1548 THE BUTTERFLIES OF NEW ENGLAND. 



performed by means of the legs or the wings, but I rather think the former. When 

 she was gone I opened the sheath formed by the leaf round the stem, and found therein 

 about thirty small white eggs deposited in a line. 



Habits of life in the early Stages. The caterpillars feed almost ex- 

 clusively upon Graniineae or allied endogenous plants and each forms for 

 itself a vei'tical nest by connecting several blades of the plant it is eating 

 by a loose, thin, open web of slender threads. I have never found any 

 that would not feed on Gramineae, or refuse to lay eggs on that family 

 of plants ; but two of the European species, Erynnis comma and Augi- 

 ades sylvanus, are said to feed on several genera of other families, the 

 former on Leguminosae, the latter on Malvaceae, though Gramineae are 

 also given as the food of the latter. There can be little doubt, however, 

 that as a general rule the caterpillars of this tribe feed on endogenous 

 plants, in contrast to the Hesperidi, which certainly prefer Leguminosae arid 

 are not known in a single instance, as far as I recall, to feed in nature on 

 endogenous families ; oviposition on these plants, however, is not essential, 

 since the eggs of butterflies whose caterpillars feed on grasses are often 

 laid indifferently on all sorts of other plants or on dead sticks, and I once 

 found an egg of this tribe on a thistle, but the caterpillar died upon a 

 thistle leaf without touching it. 



On escaping from the egg the caterpillars nearly always devour the en- 

 tire egg shell ( excepting the base ) , before their feet touch any other than 

 the shell surface. The caterpillars live almost entirely in concealment within 

 the nests mentioned, rarely leaving them, but usually reaching out for their 

 food from this cylinder and withdrawing at the slightest alarm ; to aid 

 them the caterpillars, in their earliest stage, but never so far as I have been 

 able to see in any later one, are provided with long, recurved hairs on the 

 last abdominal segment, by which a securer hold must be established within 

 the very narrow nest. Moreover, they rarely feed at any other time than 

 at night, so that a sight of one of these caterpillars, unless the nest is 

 picked to pieces, is indeed a rarity ; and this is the more striking when we 

 consider that the butterflies of this tribe form a significant proportion of our 

 butterfly fauna. In keeping with this is our almost total ignorance of any 

 parasitic attack upon the members of this tribe. I believe that none what- 

 ever have been recorded previous to those mentioned beyond. 



But there are one or two exceptions to this marked seclusion of life. 

 No one who has raised many of the species will have fixiled to notice the 

 bloom which appears on the back of some of the nearly grown caterpillars. 

 Thwaites tells us (Moore, Lep. Ceylon, i :165) , of the East Indian Gangara 

 thyrsis, which feeds on Palmae, that "from the body a loose shaggy fila- 

 mentous clothing, consisting of pure wax, is excreted, but which is easily 

 rubbed off when handled, leaving the larva quite naked," and de Nic^ville 

 tells me, what is of special interest in this connection, that this caterpillar 

 lives entirely exposed, stretching itself out upon the palm leaves. 



