510 PROCEEDINGS OF THE AMERICAN ACADEMY. 



ser. 5, xviii. .374 (1873), the type of which, collected on the Isle of 

 Bourbon by Richard, is now in the Gray Herbarium. Other speci- 

 mens of F. flcxuosa at hand show the species to have a fairly wide 

 range of variation as to leaf -contour, in some cases reaching the short, 

 subcordate, conspicuously cuspidate-caudate form of Klatt's type, in 

 others passing to elongate lance-oblong shapes more or less cuneate at 

 the base. 



Celmisia Cass. Diet. Sei. Nat. vii. 356 (1817). This genus of the 

 Senecioncae was founded by Cassini upon a single South African 

 species, which he called C. rotimdifolia and which later proved to have 

 been the plant previously named Arnica tabular is hy Thunhev^, Prod. 

 Fl. Cap. 154 (1800). The genus Celmisia, thus having been originally 

 based upon a single species, there can be no question as to its type. 

 Furthermore, the name has not been mentioned in the lists of nomina 

 conservanda or nomina rejicienda, so it must take its covu'se under the 

 rules of priority. Cassini subsequently. Diet. Sci. Nat. xxxvii. 259 

 (1825), included in his genus a plant collected by Gaudichaud in 

 Australia, namely C. longifolia Cass., a plant belonging to the Astereae. 

 When the Composiiae were treated by DeCandolle he unfortunately 

 took this latter, Australian plant as the type of the genus, and referred 

 the original South African species to a newly named genus, Alciope 

 DC. Prod. v. 210 (1836). This was, of course, contrary to the clause 

 of Article 45 of the International Rules, which reads: " If the genus 

 contains a section or some other division which, judging by its name 

 or its species, is the type or the origin of the group, the name is re- 

 served for that part of it." DeCandolle's treatment has been gener- 

 ally accepted and perpetuated until the present day. It is true. Dr. 

 Otto Kuntze noted the inconsistency and restored Crlmisia to its 

 original application, but he referred the Australian and New Zealand 

 element of the complex to the genus Aster, a reduction not likely to be 

 generally followed upon taxonomic grounds. The current Inter- 

 national Rules appear to call for the following revision of the nomen- 

 clature in the genera concerned. 



Celmisia tabularis (Thunb.), comb. nov. Arnica tahularis Thunb. 

 Prod. Fl. Cap. 154 (1800). Celmisia rotundifolia Cass. Diet. Sci. Nat. 

 vii. 357 (1817). Ligularia tabularis (Thunb.) Less. Syn. Comp. 390 

 (1832) by implication. Alciope Tabularis (Thunb.) DC. Profl. v. 

 210 (1836). 



C. tomentosa (Burm. f.), comb. nov. Conyza tomentosa Burm. f. 

 Prod. 26 (1768). Arnica lanata Thunb. Prod. Fl. Cap. 154 (1800). 

 Ligularia lanata (Thunb.) Less. Syn. Comp. 390 (1832) by implication. 

 Alciope lanata (Thunb.) DC. Prod. v. 210 (1836). 



