109 



hybrid, and eventually found three trees fairly close together, with the bark, timber and fruits about midway 

 between those of the other two, and corresponding with trees previously found in similar company (vide 

 vol. xxv, p. 716, and xxvi, p. 324). I am still unable to offer any definite opinion as to what these trees 

 really are. In a general way they seem to more nearly approach E. sideroxylon than any other local species, 

 but the fruits are smaller, the bark less rough, and inclined to be like that of the Box, and the wood yellowish, 

 while they seem too scarce to represent a distinct species. In this instance, as previously, they were only 

 found as the result of special search. Although these trees appear to be very rare, it may be mentioned 

 that in nearly every case they have been found in twos or threes. (R. H. Cambage, in Proc. Linn. Soc. 

 N.S.W., xxvii, 195, 1902). 



I wrote as follows : — 



The Ironbark -Boxes. — These seem to me to be indubitable hybrids. Mr. R. H. Cambage (these 

 Proceedings, 1900, p. 715) has pointed out the probability of an Ironbark-Box of the Lachlan being a 

 hybrid between E. sideroxylon and E. Woollsiana. He has also suggested the hybrid character of the 

 Ironbark-Box or Bastard Ironbark of Nymagee, and the White Ironbark or Ironbark-Box of Barmedman ; 

 and I would add the Cooburn or Black Box of Narrabri. I only mention these forms because they have 

 been referred to at some length in my paper on E. odorata, a species they closely resemble in bark, timber, 

 buds, &c. The foliage of the Ironbark-Boxes is duller than that of E. odorata, and the venation less marked. 



Mr. Cambage's observations as to the evolution of these forms are interesting, and must be borne 

 in mind in considering the relations of the western " Boxes." I confine myself at this moment to emphasising 

 the resemblance of these forms to E. odorata. 



It is not desirable to name these forms until after further enquiries as to their relationship. Mr. 

 R. H. Cambage writes to me : — " I should say that the Nymagee and Condoboliu trees are associated 

 with the narrow-leaved form of E. Woollsiana, and even the Barmedman ones are rather more the narrow 

 forms that the broad, but it is getting difficult in the latter place." The " Narrow-leaved form of 

 E. Woollsiana " is the form that, I recommend presently, should be known as E. Woollsiana, the broader- 

 leaved forms really belonging, in my opinion, to E. hemiphloia. (Proc. Linn. Soc. N.S.W. xxix, 763, 1904). 



Following is a description of E. Barmedmanensis, as complete as the material 

 permits : — 



A medium-sized tree. 



Bark hard, tough, dark-coloured, flaky furrowed, so hard that portions do not rub off; the 

 classification of it as an Ironbark Box is descriptive ; it is hard, but not so furrowed as an Ironbark. The 

 •bark contains a good deal of kino. 



Timber reddish brown; this is the colour after nearly twenty years' exposure, but it has darkened 

 with age. It is hard and interlocked. 



Juvenile leaves not seen. 



Mature leaves dull green, and covered with fine dots, pedunculate, straight or rarely slightly 

 falcate, not large, say up to 1 dm. long, and under 1-5 cm. in greatest width, tapering gradually into a fine 

 but soft point, and gradually at the base into a petiole of 2 cm. Venation almost invisible, secondary 

 veins making an angle of 30 deg. and more with the midrib ; intramarginal vein fine and well removed 

 from the edge. 



Inflorescence in axillary pedunculate umbels up to 5 in the head on rather slender peduncles 

 and pedicels; buds rostrate-ovoid, the rostrate or conical operculum varying in length. Although not 

 seen mature, the anthers are evidently truncate. 



Fruits about 5 mm. in diameter, nearly spherical to pear-shaped, tapering moderately sharply 

 into a pedicel of under 5 mm. Rim of moderate thickness, with usually four or five blunt valves whose 

 tips scarcely reach the orifice. Deciduous staminal ring. 



The type is " White Ironbark, Barmedman, R. H. Cambage, No. 1 (4834), suggested by Mr. Cambage 

 to be a hybrid between E. sideroxylon and E. Woollsiana. 



