Accuracy in Systematical Arrangement. 409 



one minute, a female might be observed tripping over some 

 huge block, and at the next, diving into some crevice beneath 

 it or one adjoining, and from the period which often elapsed 

 ere she and her attendant reappeared, I sometimes feared 

 that they had become inextricably involved. The fem.ales, I 

 imagine, always managed to find some outlet, and not unfre- 

 quently at a very considerable distance from the spot at which 

 they had entered. I am not aware that the Sylvia petechia 

 habitually pairs in such numbers ; the assemblage just noticed 

 might, therefore, have been in some measure accidental. At 

 the time in which it was noticed, one of tliose cold intervals 

 previously mentioned occurred, and large numbers both of tlie 

 Sylvia petechia and iS*. coronata were congregated with robins, 

 blackbirds, and other species upon an adjoining salt-Hat. 



Thougli thus pairing together in such numbers, the Sylvia 

 petechia breeds apart, and the localities which it seems to prefer 

 for this purpose, are moist, mossy barrens or swamps, sprinkled 

 with alder or stunted spruce trees. From the tops of the 

 latter they usually emit their simple but feeble notes. These 

 can scarcely be entitled to the name of sonsr, for if uttered in 

 a somewhat softer tone, they might readily be attributed to 

 the common grasshopper. As the pairing of the Sylvia pete- 

 chia occurs early in May, it is probable that two broods are 

 raised during the season. 



It might be supposed that the caudal vibrations, constantly 

 exhibited by this bird, would have awakened some suspicion 

 with respect to its present position. I am not aware, how- 

 ever, of anything of the kind having ever been expressed, 

 even by tliose writers who have had the most favorable oppor- 

 tunities for observing its habits. There is one circumstance, 

 connected with the habits of this bird, wliich is exceedingly 

 curious. In spring, as I have already mentioned, it resorts 

 freely to the ground, but in no instance have I ever observed 

 it there during its autumnal migrations. During this period 

 it seems to exhibit the habits of the true Sylvicolte, gleaning 

 among the branches of trees, and not unfrequently visiting the 

 windows of dwellings, in search of spiders and insects. This 



45 



