296 



Wycc (A. Murphy) ; near Raymond Terrace (A. Rudder), wliicli is the most 

 northerly locality known to me. 



Mt. Pleasant, Penrith (J. L. Boorman). The most westerly locality known 

 to me. 



HtjhvkUsm. In vol. ii, p. 186, of my " Forest Flora of New South Wales," I 

 have drawn attention to a probable hybrid between this species and E. robusta Sm., 

 from Erina Creek, near Gosford. I will figure this interesting form when I deal 

 with hybridism in the genus. 



AFFINITIES. 



Bentham (B.Fl. iii, IJM) associates it most closely with E.pyrljormis Turcz., 

 E. conoidea Benth. {E. erythronema Turcz.), E. urnigera Hook., E. casia Benth. 

 on an anthereal basis as usual. ^iEueller mentions the following four species: — 

 E. leucoxylon, F.v.M., E. ccesia Benth., E. erythronema Turcz., and E. cosmophylla, 

 F.v.M. 



In its anthers it is very close to E. urnigera Hook, f., E. cordata Labill., 

 E. globulus Labill., and E. megacarpa F.v.M., some of which have been dealt 

 with in Part XVIII. 



1. With E. globulus Labill. 



It certainly has a strong affinity with the E. globulus group, but the timber 

 of E. globulus is pale coloured. Like some other species, it may not be jiossiblc to 

 place it next to one species without qualification, and in the grouping and 

 " genealogical " tree I shall submit later, it will probably be found that the intensity 

 of the affinity of some species to others can only be assessed on points based on 

 examination of all known characters. 



2. With E. Planchoniana F.v.M. 



E. longifolia can readily be determined from its fruit, which is usually in 

 threes. It has the largest fruit of any of the Eastern Australian species, except 

 E. Planchonlna F.v.M., whose fruits arc not only large, but they are ribbed. The 

 shape of the fruit in E. longifolia is characteristic ; the size and the sculpture of the 

 rim rary somewhat. 



The two timbers arc differcut, that of E, longifolia being red, and i?. 



Planchoniana pale coloured. 



