THE SALIX LISTS IN THE 'LONDON CATALOGUE.' 471 



from the same two species do not all turn out exactly alike, not even 

 in the same sowing. In a crop of -S'. repens x viviuKdia seedlings, 

 there are eight or ten plants, the foliage of most of which sliow 

 some differences by which the plants can be distinguislied. Still 

 more have such differences appeared in a small crop of S. aiin'ta x 

 Lapponinii, which was raised from crossing a male and female plant 

 of the same hybrid. This fertilisation was made to test whether a 

 hybrid could reproduce itself with any constancy; and the result in 

 this case showed a good deal of variation as the result. It is not 

 surprising, then, that *S', Moorei and the *S'. herhacea x phylicifolin 

 artificially produced do not agree exactly. The former is a hybrid 

 on the herhacea side, retaining a dwarf creeping habit ; the latter is 

 on the phiilici folia side, having a strong tendency to ascend and 

 become a small bush. Tiiere is a similar discrepancy between the 

 foliage of the two ; but the catkins (female) have much in common, 

 and there is no essential feature forbidding a common descent. 



The result of the investigation is that S. Moorei = herhacea x 

 phylici folia, and replaces *S'. Grahanii in Dr. White's list ; while 

 S. Graliami in turn replaces S. Somwerfeltii Ands. (under H. Mijrsi- 

 nites) = S. herhacea X Mi/rsinites ; and S. herhacea x ni(/ricans dis- 

 appears, and has yet to be discovered as a natural hybrid. From 

 the "Addendum" list S. herhacea x nigricans? must also be erased, 

 and replaced by S. herhacea x phylicifolia {8. Moorei). 



1409 herhacea x a^trita {Maryarita F. B. White). The Eev. 

 E. S. Marshall has found this hybrid in E. Perth (89), a very 

 different-looking form from the Mid-Perth (88) plant ; consequently 

 the comital distribution is 2 instead of 1. There is a specimen 

 which I take to be the same hybrid in the Borrer Herbarium, at 

 Kew, labelled "A Myrsinites? . . . Caithness? 1808"; but since 

 the county is queried, it is open to question whether this can be 

 reckoned as an additional county. 



1410 reticulata L. There are four hybrids given in Dr. White's 

 list under this species ; two of them are entered in the Addendum 

 List, out of regard for Dr. Wljite's opinion ; but, in accordance with 

 the alphabetical arrangement there adopted for hybrids, they appear 

 one under -S'. Lapponum and the other under S. niyricans. I have 

 not seen specimens of either ; a plant supposed to be S. Lapponum 

 X reticulata, from Glen Fiagh, has grown in the garden (No. 50) 

 for years, but not flowered, and the evidence of iS. reticulata in it 

 has not become more apparent. It is therefore not reckoned. The 

 remaining two supposed reticulata hybrids are omitted. They were 

 described from imperfect specimens, one (sejuncta) from scrappy 

 leaf-specimens only, the other [soluta) from female specimens not 

 in good condition, with a good deal of uncertainty as to either 

 parent ; and I rather approve of the intention expressed at the close 

 of their description : — ■" This apparently very distinct plant I pro- 

 visionally name 8. soluta, but I am unwilling to place either it or 

 the other (viz. 8. sejuncta) in the list till they have been rediscovered " 

 {Journ. Linn. Sac. xxvii. 444). 



1411 purpurea L. On this group there is little to add to what 

 has been said above, since the differences between the two lists lie 



