rounded and their stalk very short; moreover in the above-mentioned state some of 

 the upper leaves -hecomc alternate or scattered. This particular Eucalypt was 

 noticed in Upper Gippsland by Mr. A. TV". Howitt, and near the Ovens River by 

 Mr. C. Falck." 



lie 2>ocs on to sav : — 



" There is every reason to assume that it is merely a state of E. pxdverulenta 

 (cinerea) mediating a transit to E. Stuartiana (cinerea var.) " (my parentheses). 



He goes on to say : — 



" Indeed, it was with some reluctance that E. pulverulenta became at all 

 accepted into the present work, from which all dubious species for distinct 

 illustration have been and are to be rigorously excluded." 



And again, — 



' : This finallv narrow-leaved form of E. pulceridenla, when vet in its voun<> 

 bushy state, has the leaves all broad and opposite ; but they do not continue in that 

 form, contrarily to what is noted elsewhere." 



These passages prove that Mueller felt — ■ 



(a) That E. cinerea in its typical form could not be separated from Howitt 



and Raich's specimens (p. 4). 

 (tj) That E. cinerea was connected bv a transit form with E. Stuartiana 



(that particular one which I have named E. cinerea var. multifiora). 



He accentuates his view of the dimorphic character of E. cinerea when he 

 adds : — 



" As remarked already, E. pulverulent a {cinerea) is distinguishable from E. 

 Stuartiana [cinerea multifiora) only in its foliage, holding the same relation to the 

 last mentioned congener as E. Eisdoni to E. amygdalina, as E. melanopldoia to E. 



crebra The bark of E. Stuartiana and of E. pulverulenta are very 



much alike." (My parentheses.) 



At p. 523 of my " Useful Native Plants of Australia" (18S9) I stated that 

 E. pulceridenla Sims (E. cinerea F.v.M. was meant) is very closely allied to E. 

 Stuartiana (secunda. — J.H.M.), and it is a question whether they ought not to be 

 united." 



2a. E. Stuartiana F.v.M. var. cordata Baker and Smith in " Researches 

 on the Eucalypts," 103 (1902). In Proc. Aust. Assoc. Adv. Science is, 343 (1902), 

 Mr. Raker repeats that Howitt's E. pulverulenta var. lanceolata should now be E. 

 Stuartiana P.v.M. var. cordata. Biker and Smith describe the flowers as more 

 than three in the umbel ; thus their plant is identical with my variety multifiora, 

 though only in part with Howitt's var. lanceolata. 



3. With E. Gunnii Hook. f. var. rubida Maiden (E. rubida Deane and Maiden). 

 When E. cinerea is in the lanceolate leaf stage, with its glaucous fruits in 

 threes, the general appearance of the specimens renders them very liable to be 

 B 



