NOTES ON SCOTTISH WILLOWS 65 



As I have shown in the " Revision," the name Croweana 

 Sm. properly belongs to the abnormal state of 5. pJiylicifolia, 

 and since the hybrid seems to be undescribed, I propose for 

 it the title X Salix secerneta (S. purpurea x vS. phy lie if olid). 

 Of 6\ secerneta there are in Mr. Fingland's collection examples 

 from two different bushes ; and as these are markedly different, 

 it is desirable to give a description of each of them. 



No. 59. — Glenairly Bridge, Sanquhar, Dumfriesshire; 

 Sept. 3d and May 30th. — Shoots and twigs like those of S. 

 pJiylicifolia, quite glabrous. Leaves long and narrow for their 

 size, the largest about 2 inches long by \ inch wide, ovate- 

 lanceolate acuminate (often obliquely), paler green but scarcely 

 glaucous below ; margins finely serrate ; mature leaves quite 

 glabrous, some of the younger leaves sparingly hairy below 

 with long straight hairs. No stipules. Catkins (6) small 

 and narrow, the largest \ inch long, shortly peduncled, the 

 peduncle with about 3 small lanceolate leaves, which are some- 

 times hairy below. Scales spathulate, often concave, rounded 

 at the apex, base brown, upper half black, subcoriaceous and 

 often slightly shining, clothed with long white hairs. Nectary 

 short, broadly quadrate. Flowers monandrous, the filaments 

 of the stamens being connate for their whole length ; anthers 

 small, four-celled ; empty anthers not fuscous. 



From the leaves alone this plant might be passed over as 

 merely a form of 6\ pJiylicifolia, but in the flowers the pur- 

 purea element is shown by the structure (but not the colour) 

 of the stamens and also of the scales. At the date (3d May) 

 when the flowering specimens were gathered the catkins were 

 rather passe, which suggests that the flowering period is inter- 

 mediate between those of the parents. 



No. 60. — Same locality as No. 59. — Leaves and twigs of 

 5. pJiylicifolia. Catkins (c?) short and narrow, shortly 

 pcdunculated, with about two small peduncular leaves. Scales 

 spathulate, acute, lower half brown, upper half very black, sub- 

 coriaceous, somewhat shining. Nectary quadrate. Filaments 

 either nearly or quite free, or connate for a short distance. 



This form is very near S. pkylicifoliq, but the emulation 

 of the filaments, and in a less degree the structure of the seal 

 indicate its relationship with .V. purpurea. 



Mr. Fingland has found another very remarkable willow 

 1 1 



