1921] Fernald, — Expedition to Nova Scotia 223 



undoubtedly a fertile source of introduction of many railroad weeds. 

 And possibly this colony of Croton glanihdosus may have originated 

 in such manner. 



Although the plant was not detected at any other station along the 

 railroad, the vigorous manner in which the species is establishing 

 Itself at this spot leaves no doubt in the mind that it is only a ques- 

 tion of time when it will have extended further along the railroad, 

 or even out into the adjacent sand-barrens. 



Jt is said to be a weed in portions of its range, and like many cro- 

 tons and allied species of the spurge family, to have an especial fond- 

 ness for railroads. At least one of its stations in Virginia, the state 

 usually given as the northern limit of the species, is " along the rail- 

 road between Lynchburg and Danville" and probably other occur- 

 rences on the outer edge of its range are of a similar character. It 

 would thus appear that the plant at Bishops Bridge has found a very 

 congenial habitat and it would not be surprising if what is now such 

 an unfamiliar species to local collectors should ultimately become a 

 familiar member of the New Jersey flora. 



Academy of Natural Sciences of Philadelphia. 



THE CRAY HERBARIUM EXPEDITION TO NOVA SCOTIA, 



1920. 



M. L. Fernald. 



(Continued from p. 105.) 



Panicum Lindheimeri Nash. As already noted (p. 141) there is 

 no constant character by which to distinguish from P. Lindheimeri 

 the several plants subsequently published as species and separated 

 by Hitchcock & Chase upon the minutest differences in size of spike- 

 lets and varying degrees of pubescence on the foliage. These plants, 

 /'. tnniesseen.se Ashe, P. huachucae Ashe, and P. implicatam Scribner, 

 have for a quarter-century baffled those who, not restricting their 

 studies to the grasses, are in the habit of looking in other plants for 

 essentially constant characters in species and who have long since 

 learned that in other groups at least, fluctuating degrees of the same 

 type of pubescence when unaccompanied by definite characters of 

 the inflorescence give very unsatisfactory grounds for specific separa- 



