340 THE BUTTERFLIES OF NEW ENGLAND. 



betakes itself to another and repeats the same process. Owing to this 

 peculiarity, this species is easily discovered, as the bent and eaten leaf 

 betrays its presence" (Edwards). Mr. Edwards found that it first shel- 

 tered itself very shortly after reaching its third larval stage, and that 

 it effected its object by biting off the principal ribs of the base of a 

 leaf, on either side of a midrib, after the manner of Vanessa atalanta, 

 and was thus the better able to pull the sides of the leaf together. But 

 in the later stages the ribs were not bitten, the caterpillar being able to 

 draw the edges together without that precaution. His observations were 

 made on Boehmeria ; with the elm, a more refractory leaf, the process is 

 slightly different (81:9); the caterpillar bites a couple of channels on 

 either side the leaf, starting from a centimetre's distance from the base 

 and cutting obliquely toward but not to the midrib, through two or more 

 of the ribs ; the corners of the flaps thus formed of the larger part of the 

 leaf are then fastened together by a few strands of weak silk, rarely extend- 

 ing more than a centimetre and a half beyond the corner. The rest, 

 therefore, flares open apically, and w^hen half eaten bears some resemb- 

 lance to a saddle. In these nests one never finds more than a single 

 inhabitant. 



Life history. In New England the insect is double-brooded. The 

 butterflies hibernate and appear upon the wing all through the months of 

 March, April and the greater part of May, occasionally into June, few in 

 number and generally worn and faded ; they lay their eggs on the tender 

 leaves of the hop and elm soon after their budding ; the eggs hatch in 

 four or five days ; the caterpillars grow rather slowly, the chrysalis state 

 continues from seven to eleven days, and they begin to transform to but- 

 terflies toward the end of June ; they do not become abundant until the 

 beginning of July, and a few stragglers emerge from the chrysalis as late 

 as the middle of the mouth. These fly throughout August and later. 

 The eggs are again laid late in July and throughout August, and the 

 second brood begins to appear about the 25th of August ; they continue 

 to emerge from the chrysalis until the latter part of September, and evi- 

 dently hibernate without much delay, as they become scarce by the 

 end of the month and are rarely seen in October. In one instance 

 some late caterpillars I found in Cambridge on nettle changed to chrys- 

 alis as late as October 9, and the butterfly appeared October 27. 



The butterflies of the first brood (i.e., first from eggs of the same sea- 

 son) are almost always of the form dryas, and those of the second, which 

 winter, are of the form harrisii, but not invariably. For instance, Mr. 

 Lintner took both forms on July 24, when only dryas should be flying, 

 and I have several times taken wintered dryas on the wnng about Boston 

 the first week in June, and as late as the 19th, and Mr. F. H. Sprague 

 has reared dryas (one out of more than fifty) from caterpillars found in 



