EUCALYPTUS COEDATA. 



second stalk, bearing however but one flowerj from tbe same axis. Lid nearly half as long 

 as the tube of the calyx or variously shorter. Filaments pale ; anthers dorsifixed, gradually 

 attenuated at the lower end, the dorsal gland under the summit conspicuous. Style not 

 equalling the length of the stamens. Fruit from | to fully ^ inch in diameter, sometimes slightly 

 contracted at the summit, not rarely a little constricted beneath the ring-like margin ; capsular 

 vertex slightly convex. Sterile seeds partly very short, partly slender, and rather above a line 

 long ; fertile seeds obtusangular, measuring ^ to nearly one line. 



The shrub or tree flowers during July and August. It seems not to descend to the coast. 



As remarked by Bentham, ^T^cordata stands in relation near to E. cosmophylla and 

 particularly to E. pulverulenta, indeed the latter being illustrated as E. cordata by Loddiges. 

 E. cosmophylla however has elongated stalked and scattered leaves, the rim of the fruit broad 

 and the seeds more angular. E. pulverulenta has the branchlets generally more slender and not 

 acute-angular, the leaves not crenulated, but dotted with roundish almost uniform oil-pores, 

 the flowers generally smaller, the tube of the flowering calyx downward obconically attenuated, 

 while the lid is less depressed ; the fruit is smaller, more topshaped and has a comparatively 

 broader rim, the convergent free part of the valves emanates almost at a level with the calyx- 

 edge and arises not distinctly beneath the rim ; the furrow between the discal lining and the 

 calyx-tube is running just beneath the edge of the fruit, not forming a faint vertical channel 

 around the rim. Crenulated leaves occur also in E. urnigera, and, strange as it may appear, it is 

 to this species, that E. cordata bears the closest alliance ; for although the aged state of E. 

 urnigera has scattered long-stalked dark-green and lanceolar-sickleshaped leaves, more slender 

 elongated and downward more attenuated calyces on conspicuous stalklets with ampler lid and 

 urnshaped fruits with deeply enclosed valves, yet trees are now known (through Mr. Stephens 

 from " Old Man's Head," a subalpine mountain near Lake Crescent), which to all appearance form 

 a complete transit from E. urnigera to E. cordata. Moreover Mr. Aug. Oldfield sent many years 

 ago from the middle-regions of Mount Wellington sterile saplings as the young state of E. urnigera, 

 the adventitious lower shoots of which can in no way be distinguished ia foliage from E. cordata, 

 and which are also partly pruinous. On the summit of Mount Wellington I collected a state of 

 E. urnigera with all leaves nearly oval and with simply truncate-ovate fruits. Hybridism does not 

 seem to explain the origin of these aberrant forms in a genus, where against cross-fertilisation 

 is guarded by a calycine lid ; though — as pointed out by Mr. W. Sh. McLeay — the possibility of 

 such a process is thereby not absolutely excluded, as Parrots, Kakatoos and some other birds, 

 while feeding on young Eucalyptus flowers, may carry the pollen of one species to the stigma of 

 another. 



ExptANATioir OP Analytic Details. — 1, an unexpanded flower, the lid lifted ; 2, longitudinal section of an 

 unexpanded flower ; 3, some stamens in situ ; 4 and 5, front- and back-view of an anther, with part of its filament ; 

 6, style and stigma ; 7 and 8, transverse sections of fruit ; 9, longitudinal section of a fruit ; 10 and 11, sterile 

 and fertile seeds ; 12, portion of a leaf ; all more or less magnified. 



