BY J. H. MAIDEN AND J. H. CAMFIELD. 265 



from that of the normal species. It is obtuse, so that it is some- 

 what difficult to select the highest flower in the cluster, but this 

 is an easy matter in /S*. incarnata. When fully expanded the 

 flowers of the latter, i.e.., sepals and petals, are nearly at right 

 angles to the axis of the flower; in our variety they spread only 

 to an angle of about 30° to the axis. But the chief point of 

 difference consists, as already indicated, in the shortness of the 

 style (to which we have drawn attention in the name of our 

 variety) and the fact that it is always completely hidden by the 

 anthers. The style is generally produced beyond the anthers in 

 S. incarnata for ^ line. The anthers of the variety are also very 

 short. 



We have been giving attention to some of the New South 

 Wales Banksias, particularly those which are found in the Port 

 Jackson district (the original collecting ground for most of the 

 east coast species), and desire to offer some notes in regard to two 

 of them. 



Banksia latifolia, R.Br., was originally described in Trans. 

 Linn. Soc. x. 208. In Prod. 394 are Brown's own words, as 

 follows : — " Foliis obovato-oblongis spinuloso-serratis basi acutis: 

 subtus costatis reticulatis cinereo-tomentosis, perianthii unguibus 

 sericeis ; laminis glabris, caule fruticoso." Nothing is here said 

 in regard to the size, varation in outline, &c., of the leaves. 



In the figure in JBof. Mag. t. 2406 and in Bentham's descrip- 

 tion {B.Fl. V. 55.5) and all others we have observed, the plant is 

 figured or described as with leaves " obovate-oblong, often 

 truncate, irregularly serrate with short usually prickly teeth, 

 contracted at the base, 4 to 8 inches long, 1^ to 3 inches broad." 



We desire to invite attention to the fact that the species is 

 more varial^le than it is usually supposed to be, some forms, not 

 specifically different, having been confused (quite reasonably) with 

 B. integrifolia and even B. marginata. The confusion between 

 B. paludosa and this variety will also be alluded to. We think 

 that the great amount of variation in forms indubitably included 

 under B. latifolia should be indicated in a name, and suggest the 

 18" 



