70 THE JOURNAL OF BOTANY 
S. triqueter. S. carinatus. S. lacustris. — 
Flowering-stems acutely | Flowering-stems terete Flowering-stems stout, 
triangular about 3 ft. below, obtusely trian- terete, 6-10 ft. 
high. i t. 
‘; igh. 
Leaves reduced to a taaves with or without | Leaves with lamina more 
sheath, lamina not | lamina. or less developed. 
developed. 
| 
Spikelets varying in | Spikelets never sessile, 
number, sessile, or on often long pedicelled. 
elongated branches. 
Spikelets two to many, 
sessile, and compact at 
the apex of elongated 
branches. 
Hypogynous bristles | Hypogynous bristles Hypogynous bristles 
equalling or a little mostly equalling the equalling or slightly 
shorter than the nut. ovaries, exceeding nuts. 
Anthers glabrous. Anthers glabrous or | Anthers fringed at apex. 
shortly ciliate. 
Styles 2. Styles 2-3. Styles 3. 
The character of the nuts we do not mention as nearly all the 
spikelets of S. triqueter and S. carinatus were found to be barren. 
n the case of S. triqueter this sterility may have been due to an 
abnormal season. Jackson found well-developed nuts in one or 
late good species, and may even become so; as, for instance, 
Potentilla aurulenta, which has arisen from the primary hybrid 
combination, P. verna x opac : 
ari nd S. triqueter-are associated on the banks of the 
Arun, near tley, Sussex, and by the Tamar, near Calstock, 
Cornw Tam unable to ascertain whether S. lacustris grows 
Aarostis Tenvrs Sibth. var. pumra (Linn,). Yeldersley Fields 
South Derbyshire, wild, 1906; the same cult., August, 1907. It 
is, I imagine, well known, since Mr. FE. S. Salmon’s investigations, 
