BY J. H. MAIDEN. 223 



F.V.M., is near to E. arnygdaluia, Labill., applies with a good 

 deal of force to Tasmanian specimens, the superficial resemblances 

 of herbarium specimens being frequently very considerable. 



The type of E. gracilis, F.v.M., is a South Australian specimen, 

 and was collected by Dr. H. Behr. It was labelled by Mueller in 

 Herb. Melb. as "Eucalyptus gracilis, Ferd. Muell. Murray 

 Scrub. Dec. 1848. Behr, Nov. Holl. Austr., Dr. Ferd. Miiller." 



It will be observed that neither in Mueller's nor Miquel's 

 description is there any mention of an angular calj^x which is so 

 obvious a feature of the typical E. calycogona, Turcz. 



It, however, passes by insensible gradations into the angular 

 form. 



3, E. gracilis, F.v.M., var. hreciflora, Benth. Calyx-tube scarcely angled, 

 1^ to nearly 2 lines long. Fruit about 2 lines only, but the deeply sunk 

 capsule and the stamens entirely as in the ordinary form — Darling and 

 Murray Desert, also F. Mueller's Spencer's Gulf specimens, which being in 

 fruit only are somewhat doubtful (B.Fl. iii. 211). 



Bentham's specimens cannot be traced in Herb. Melb., but I 

 am indebted to Kew for a fine drawing of the original specimens, 

 and for fragments of the specimen which place its identity beyond 

 doubt. 



I am of opinion that E. gracilis, F.v.M., and E. gracilis, F.v.M., 

 var. hreviflora, Benth., are so closely allied that it is impossible 

 to separate them even as two varieties. I think that they should 

 form one variety readily noted by its hemispherical operculum 

 and almost entire absence of angularity in calyx or operculum. 

 I propose the name gracilis for this variety, i.e., E. calycogona, 

 Turcz., var. gracilis. 



Range. 



Western Australia. -Coolgardie (Nos. 100 and 101; 1899; K. 

 Helms). Sap-green leaves, very shiny; fruits small, pear-shaped, 

 constricted at the mouth, but not ripe ; operculum a little 

 pointed. 



Specimens from South Australia [no locality] (W. Gill, 1896 

 and 1900) are very close to this form. 



