352 MR. R. A. ROLFE ON THE SELAGINEJE 
sheet. Choisy enumerates this among the “Species minus note ;” 
but his S. geniculata is the same plant. The true S. geniculata, 
L. fil., he wrongly calls S. polygaloides, L. fil. 
S. DIFFUSA, Thunb.! There are three sheets under this name: 
sheet æ contains three specimens, sheet B one specimen, which I 
shall consider the types, as they agree with the description. 
I have not seen Choisy's specimen, but his description agrees 
admirably. Thunberg’s sheet y is another species, for which I 
find no name. The following description will serve to identify 
it, though I have only met with this single example as yet :— 
S. NIGRESCENS, sp. nov.— Perennis, ramulis virgatis dense cinereo- 
puberulis ; foliis solitariis subadpressis linearibus obtusis earinatis 
levibus ; spicis terminalibus elongatis v. oblongis densis, bracteis 
lineari-lanceolatis acutis ciliatis; calyce campanulato hirsuto 5-fido 
segmentis parum inzequalibus subulatis acutis; corolle tubo gracili, 
segmentis oblongis quam tubus triplo brevioribus; staminibus 
styloque breviter exsertis. 
Ad Cap. b. spei. Thunberg! 
Frutieulus. Folia nigrescentia, 1-2 lin. longa. Spica 1-2 poll. 
longa. Bractez 14 lin. longs. Calyx $lin. longus. Corolla 2 
lin. longa. 
Although the name ** Selago diffusa” appears on the sheet, it 
does not appear to have any thing to do with the description of 
that plant, of which Thunberg says “ ramis diffusis."' 
S. scaBRIDA, Thunb.! Two sheets with three specimens. Both 
Meyer and Choisy rightly understood the species ; but the latter 
described a new species as S. glandulosa! (DC. Prodr. p. 20) from 
a vigorous specimen of this plant with abortive spikes and a few 
notes under those species). The following appears to me to be the explanation 
of the difficulty :— Thunberg first identified, as far as he could, his plants with 
previously described ones, giving new names to the remainder; all of the names 
he wrote with ink at the bottom right-hand corner of the sheet. Afterwards, in 
his * Prodromus’ (and later in the MSS. of his ‘Flora Capensis’), some discre- 
pancies were discovered and rectified ; but the corrections were not noted on the 
sheets at the time, and never afterwards checked from the book. Adopting this 
hypothesis, which has been arrived at by a careful and exhaustive comparison, 
not a single species is missing. For an example, see remarks on a later page 
under Hebenstreitia scabra, Thunb. One reason why I call special attention to 
the above is, that the same difficulty occurs in other orders received at Kew for 
comparison, and doubtless throughout his herbarium, thus providing innume- 
rable pitfalls for any one working with it, unless each plant is carefully com- 
pared with the corresponding description, and vice versd. 
