CEWT. I. 7 



fihf ob, apparently very distinct from CL alni- 

 folia^ leaves quite narrow although as long. 



19, C/. hradeata Raf. ramulis teretib* fulvo to- 

 mentosis) fbl. petiol. latoblongis, basi neutis^ 

 apice acum. ineq. serrulatis, subtus fulvescens 

 granulatis, nervis tomentosis : raceinis laxis> 

 bracteis lanceolatis fl. longior, calicib. fulvis, 

 toment acutis — very peculiar sp. of Florida, 

 leaves about 2 inches long and one wide, some 

 bracts one inch long, t 



20, C^. pumila Raf. repens, caulib. erectis bre- 

 vis slmpl. apice angul. pubens, fol. petiolis 

 hirsutis. obovatis utrinque acutis, mucronato- 

 serrulatis, glabris, nervis subtus hirsutis : ra- 

 ceme brevis basi laxo, bracteis linear, ped. 

 eq. calicib. obtusis furfuraceis — ^m South New 

 Jersey, stem only 4 to 10 inches high with a 

 singlri raceme and few small uncial leaves. If 

 it is a var. of CI. alnifolm, it is a singular de- 

 viatiait; 



21^ CI. oomdatn Raf. ramulis juigul. villosis, fol. 

 ovatis, basi oordatis, apice. obtuse acum. sub- 

 integris vix erosis glabris, subtus reticul. ner- 

 vis puberis : racemis laxis ebracteatis, cano 

 villosis, calicib. obtusis Canis furfuraceis — 

 sent me as the CL arborea, but totally un- 

 like, by cordate leaves 3 or 4 inches long, 2 

 broad, not serrate nor oblong, racemes not 

 spiked with rather long pedicels. Probably 

 African, t 



23, DicoNANGiA heterophyla Raf. Itea Virgin- 

 ica L. &c. from New Jersey to Louisiana. I 

 restore the excellent generic name of Mit- 

 chell and Adanson, the bad one of Linneus 

 being too short and meaning a Willow ! I give 

 it a new specific name quite appropriate,since 

 the same shrub has leaves ovate and obovate, 



