382 ruoCREDiNGS of societies. 



A salient objection to the above may be the wide remove between 

 the Ilydrocharidea and Alismacea, which have rosemblanocs in afjuatic 

 habit, embryo, perianth, and stamens. The Scitaminem take the above 

 phxce rather, because elsewhere they would break through more natural 

 series than that they are closely affined to Ilydrocharidem. OrchideiE are 

 somewhat circumscribed, the approach to Scitaminem being in irregula- 

 rity of flower, and to Bunnanniacece in minute exalbuminous seeds and 

 homogeneous embryo. The hitter order is connected with the Orchidece 

 by ovary and seeds, to Iridecn by centrifugal inflorescence. The 

 Iridece commence a long series of orders extending to and including 

 Cyperacece, the boundary lines being indefinite, and cross-relationships 

 existing. The former arc distinguished by equitant leaves, centri- 

 fugal inflorescence, regular or oblique triandrous flowers with anther- 

 cells turned outwards, though none of these characters are constmt. The 

 division Coronariccc is taken in Endlicher's extended sense. As to the 

 Liliacece, the author in the main accepts Mr. Baker's revision of that 

 order. Pontederiacecb differs but little from the Lily group. Phillnj- 

 dracecB, comprising three genera, in ovary, &c. may be referred to 

 Xyridece, but in ovule and basal embryo is linked to Liliacece ; 

 flowers, stamens, and staminodia are those of Scitamiiiecc. The 

 JuncacecB are doubtless connected with the Liliacece. Palms are 

 universally recognised as a substantive order, some even would raise 

 them to a separate class, yet their woody stem and peculiarities of leaf 

 are the main characters ; flower and fruit being very like those of 

 JuncacecG. The Nudijivra proper, with five orders, are by all acknow- 

 ledged as closely affined, though variable in certain structures. The 

 Alistnacece are regarded as an anomalous order connecting in some 

 measure the Naiadece with Hydrocharidece. The Glamales are a 

 group now very generally admitted, although with different limits, 

 according to the notion attached to the term glume. Ou 

 this point the author holds that the organs enclosing the essen- 

 tial parts of the flowers of Glumales are all either perianth - 

 segments, or more frequently bracts or bracteoles performing 

 the functions of very much reduced or absent perianths. And 

 there would be no absolute error in the uniform employment of these 

 three terras whenever the distinction between them is clear, and of 

 the more general terms scales or settc when doubtful. But as the 

 biological functions of bracts or bracteoles increase in importance, so 

 also does their character acquire fixity, and convenience arises in the 

 introduction of the term glume for the primary bracts of the spikelet, 

 and of special ones — perigyniura, palea, lodicule, &c — for modifications 

 of the secondary scales in particular groups. It is, however, very 

 essential for a coriect understanding of the floral structure that wlien- 

 ever possible a definite meaning should be given to each term used. 

 The term glume has been too widely employed — e.g., for the perianth- 

 segments of Itestiacea. Mr. Beutham follows out this subject 

 by comparisons of the structures extant in a great number of genera, 

 first insisting on two great laws in terminology: — 1. Homo- 

 logous and generally similar organs should be designated by the 

 same name. 2. Where the want of homology of two organs has 

 been demonstrated they should be called by difiereut nanu's. 

 An exception to the first would be — where in a series of homo- 



