136 BOTANICAL GAZETTE [ FEBRUARY 
knowledge of the order, and his long expected monograph™ of its most 
difficult genus is a very welcome addition to botanical literature. 
e work, which fills nearly 200 octavo pages and recognizes 390 species, 
is regarded as supplementary to Rohrbach’s Monographie der Gattung 
Silene. Species fully treated by Rohrbach are not redescribed, but only 
enumerated with brief bibliography. Species of later date, however, are 
well characterized in Latin. In its scientific aspect the work is decidedly 
British. The species and varieties are of the Benthamian sort, and there is 
no attempt at the elaborate varietal and formal subdivisions popular with 
and sometimes inordinately multiplied by continental monographers. Unfor- 
tunately exsiccazi are not cited, which is a considerable defect. Surely the 
enumeration, under each species and variety, of a very few authenticated speci- 
mens would have added much more to the value than the bulk of the work. 
One of the most interesting features of Mr. Williams’ monograph is the 
attempt to transfer from Silene to Melandryum a considerable number of 
American and Asiatic species, chiefly those of Watson, Franchet, and Max- 
imowicz. Recognizing the close affinities of certain large-flowered Silenes 
to species of Lychnis of the ZL. dioica type, various continental botanists 
have, since the beginning of the century, sought to unite them as an inde- 
pendent genus, Melandryum, or, as originally spelled, MWelandrium Rohl. 
Various combinations of inconstant characters have been devised to limit 
this natural but ill-defined group, the strongest being the greater inflation of 
the calyx and the complete absence of the partial septation usual in the cap- 
sules of Silenes. While restricted as by Rohrbach to such species as 5. 
Baldwinii, S. Virginica, etc., the genus Melandryum seemed to have, as to its 
American representatives, a tolerable habital unity, which gave it a certain 
raison @étre. Mr. Williams, however, by giving up all distinctions except the 
septation of the capsule, and attempting to apply this consistently, feels him- 
self forced to transfer to Melandryum also.a number of species of the char- 
. acteristic Eusilene type, such as S. Palmeri, S. Lemmoni, S. Bernardina, S. 
 platyota, S. Shockleyi, and S. Thurbert. Large genera, however, are seldom 
sfactoril I 
- especially, in advised when ‘based, as in this case, upon the p resence of a 
se val such as these partial septa, which exhibit all stages of 7 
— But even if the desirability of such a generic distinction were 
litted, eek writer E conld not saree with Mr. Williams in excluding foe - 
, both of which i i 
