384 



(As a matter of fact, this was done by Deane and Maiden on morphological 

 grounds in Proc. Linn. Soc. N.S.W., xxiv. 400 (1899), as soon as they obtained access 

 to a tvpe or co-type of E. dives. The question of the oil had nothing to do with 

 recognition of the species). 



The subject is more fully treated in the article " The comparative constancy of 

 the oil products from individual species of Eucalyptus."' at p. 423, which, however. 

 introduces no new facts. 



I have on no occasion disputed the usefulness of oil determinations as a supplement 

 in or a cheek on taxonomy, but 1 dispute the lengths to which the interpretation of them 

 has been pushed. The vast majority of Eucalyptus names have been given, and will 

 continue to be given, without reference to oils. 



Ai p. 68, under E. Blaxkmdi .Maiden and Cambage, we are told— 



The Fruits cannot !»■ separated on herbarium material from those ol E. cwpitcllata, II requires a 

 knowledge of the barks, and a chemical test of tin- oils to differentiate the species from li.capitdlala, 

 as the two are very closely related." 



As a matter of fact, the authors separated this species on morphological grounds. 

 As is usually the case, the assistance of oils, if it conies at all.- conies after a derision has 

 been arrived at otherwise. 



1. At ]>. 73 the authors' statement that 1 confirmed the specific differences 

 between E. panicukita Sm. and E.fascieulosa F.v.M. by calling in the evidence of bark 

 and timber is testimony that 1 am broad-minded as to the use of characters. 



2. E. leucoxylon and E. sideroxylon were separated on the evidence of seedlings 

 and juvenile leaves, bark and timber. 



3. At p. 177, the authors state— 



Bentham. however, in his Flora Aiislraliensis, iii, 230, places E. elceopliora with A', goniocah/x 

 . . . Bentham's results were founded on herbarium specimens, and as recent experience has shown 

 thai Bucalypts cannot be determined on such material alone, &c. . . ." 



As a matter of fact. I restored R. ekt'ophora to specific rank on morphological 

 grounds. I only mention these last three cases as well-known examples of the use of as 

 many morphological characters as possible. The use of oils does not come into the 

 question. 



Turning to p. 42 (2nd Edition) of Messrs. Baker and Smith's work', in saligna 

 var. pallidwahis, there are certain morphological characters together with glaucousness 

 and the texture of the timber, which separate il from E. mliyna. I call it a new species, 

 /•.. grandis, bul Mr. Baker, while admitting there is considerable variation, still keeps il 

 under E. saligna as a variety. 



I reiterate the question which I asked in Part VIII, p. 251 (referring to oils) — 

 Wh.it variation in amounl of a constituent, or what constituent must be present or 

 absenl in any particular case to constitute a valid species? '' Certainly the oil character 

 j 8 no t the one invariable (Part VIII, p. 249). 



