384 PROCEEDINGS OF THE ACADEMY OF [1892. 



should be responsive at once to the force that induced it. One can 

 scarcely imagine a change in form to occur in a plant, responsive to 

 irritation received from external causes by its ancestors hundreds, 

 perha|js thousands, of generations previous. But if we take the 

 succession of forms we know to have occurred, and know to have 

 been evolved from closely related forms, as following in obedience 

 to some law of growth as yet hidden from us, we can have no diffi- 

 culty in suspecting that when the fulness of time shall arrive these 

 analogues of haustoria will have full parasitic functions. 



Polygonum cilinode. 



In a few instances, by no means common, I found July 12th 

 at Seal Harbor, Mt. Desert Island, branches of Polygonum cilinode, 

 which instead of being climbing were sarraentose, and, bending 

 over, had rooted at the tip and formed a large terminal bud as we 

 find in some species of Rubus. Many were found of a sarmentose 

 or trailing habit, with no disposition to climb, though the facilities 

 for climbing were within reach. In among the ramifications of the 

 roots of these rooting tips were numerous cleistogene flowers, perfect- 

 ing sometimes wholly underground. The flowers in the climbing 

 branches are of two kinds as I have noted in other Polygonums, one 

 always closed and fertile, the other open, apparently perfect in all 

 respects, but barren. The inflorescence is formed of continuously 

 branching axillary buds, and the only check to a further contin- 

 uance of growth, seems to be exhaustion. The growth ends with 

 depauperate buds. The species is evidently on the border line evolu- 

 tionarily speaking between the merely upright and the climbing 



species. 



Aster tatarica. 



This Asiatic species exhibits in its inflorescence a curious mixture 

 of characters separate in other species. The upper portion of the 

 panicle is corymbose and comprises about a dozen floAvers, which 

 are centripetal, the central flower opening first. The flowers are 

 quite gay with their numerous violet-colored rays. The lower 

 flowers on the outer branches of the corymb, are, however, rayless; 

 below this corymb is a conical panicle of a foot or more in length. 

 The lower branches of this panicle extend six inches or more, and 

 these lateral branches gradually shorten till they are scarcely an 

 inch long. These branches are all centripetal in their growth from the 

 central axis, but the flowers are centrifugal. In all the upper flowers, 



