198 THE JOURNAL OF BOTANY 



between cases where Linnaeus had an a and those wliere he has a 

 "type," hut that svich distmction would be wrong the present case 

 demonstrates, since in a note Linnaeus himself says " a in Virginia." 

 This shows that a may be understood in the Species Plantarum in all 

 cases where it does not occur, and that the first splitter may choose the 

 var. /3 as the type if he sees any good reason to do so. For this same 

 reason it seems right to use Vhmis glabra Huds., 17G2 (quoad p 

 '*typo" excluso), emend. Miller, 1768. 



DicUTARiA IsciiyEMUM comb. uov. Panicum Jschcemum Schreber 

 ex Schweigger, 1804, employs the earliest definite trivial for our 

 D. humifiisa Rich. The plant ("foliis . . . passim pilosis") referred 

 to under Digitaria Isclicemum Muhl. Descr. gram. 131 is Panicuni 

 sanguinale. 



Set ARIA Beauv. is nomen conservamlum, me judice. It has been 

 in use for a hundred years and Sefarla [Acharius] Michaux has been 

 neglected for the same period. 



Agrostis tenuis Sibth., 1794, is replaced by A. capiUaris L., 

 1753 [? partim ; non Linn, herb.] ; Hudson, 1762 ; Leers, 1775 ; 

 Hitchcock, 1920. I have not yet investigated this case. 



Agrostis alba L. is certainly Poa nemoralis var. unijiorn. 

 The correct name for ^. aiha auct. is either^, palustris Hudson, 1762 

 (Fl. Angl. 27), or A. stolonifera L., 1762, emend, (partim), but it is 

 not yet certain whether the latter name is more correctly referred to 

 A. veriiciUata VilL, cf. Asch. & Graebn. Syn. II. i. 172 (1899); 

 Hitchcock, 1904 (in Bot. Gaz. xxxviii. 141), and 1920 (in U.S. Dep. 

 Agric. Pnill. 772, pp. 128, 129). 



Phragahtes is retained by S. Si T., who show that Adanson had 

 chietly in view the common Reed. For those who reject non -binominal 

 works such as that of Adanson, it should be noted that the first 

 author to precise Arundo L. was Pal. de Beauv., 1812 (Agrost. 60), 

 who retained the name for Phragmites and made a new genus of 

 Donax, with which position I concur. Hitchcock (1920), however, 

 fixes Arundo Donax as the type of the genus Arundo from the 

 citation in Gen. Plant., viz., " Scheuch. t. iii. 14, 3," saying that 

 Scheuchzer's ixg. 14 of Tab. iii. (A, B, and C) represents A. Donax. 

 But D of the same figure is Phraginites, Avhich Scheuchzer describes 

 equally thoroughly. If the *'3" of Linnajus stands for C, Hitch- 

 cock's argument might hold, but, as it seems to be an assumption, we 

 ma}^ by the principle of residues, retain Arundo L. emend. Beauv. 

 for Phragmites [Adans.] Trin., 1820. 



PucciNELLiA Parlatore, 1850, is accepted as earlier than Afropis 

 Grisebach in Ledebour, 1853 [Trin., 1838, and Ruprecht, 1845, as 

 section onW]. 



Lepturus Trin. partim, et auct. recent, plur., non R. Br. is 

 replaced by Pholiurus Trin. (P. Jiliformis Schinz et ThelL 

 op. cit. 265). 



S. & T. remark that Druce (1917) in B. E. C. 1915 Report 416, 

 has taken up Scahiosa Virga-pastoris Miller for S. snaveolens 

 W. & K., because the Index Keivensis identifies it tlms. They point 

 out that this identification camiot be right and that tlie Taberna^- 



