6 AUTHOR'S PREFACE. 



from Naracoorte to Beachport and Mount Gambier. In cases where a 

 plant has only been collected once or twice, the precise localities are 

 given. 



The system of classification adopted is almost entirely that of Engler 

 and Gilg's "Syllabus der Pflanzenf amilien, " which is essentially the 

 same as that followed by Engler and Prantl in the "Naturlichen 

 Pflanzenf amilien. " Th's classification, proceeding from the simpler 

 to the more highly developed forms of plant life, has been utilised 

 by the authors of all the more recent floras, and by Maiden and Betche 

 in their "Census of New South Wales Plants," published in 1916. 



The comparative characters given in the keys refer only to South 

 Australian species, native or naturalised. 



In respect to nomenclature, the rules and recommendations adopted 

 by the International Botanical Congress of Vienna in 1905 have been 

 strictly followed. One of the most important rules provides that the 

 earliest specific name given to a plant must be carried forward if 

 the species is transferred to another genus. For instance: a common 

 shrub in South Australia was originally named Stenochilus glaber by 

 Robert Brown in 1810 ; it was removed to Eremophila by Mueller in 

 the third volume of the Proceedings of the Royal Society of Tasmania, 

 published in 1858, under the name of E. Brownii. As this neglect to 

 preserve Brown's specific name constitutes a breach of the rules of 

 nomenclature, the name was altered to E. glabra by C. H. Ostenfeld, 

 in the third part of his "Contributions to West Australian Botany," 

 published in 1921, and now stands as Eremophila glabra, (R. Br.) 

 Ostenf. 



The same rule of priority of publication was applied to the names 

 of genera, but here a modification was introduced, by which, in order 

 to prevent confusion resulting from too many changes, certain generic 

 names which have been in general use for a great number of years are 

 to be retained, even though they are not the earliest. A list of "names 

 to be retained" ("nomina conservanda") is appended to the rules 

 of the Congress, but it must be allowed that it is drawn up in a some- 

 what arbitrary manner. Thus the generic names Hypolaena and 

 Patersonia, established by Brown in 1810, are retained in preference to 

 Color ophus and Genosiris, published by Labillardiere in 1806 and 1804 

 respectively, but no reference is made to Brown's Xerotes and 

 Angiantlius (1810), which have therefore to give way to Labillar- 

 diere 's Lomandra (1804) and Siloxerus (1806). However, it is only 

 by adhering closely to the international rules and recommendations 

 that we can ever hope to attain uniformity in nomenclature. 



Recommendation 11 states that specific names commemorating men, 

 when ending in a consonant and used in the genitive case, are to be 

 formed by adding the letters -ii, except when the name ends in -er, and 



