202 PHYTOGRAPHIC EXPRESSIONS AND ARRANGEMENTS. 



Ascherson and Baillon, for Salsolacese. Like Candollea it was 

 clearly and fully described ; both were also illustrated at once 

 by delineations and have indisputable rights of precedence. The 

 priority of Siebera among Composite renders the homonymous 

 genus among Umbelliferse superseded by Didiscus, a name 

 universally familiar to Botanists and to Horticulturists also for 

 the last 60 years. If the law of strict priority is not throughout 

 to be observed, it will be quite incomprehensible, by what code all 

 exceptions and deviations are to be admitted. In a case like 

 that of Gahnia against Cladium, the first dual name in the 

 former genus establishes the claim, though as a genus Cladium is 

 of a much earlier date ; their supposed distinctions in this case 

 are rendered invalid by Carex within the same complex ordinal. 

 Had the due restoration of such genera as Wilckia, Uybanthus, 

 Bramia and Laxmannia been effected early in the century much 

 synonymy would have been obviated, and less hesitation existed 

 as to their recognition. In all this it must also be considered, 

 that a phytographic system is to serve for centuries, and not 

 merely to be in consonance with the traditions of a few 

 generations. 



A curious intrication exists as regards the name of one of the 

 most common of our Senecios, which since 1803 passed as S. 

 australis; it has recently been ascertained that its description by 

 Willdenow is based on a variety of Senecio lautus from New 

 Zealand, where our plant does not occur, so that the formerly 

 current name has only the authority of A. Richard (Sert. Astrolab. 

 t. 39) from 1833 for it ; but the plant constitutes the#. dryadeus- 

 of Sieber, mentioned in 1826 by Sprengel (Syst. Yeget. iii., 562). 

 This case is merely given to show what accuracy is required for 

 incontestable records in phytographic writings. Todea might well 

 be reabsorbed in Osmunda (perhaps with exception of the section 

 Leptopteris) in as much as 0. bipinnata effects the transit. (See 

 Hooker's Filices Exotica?, t. 9.) 



Now only remain some concluding remarks on ordinal affinities; 

 for but few of these characteristics are by themselves absolute, 

 not even that established already by Ptay in reference to the 

 number of Cotyledons ; thus in Australia Ceratophyllum, JVuytsia, 

 some Persoonias and Ccdlitris have more than two cotyledons, 

 while they are undeveloped in Cuscuta. The linear sequence of 

 course is the only one available for practical descriptive purposes;, 

 so that, even in the most careful systematic arrangement, it becomes 

 necessary to rely not too strictly on even main distinctive notes. 

 To vindicate therefore some of the changes ventured on in the 

 new publication, let us remember that numerous stamens occur in 

 the Tropic-Australian genus Distichostemon among Sapindacese, 

 and in two species of the Asiatic genus Megacarpa^a among 



