1092 THE BUTTERFLIES OF XEW ENGLAND. 



continuous flight. We are, however, inclined to tliink tliat tlie presence of this vast 

 concourse of insects at the Bermudas was not owing to ordinary causes, and that we 

 must looli to some extraordinary means to solve the mystery. From a very extended 

 series of observations . . . we have become impressed with the fact that the largest 

 flights of birds occur there during the period of great atmospheric disturbance. From 

 the latter end of September to that of October, violent revolving gales are prevalent 

 throughout the region which comprises the east coast of the southern and middle 

 states and the Nortli Atlantic in those latitudes, for some 600 or 800 miles from land 

 [which might eugulph large numbers of these butterflies] . . .Hurled with amazing 

 rapidity along this cool aerial current, in the course of about three or four hours the 

 heated vapor arising from tlie Gulf Stream would be met with, and would it be con- 

 sidered as too imaginative to grant that the ascending warmth of that stream has 

 power suflicient to ameliorate the condition of the cool current, to stay its rapid 

 course and allow the animal freight to descend, which, then within a comparatively 

 short distance of the Bermudas, would seek the nearest land, . . . and aided perhaps by 

 perfect calm or favorable breeze, arrive at those distant isles, without encountering 

 the dangers, which, in the form of contrary winds, would most certainly accompany 

 an intentional migration to the islands? 



Oviposition. Eggs sent me hj ]\Iiss Eliot were laid, some on the 

 stem, some on the upper surface of the leaf of the food plant. They 

 hatch, in New England, in the latter part of the season, in five or six 

 days ; thus some, laid "on or about August 7 or 8," hatched in the night of 

 August 14-15 ; others, laid August 18-19, hatched in the night of August 

 25-2(! ; and others, laid August 19 and 20, hatched August 25 and 2(). 



Food plants. Dr. A. W. Chapman reared the caterpillar many years 

 ago on pai'tridgc pea. Cassia chamaecrista Linn. Abbot states in his 

 notes that it feeds on C. occidentalis Linn., but is "more frequent" on C. 

 chamaecrista : Miss Eliot obtained the eggs freely from butterflies caged 

 on the same, and I reared them easily on plants repeatedly sent me for 

 the purpose from New Bedford by Miss Eliot. Edwards tells me that he 

 fed on C. nictitans Missouri specimens sent him, and it is this jilant and 

 not C. chamaecrista, which grows in great beds in the railway cutting 

 north of Holyoke, where the butterfly is always found ; I cut off^ at the 

 root, however, and examined inch by inch some hundred plants on the 

 spot one mid-September day in a vain .search for egg or larva. Boisduval 

 and LeConte state that it feeds on various species of Trifolium and Gly- 

 cine, as well as Cassia, but I am inclined to doubt it, for before Miss Eliot 

 sent me plants from New Bedford, I lost specimens by not getting food to 

 their liking. I did not try either of these genera, but the yoimg cater- 

 pillars utterly I'efused Cassia marylandica, and rather than eat Desmodium 

 or Coronilla, preferred the entirely dry leaves of Cassia chamaecrista. 

 One, indeed, ate some Coronilla and filled himself green w-ith it, and 

 passed half a dozen pellets of excrements, but would touch no more of it. 

 Another to escape hung himself from a Coronilla leaf by a thread at least 

 two inches long. Another experimented with Desmodium, and ate the 

 parenchyma of the upper surface in two or three minute holes, indicating, 

 perhaps, the number of meals it had made, but died on the fresh leaf with- 



