174 



3. E. perfoliata, Dura.-Cours., described as follows, from leaf specimens only, is 



" very doubtful."— (B.F1. iii, 200.) 



It may be convenient to draw attention to Dumont's imperfect descriptions 

 at this place, especially as he says his plant is the E. hypericifolia of English gardens. 

 With the aid of Kew I have been successful in obtaining Dumont's original 

 description, which is instructive, as showing how Eucalypts were described in the 

 early days. I have not been able to see the type : — 



Eucalyptus hypericifolia, Dum.-Cours., Bot. Cult, ed. 2, vii, p. 279. E. a feuilles de millepertuis, 

 E. hypericifolia, Hort. angl. Cette espece ne me semble former qu'un arbrisseau. Ses rameaux sont tres 

 menus et n'ont guere, surtout vers leur sommet, que la grosseur d'un fil. Feuilles opposees, lanceolees, 

 oblongues, pointues, tres-entieres, glabres, d'un beau vert en-dessus, un peu glauques en-dessous, de 4 

 centimetres de longueur, et de 4 a 5 millimetres de largeur. Lieu id [La Nouvelle-Hollande]. Toujours vert. 



E. hypericifolia, Link, according to a specimen in the Prodromus herbarium 

 is E. cneori folia. 



E. perfoliata, R.Br. (B.F1. iii, 253), is a Northern Territory shrub. There 

 is, however, E. perfoliata, R.Br, in Herb. Kew, which is E. Risdoni, Hook, f., 

 E. perfoliata, Tausch, in the Vienna herbarium, ex Herb. Ferd. Bauer, is also 

 E. Risdoni. The name perfoliata was loosely employed in regard to Eucalypts by 

 the early botanists. (See Proc. Linn. Soc., N.S.W., 1901, 550.) 



4. E. connata, Dum.-Cours. 



E. a feuilles cornices, E. connata, I but. angl. Tige droifce, grisatre, d'un rouge brun, ainsi que les 

 rameaux dans leur jeunesse, cylindriques el glabres. Feuilles connees, presque perfoliees, ovales a leur base, 

 oblongues, pointues, l^gerement ondulees a leura bords, tres-entieres, glabres, relev^es en-dessous d'une 

 nervure rouge et saillante, tres -ponctuees de points fcransparens, longues dun decimetre environ (4 polices), 

 larges de 16 a 17 millimetres (7 a 8 lignes). Lieu id [La Nouvelle-Hollande]. Toujours vert. — (Bot. Cult., 

 ed. 2, vii, 280.) 



Very doubtful (Benth.) ; E. diversifolia, Bonpl. (DC. Pi'od. iii, 220), with a 

 query. E. connata, J. Schauer, from Tasmania, in Herb. Vienna, is E. Risdoni. 

 If this species was described it would perhaps take priority of Hooker's name. 



Var. elata, Benth. 



A beautiful tree of the largest size, the bark of the trunk grey and deciduous, that of the extremities 

 of the branches purplish-red or reddish-brown (Gunn). Leaves broadly lanceolate-falcate, 2 to 4 in. long 

 rather thick, sometimes almost as in E. obliqua. Flowers of E. Risdoni. Fruit pear-shaped, 4 lines 

 diameter, with a broad convex rim — Lake St. Clair (Gunn). This variety in the dried specimens appears 

 to connect E. amygdalina with E. obliqua, but without doubt belongs to E. Risdoni, as observed by 

 Oldfield, although the dried specimens were included by J. U. Hooker among the varieties of E. radiata, 

 Hieb., now united to E. amygdalina. — (B.F1. iii, 203.) 



See my notes on Gum-topped Stringybark, p. 177. 



Var. elata is a drooping broad-leaved glaucous form, with broadish sucker 

 eaves, common (Mr. Rodway states) in mudstone country in Tasmania, and one of 



