246 THE ENTOMOLOGIST. 



remainder of the segments taper rapidly and regularly to the 

 anal segment, which at the rear is '4 ram. wide. The rear 

 outline of this segment when viewed dorsally is square, hut 

 when viewed laterally it is hooked, with the. point of the 

 hook directed ventrally. The segmental divisions are very 

 clearly defined ; the skin is rough and without noticeable hairs ; 

 the head and thoracic segments are smooth and slightly 

 polished; they are reddish-brown in colour; the head is rounded 

 in front and it is without a cocoon opener; the eye cases are 

 rather prominent ; the wing cases are slightly polished, rough, 

 and striated longitudinally ; they reach to the rear end of the 

 third abdominal segment; the abdominal segments have only 

 rudiments of spikelets. 



The pupa3 were kept in a room under conditions as to heat 

 that would be approximately normal, and the first imagines were 

 noticed on June 30th, on wdiich day nine had emerged. They 

 continued to emerge in about equal numbers each day until 

 July 10th, after which date odd ones appeared until the 19th, 

 on which day the last specimen emerged. 



Altogether eighty examples were bred ; these consisted of 

 37 ab. nigrana, 35 ah. semiustana, and 8 ab. cristalana. This 

 remarkable result means that although there are some fifty or 

 more forms of I\ cristana to be found in the locality from which 

 the parent moths were obtained, every one of the specimens bred 

 followed one of the three parent forms. It also seems to point out 

 that the union of the sexes takes place in the winter or early spring. 



The bulk of the specimens emerged in the morning, but this 

 was not invariably the case, for a considerable number were seen 

 with wings undeveloped at or about sunset. They usually sat 

 quietly on the leaf which had contained the pupa, but on being 

 gently touched they took a short flight, settled on tbe side of the 

 cage, ran quickly up it, and if allowed took flight at once, rising 

 until they reached the ceiling of the room ; on this they would 

 generahy settle, and when once fixed they would allow themselves 

 to be boxed without difficulty. 



With respect to the important question of the food-plant, or 

 plants, I confined single larvae separately with foliage of every 

 tree and shrub grown in the localities where the moths occurred, 

 and which I was able to obtain ; I found they would eat the 

 following: blackthorn, whitethorn, pear, crab-apple, dog rose, 

 Pyrus aria, plum, beech, oak, hornbeam, birch, elm, lime, and 

 hazel ; and that they rejected holly, privet, maple, sycamore, 

 bramble, sallow. Viburnum opuliis, Rkamnus frangxda, and aspeu. 



From this experiment it will be seen that although not quite 

 a general feeder, P. cristana has a very wide range of available 

 food, and this is perhaps unexpected in view of its decidedly 

 local distribution. 



My next experiment was to try and ascertain the food-plant 



