hin tee ies OS 
ER ESE SS RP Ea Legh eo > PR 
Se dase CA pe rine eet s Sse mee Ry As Ne eee ae Oe a i ie gg a eho i oust ee 
Be ah | igh adh eased eae ayia ih T SE PEM Aesth 1p alae sees Pee rt LS wey Veneer on eh 
1897 | CURRENT LITERATURE 137 
may be noted that the tricarpellary Melandryum of Mr. Williams differs 
materially in its limitation from the genus of Réhling, Rohrbach, Garcke, 
and other continental writers. 
Considering the extent of his task and the great number of forms 
treated, Mr. Williams has described few new species, and those made appear 
to rest upon strong characters. A few changes of name, which affect our 
North American species, may be noted. In S. campanulata Wats., Mr. 
nomenclature. For, if the type of a species is to be taken, not as that form 
which was originally described, but that which any subsequent writer may 
{from abundance of material in his own herbarium or the statement of others) 
regard as the commonest, agreement will be difficult indeed. The name 
S. Cucubalus, restored by Rohrbach and to be accepted by strict followers of 
the “ Kew rule,” is rejected on a combination of what would seem very weak 
grounds; first, Cucubalus is a generic name, although S. Armeria is kept up 
without question. Then S$. Cucubalus is said to be pedantic; why more so 
than various other longer and less euphonious binomials retained, does not 
appear. Furthermore, it is stated that there is a name, Cucudbalus inflatus 
Salisb., three years older than S. Cucubalus. What this has to do with the case, 
it is difficult to understand, for being under another genus this cannot come 
under the “ Kew rule,” and if Mr. Williams adopts the continental practice 
of taking up the earliest specific name, he must be aware that in this case 
there are earlier ones than that of Salisbury. Finally, the doctrine of usage 
is brought in to support S. zmflatus, yet Mr. Williams does not hesitate at 
another point in his work to replace the well known North American Ss. vere- 
cunda by S. Behrii Williams, an elevated varietal name never current in any 
flora. 
In the arrangement of species it is hard to see why S. monantha Wats., 
which, if not actually a variety of S. Doug/asii, must be a near ally, is rele- 
gated to § GasTROSILENE, with which it has no close affinity. In a prelim- 
Sees wpe ts eat S. purpurata placed 1 under S. Scouders “ex B. L. Rob- 
inson,” while as a matter of fact the type of S. urpurata, kindly loaned by 
Professor Greene, has proved on examination identical with the Siberian S. 
_ repens Patr. S. Hallii Wats., upon which (together with the ill-starred S. 
- purpurata oS Williams bases his. S. Scouleré var. costata, differs 
from S. S 
‘he : fth calyx, so that its spe- 
-wair anted. 
