52 



botanical gazette. [ March, 



the old, blackened, still clinging stipules. The capsules are 

 rather larger than in the type specimens, with the same bifid 



stigmas ; scale narrower. 



I can not imagine why Prof. Andersson went out of 

 his way to give as the habitat of this species, "Near Fort 

 Franklin, subarctic America, Richardson, Douglas," for 

 Richardson never collected S. Barrattiana anywhere. Doug- 

 las never collected anything at Fort Franklin, and finally 

 the credit belonged exclusively to a third person, not men- 

 tioned at all, who found the plant in an entirely different lo- 

 cality, a locality confirmed by Macoun's rediscovery. 



4. S. Hookeriana Barratt. This, as observed by Prof. 

 Sargent, is a small tree 20 to 30 feet in height, with a trunk 

 rarely 12 to 18 inches in diameter, *' more often a low strag- 

 gling shrub with many prostrate stems : on the coast gener- 

 ally along the edge of sea beaches, or in low, rather moist 

 sandy soil." Mr. Howell's experience of the general habit 

 of the species is : " a small tree or large shrub, 10 to 20, or 

 rarely even 30 feet in height, with usually several stems from 

 the same root 4 to 8 inches or more in diameter. It is a plant 

 of the sea coast and salt marshes, usually growing on the 

 margin of ponds, but confined in its range to the immediate 

 proximity of the sea. Very abundant about the mouth of 

 the Columbia, and perhaps the commonest willow of the coast 

 from Port Oxford, near the south line of Oregon, north to 

 Vancouver Island. I have never found it away from the di- 

 rect influence of the sea. Mr. Nuttall's locality, i outlet of 

 the Wahlamet,' must be a slip of the pen for outlet of the 

 Oregon (Columbia). I live at the mouth of the Willamette, 

 within a few miles of Wyeth's old station, where Nuttall 

 made his headquarters, and I know that no such tree grows 

 here. I am quite satisfied in my own mind that S. Hooker- 

 ania does not occur inland, though it might possibly be found 

 near a saline spring." 



Dr. Barratt describes the capsules as " glaberrimis," and 

 this is emphasized by Hooker's remark, "pistil perfectly gla- 

 brous, even its stipe ; M but while true of the type specimens 

 (the capsules of which are past maturity), I find that more 

 frequently the capsules, especially when young, are silky. 

 This is quite noticeable in a form collected by Prof. Macoun 

 on Vancouver Island, and goes to confirm the seeming affinity 

 with S. Barrattiana. Furthermore, mixed with the type 

 specimens of S* Barrattiana, coll. Douglas, are pistillate 

 aments without leaves, but showing shorter stigmas invaria- 



