IXVII. EPACKIDE^. 145 



Corolla-tube very short; lobes very spreading, glabrous, more 



or less imbricate 21. Sprengelia, 



torolla-tube cylindrical; lobes erect, recurved or revolute, 



bearded inside, valvate in the bud 22. Andeesonia. 



leaves with an adiiate, sheathing base, which falls off with the leaf, 

 leaving annular scars on the denuded branches. Placentas de- 

 pendiDg from an ascendina;, recurved stipes. 

 i^orolJa circiunsciss near the base, calyptriform, the lobes not 



p°P;™g 23. RrcHEA. 



toroha not circnmsciss, the lobes spreading 24. Deacophyixum:. 



Tribe I. Styphelie^.— Ovules solitaiy in each cell of the ovarv, and 

 pendulous from the summit of the cavity. Style terminal. Fruit indehiscent, 

 usually drupaceous, 



1. STYPHELIA, Sm. 



(Soleniscia^ DC.) 



il ^"^^'^'^^^^ elongated, cylindrical or slightly ventricose, hairy inside at 

 m"d 11^^^^' ^^1^1 with 5 tufts of hairs, sometimes confluent in a ring below the 



1 ale or rarely glabrous ; lobes linear, bearded inside, much revolute, ex- 

 posing the stamens. Pilaments free from the throat, fflifonn, glabrous ^ an- 



ers exserted, free, linear, l.c(dled, attached about the middle. Hvpogynous 

 ^^caies distmct or united in a cup-shaped disk. Ovary 5-celled, with 1 ovule 

 ^^ each cell. Style filiform, longer than the corolla-tube; stigma smalL 

 cam ^- /^P^' ^^*h a dry or slightly pulpy mesocarp, and a hard, bonyendo- 

 pef iT ^ ^^^'^ ^"^^ ^^^^^ ^^' ^^^"^^^ ^y abortion.— Leaves sessile or scarcely 

 rar 1 9 ' ^'^^ers axillary, solitary, with the rudiment of a second or very 

 moa/ ^^ ^' ^^^ ^ ^^^^'y ^^^^^ peduncle. Bracts several, 1 or 2 of the upper- 

 the fTr ^^ ^^^^ enlarged, and embracing the base of the calyx as well as 

 still larger bracteoles. Calyx usually coloured. 



stanll^ ^T^^ ^^ hmited to Australia. The very much revolute coroUa-lohes and exserted 

 SDi«ll fl^ *^^s^inguish it from all its allies except Coleanthera, which is readily known by the 



tvTr^ ^^'d connate anthers. 

 then r ^^^ ongiually intended by Smith to comprehend the few drupaceous Epacride^e 

 jrrounr7-"u ^' ■^^^'''°' "^ ^*^*^inS a vast number of species, divided it into several 

 he 1a ^ * t. Proposed as distinct genera, considering the original genus as a tribe, but 

 heusiv ^^°*^ ^*"^^ ^^^t others might prefer to treat them as sections of one compre- 



k^ tip ^''''^^' Sprengel attempted to consoUdate some of these genera, but, by rearrang- 

 cr^t "f ^P^^i.es upon technical characters taken from books, without actual observation, he 

 have !, ? ^""^ confusion. De Candollc, Endlicher, Liiidley, Sonder, and others, who 



whos^ 1 ^'^ ^"^^ ^^"*^^^^ the Order, have adopted the views of Brown, the excellence of 

 »iadr ^""11 H^' ^'^^ ^^^ "ii"o^ exceptioiis, been fully confirmed by the additions since 

 ^th 't^n ^^^^^^^' ^'^^ter a careful study of a large number of species, which has supplied ns 

 genus di?ff' ^^^^'^'^^^^ observations (Fragm. iv. and vi.), returns to the idea of one large 

 ractera r ^ ^^^° sections, justifying the chan-e on the observation that some of the cha- 

 ^ediat/c - "^^"^ ^^^ *^^ distinction of genera"^ have failed, that there are frequently inter- 

 comprpV -'^^ connecting the several groups, and that Sti/phelia, if retained in its ongmal 

 oaeatterr^f ? T^^^' '® ^^*' ^^^^"^ ^^^> SO numerous iu species as Acacia and others, which no 

 ^ observ 1 1^ ""^ "^P- ^""J admitting the correctness of these statements, it may, however, 

 are not m ' ^° ^^^ ^^^^^'^ Vegetable Kingdom, there are few, if any, large genera which 



"^^^acia nr^^^ °^ ^^^^ connected with others by intermediate species; that in such genera as 

 ia the Jl , ""^^'"^^ ^^^ instance, throughout the uhole of the 500 odd species known there 

 ^'^erabW V^^^f^'-o^itj ia the structure of the flowers ; whilst in Stypheliece there is con- 



J«e aiversity, as well in the corolla as in the stamens and pistil; and that, if m the 



L 



