158 THE JOURNAL OF BOTANY 



although Kohert Brown regarded it as masculine. Polyyala (to ya\a) 

 should he neuter, although it has has heen treated as feminine until 

 recently (Chodat in Engl. Jahrh. Hi. Beibl. 115, 70 ; 1914). Abroma 

 anl Theobroma (to flpuifxa) should he neuter, not feminine. 



In order to show how inconsistent has been the allocation of 

 genders to generic names, a few examples are taken from the Index 

 Kewensis, which follows the general usage as regards gender. 



Ayathosma and Coprosma are treated as feminine, but JCylosma 

 and Saprosma as neuter. Dipterocarpus and JElcsocarpus are made 

 masculine, but Artocarpus and Didymocarpus feminine. J) i it it /hu a 

 and Phyllanthus are regarded as masculine, ^Eschy nan thus and 

 Adenanthos as feminine. Sclerochiton is given as masculine, Bra- 

 chychiton and lihodochiton as neuter. In most of these cases the 

 gender given in the Index was the one used by the author of the 

 genus. 



B. When the termination of the final word of a compound is 

 altered, the gender should follow it. 



Examples: — Cercocarpus (m.), Callicarpa (f.), Amphicarpcea (f.), 

 Onnocarpum (n.) ; Ceratocephalus (m.), Basycephala (f.), Dracoce- 

 phalum (n.), Polycephalium (n.); CaryophyUus (m.), Aciphylla (f.), 

 Bryophyllum (n.) ; Circceaster (m.), Cremastra (f.), Bellidi- 

 astrum (n.). 



C. Indeclinable names borrowed from non-classical languages 

 should be treated as neuter. 



Examples : — Manihot, Taonabo, Amelanchier, Ayati, Batatas. 



D. Compound names in which the final word is -anthos or -anthus 

 should be masculine. 



Examples : — Adenanthos, Galanthus, HeUanthus, Loranthvs, 

 Phyllanthus. Strictly speaking, these should be neuter (to artios), 

 but names of this kind are so numerous, and have been with such few 

 exceptions treated as masculine, that it would be very inconvenient 

 to change the accepted gender. 



10. Orthographic corrections should be made in cases where the 

 original spelling suggests a false etymology. 



Examples : — The generic name Anoua was derived from Anon, 

 the Hispaniola name for Anona squamosa. It was first mentioned 

 by Oviedo, Hist. lib. 8, cap. 18, as " Hanon " (1535); and in his 

 second edition (1547) under the form " Anon" (Mart. Fl. Bras. xiii. 

 pars 1, 14; 1841). Acosta, Hist. lib. 4, cap. 25, gave it the form 

 Anona, which was generally accepted by subsequent authors until 

 Linnaeus (Hort. Cliff. 222; 1737) altered it to Annona (Lat., a 

 year's produce), on the ground that Anona was a " barbarous " word. 

 To some it might appear that it was the alteration of the spelling 

 which was undesirable, as suggesting a false et3 r mology. Jussieu 

 (Gen. PI. 283 ; 1789) restored the form Anona, which has been used 

 until recently, and was accepted by Dalle Torre and Harms (Gen. 

 Siphonog. 174; 1901). Safford (Journ. Wash. Acad. Sc. i. 118; 

 1911) and Fawcett and Rendle (Fl. Jamaica, iii. pt. 1, 194; 1914) 

 have, however, adopted the spelling Annona used in Sp. PI. ed. 1. 



Trichodesma and Pcntadesma (derived from >/ cea^nj, bundle) 

 should be written Trichodcsme and Pentadesme (cf. Sphenodesmc) 



