144 Rhodora [June 



crowded buttons. This plant was seen or collected throughout Yar- 

 mouth County and eastward in the silicious belt as far as Queens, 

 everywhere dominant and thoroughly characteristic of these sandy 

 and cobbly lake-margins. Close study, however, fails to reveal any 

 specific characters in the heads by which the Nova Scotia plant can 

 be constantly distinguished from the coastal plain Solidago tcnui- 

 folia and it is, consequently, here treated as a pronounced geographic 

 variety of the southern S. trnuifolia. 



We had been closely watching Utrictdaria cornuta for, when the 

 plants were still young and before the corollas expanded, we had 

 noticed that in some colonies the flowers were approximate at the 

 summit of the stem as good U. cornuta is supposed to have them, 

 while in other colonies or often in the same colony were plants with 

 the flowers scattered along the upper part of the stem, a character 

 which, with its smaller flowers, is supposed to distinguish U. juncca 

 of South America, the West Indies and the southern coastal plain. 

 U. cornuta was at last in prime condition and here, on the beach of 

 Trefry s Lake, were many plants with flowers as small as in the 

 smallest-flowered U. juncca, but dosely approximate; while at 

 neighb >ring lakes we found colonies with flowers larger than we had 

 ever before seen in U. cornuta but as remote as in U. juncca. U. 

 juncca is said to have a less spreading margin to the lower lip but if 

 this character proves no better than the others ascribed to it, it will 

 be evident that, when in 1847 Benjamin 1 treated the two as one 

 species, he was not far from the truth. 



Slightly beyond the Smilax tangle where Pease and I had turned 

 back there was a second mass of Cat Brier, only this was S. rotundi- 

 folia, var. quadrangidaris, a coastal plain variety previously known 

 northward to Nantucket and Cape Cod. The name quadrangularis 

 is most unfortunate, since the finer branches and branchlets of typical 

 S. rotundifolia are as often as not quadrangular, the distinctive 

 feature of the variety being its ciliate leaf. 2 While Long was gather- 

 ing specimens of the Smilax, I was absorbed in contemplation of the 

 golden-rod growing at the border of the spruce swamp, still immature 

 but surely Solidago KUiottii, a thoroughly distinctive species, origin- 

 ally from Carolina and Georgia, named for Stephen Elliott, the 

 great botanist of South Carolina, and "rare and local" even in south- 



1 Benj. Linnaea, xx. 305 (1847). 



; See Bk-knell, Bull. Torr. Bot. CI. xxxvi. 10 (1909). 



