366 



Tapinanthus, series Euhracteati, founded on Loranthus protei- 

 cola, EngL and L. nigritanus, Hook, f . ;* DendrophtJwe, series 

 Diplohracteati, based on Loranthus Tcwaiensis Engl. ;t Euloran- 

 thus, sect. Lepidotepalum, ba&ed on L. periclymenoidesj Engl. & 

 Krause;:|: Tapinanthus, series Breviflori, founded on L. mminalis , 



Engl. & Krause._ 



When the writer described tbe Loranthaceae of tropical Africa 

 in 1910, ke rejected the division of Loranthus into subgenera, sec- 

 tions and series, as being unnecessarily complicated, and leading 

 to tlie formation of artificial groups.ll For example, Euloranthus, 

 sect, Sycophila, is unquestionably more closely allied to Plicote- 

 palus, sect. Acrostachys, tlian the latter is to the two other sections 

 of PUcotepalus. The subgenera Deridrophthoe and Tapinanthus 

 appear to be artificial, individual groups assigned to the one sub- 

 genus having their closest allies in groups belonging to the other. 

 Thus the Rufesccntes appear to be allied to the Euhracteati and 

 Erectilohi, whilst the Infundihuli formes seem to be related to the 

 Furpureiflori. The only character by which Tapinanthus can be 

 distinguished from Dendrophthoe is the presence of a tooth on the 

 filament; this tooth is sometimes very minute (0- 1-0-17 mm. in 

 L. Hohtii) and may be easily overlooked. Thus the genus 

 Englerina {Dendrophthoe, series Englerina, Engl.) was founded 

 by Yjxn Tieghem on Loranthus Holstii, Engl., on the strength of 

 the inaccurate statement that this species had no tooth.l Engler 

 himself originally referred L. Holstii correctly to the section Lu:h- 

 nanthus, in spite of the tocth being, as he thought, missing. Simi- 

 larly L.irangensis, Engl. [Tapinanthus series P ur pur ei fiori) was 

 re-described fourteen years later as a new species, assigned to 

 Dendrophthoe, series I nfundihidi formes, owing to the tooth on the 



filament being overlooked.** The latter case illustrates the close 

 resemblance which i ndividual groups of the subgenus Dendro- 

 phthoe^ bear to others of the subgenus Tapinanthus, which suggests 

 that the presence of a tooth on the filament has no greater 

 taxonomic value than transverse septation of the anthers, or the 

 presence of verticillately branched hairs. 



The writer accordingly divided the tropical African species of 

 Loranthus uiio 29 sections: twenty-four of these representing 

 groups previously recognized by Engler, and five being new.tt He 

 reduced the following groups : Englerina, Metula, Inflath Denfi- 

 tnctulaixnd Breviflori. Loranthus viminalis, Engl. & Krause. 

 the type of the Breviflori is in the writer's opinion a species allied 

 to £.. Adolfi-Friderici and L. rugegensis. The flowers appear to be 

 fn r Tf?'/"^ - '■ ^T""' ''''''?''' along with normally developed ones, 



tain lf.t '''"r^^ ^- ''?,^/f^^<^^oides, and have been found to < 

 norm^™'i v ^'^'^ '!•" ?^i^^^i^«' however, that the flowers 



con- 



and has accordingly revived the series Breviflori, tt 



Engl. Jahrb., vol. xxx., p. .303 fl901). 

 ; . vol. xl., p. 522 (1908K 



t ' vol. sliii., p. 400 (1.909). 



»T^ '^, m ■' ^'^y ^^"'•' P- 314 (1909). 



J^;.^h|!ir^!';^ ^l^l 1-^- ^aHH,.. .1. .., p. 1 .6., 

 n ^'7Tv'r^ntN^^^^^^ and BreetUoI^L 



