^'i FLOR. TELL. 



A(l. Differ, ad Amaranthus cal. 3part. stam. 3 

 — Types all. the triandrou5 Amaranthus that 

 ought never to have been united with the pen- 

 tandrous, half the Genus belongs here. It is 

 said that A. oleraceus appears to unite both,, 

 having 3 or 5 stamens, if so it is XWieQononcus 

 a dimorphous sp. that ought to be a peculiar 

 Genus, Pentrius Raf. 



556. EuxoLus Raf. (well shut) diff. ad Di- 

 meianthus, fructus utriculus ovatus indehiscens 

 sem. ovata non lenticulata cal. longior — Type 

 E. deflexus, or Amar. do L. 



557. Amblogyna Raf. (obt. fern, fl.) diff, ad 

 Dimeianthns fl femineis infundibulif. Sfidis, 

 lac. obi. — Type Ambl. or Amar. polygonoi- 

 des L. 



558. Amaranthus L. Bajan Ad. as A. san- 

 guineus appears the type of this Genus, it must 

 with the pentandrous sp. retain this name. 

 Adanson's Amaranthus was Qelosia L. 



559. CELOSIDIA Raf. 1814. This family 

 differs from Alsinidia just like the Gomphreni- 

 des from the Amaranthides by having united 

 stamens. It differs from Gomphrenides by sev- 

 eral central seeds in the capsule. The habit 

 is nearer Amaranthides, leaves commonly al- 

 ternate^ — Types Qelosia^ Lophoxera, Snkana^ 

 Hyparete, but Cedrela does not belong to it, 

 nor Coilosperma^ nor Belutta, 



560. LopHOXERA Raf. (crest dry) ad Celo- 

 sia diff. stam. vix. coalitis, stylo trifido stig. 3, 

 caps. 3sperma et 6sp. — Types Loph. comosa, 

 paniculata, caudata, polygonoides, and race- 

 mosa (Cel. 3gyna L.) all Celosias. The true 

 Celosias have one style, 2 stigmas, 2 ar 4 seeds. 

 1'he Cel. lanata is now Aerua tomentosa. 



561. SuKANA Ad. Diff. Celosia, stam, 5 sine 



