2 



SALIX FRAGILIS, S. RUSSELLIANA, AND S. VIRIDIS. 197 



Turning now to some of the chief Continental salicologists, 

 Koch, in both editions of the ■ Synopsis 1 (1838 & 1844), makes 

 Russelliana a variety of fragilis. Grenier, in Grenier & Godron's 

 1 Flore de France (1856), cites S. Russelliana Sm. as a synonym of 

 S.fragilis L., 0. pendula Fr. Wiminer (' Salices Europseae,' 1866), 

 while quoting Russelliana under S. frag His- alba Wimm., thinks that 

 the name should be abandoned as dubious. Lastly, N. J. Andersson 

 (' Monographia' 1863, and DeCandolle's 'Prodronius' 1868), after 

 mentioning the views of many authors, thinks that the name 

 Russelliana should be consigned to oblivion since it is impossible to 

 find out now whether it was to a species or definite variety, or 

 rather to a single individual that Smith applied it, and since it 

 has been so variously used by different writers to denote various 

 ambiguous forms between S. fragilis and S. alba. He, however, 

 gives the name as a synonym of S. viridis Fr. Nyman (' Conspecti 

 quotes S. Russelliana Sm. as a major variety of S. viridis Fr., 

 8. Russelliana Koch as a minor variety. 



Amongst all this diversity of opinion, a glance at the descrip- 

 tions makes one thing clear, namely this, that the Russelliana 

 of many authors is not the Russelliana of Smith. 



To clear up, if possible, the confusion, it is necessary to go to 

 the fountain-head, and ascertain that which I suspect many writers 

 on the subject have neglected to do, namely Smith's own views, 



not of Russelliana only, but of fragilis. 



For willows, more than almost any other class of plants, it is 

 desirable to examine and compare authentic specimens. Unfortu- 

 nately in Linne's Herbarium, in the possession of the Linnean 

 Society, the only specimen named by Linne S. fragilis is one of S. 

 alba, and Linne's description of S.fragilis, though probably intended 

 for that species, is too vague to discriminate it with certainty from 

 some allied willows. In Smith's Herbarium, also belonging to the 

 Linnean Society, there are either no specimens of S.fragilis and its 

 allies, or what seems more probable, the packet containing them 

 has been mislaid. A comparison, therefore, of specimens is impos- 

 sible. Fortunately, however, in Smith's descriptions and drawings 

 there is no dubiety, and an examination of these reveals the 

 v ery curious fact that the fragilis of Smith is not the fragilis of 

 many modern botanists ! 



In this group of willows the most important characters lie in 

 the structure of the female flowers and fruit. Smith's description 



nana, << germinibus pedicellatis subulatis." itoch says ot m _ 

 "capsulis ex ovata basi lanceolatis glabris pedicellatis, pedicello 

 **ctarium bis terve superante;" and distinguishes Russelliana as a 

 variety, with more pubescent and usually more finely serrated 

 leaves; Grenier, that fragilis has the "capsule ovolde-conique, 

 attenuee au sommet, glabre, a pedicelle court (1 mill.), deux- trots 

 Ms pl m long que les glandes a peine visibles ;" and of 8. pendula Fr. 

 y^S* Russelliana Sm.), " capsules plus petites;" Andersson, for 

 Jwyilis, " capsulis eloneato-conicis, attenuatis, glaberrimis, pedicello 



