of Mathew, etc., that 

 this is not accurate and that it does do so. The pupal stage lasted from 

 June 18th to July 8th. Snellen notes the pupal period as lasting 13 

 days, from July 15th to the 28th. 



Pupa. — The pupa was very remarkable, as might be expected of 

 a Hesperiid. [It is figured, front and back, in figs. 4-5, and an 

 enlarged figure of the head is given in fig. 6.] On the ventral side, the 

 pupa was yellow-green, on the dorsum darker green with four yellow 

 longitudinal lines not extending the whole length, the wing-covers and 

 the tail were yellow. On the green head stood a pin-shaped, slightly 

 forward bent little horn, which, on its red extremity was beset with 

 little black-red hooked bristles. The proboscis-sheath, which is basally, 

 naturally fixed in the covering of the pectus and feet, was beyond the 

 end of the wing coverings free, but unbent and stuck out beyond the 

 half of the abdomen. The three last segments retreated to the ex- 

 tremity in a solid cone. Hound the stigmata on the abdominal segments 

 are some brown spots. At the tail end no hooks were to be found. 

 Circumstances are here reversed, for, in the pupsB of most moths, it is by 

 the tail end that the pupa is iixed to the spinning. The booklets or 

 crotchets that one there finds on the anal extremity one here obsencs 

 on the nose-horn. When the pupa-skin ruptures in that case for the 

 emergence of the imago, the suture at the back of the head splits and 

 the head-cover remains united with the proboscis-sheath. 1 [ere, on the 

 contrary, the suture splits across the forehead and the headpiece 



