Genus Pultenaea. 211 



reduction of species could place five or six, including P. stricta^ 

 P. retusa, P. Gunnii, under one species name, and cultural experi- 

 ments may yet prove that they are not all valid species, but few 

 Avould he bold enough to make such a drastic reduction. 



Even if it were proposed to unite, say, the three mentioned,, 

 which certainly are very close as regards floral structures, the work 

 of the great systematists, Bentham, Smith, Hooker, Mueller and 

 others, would be discredited, and their ideas of what constitute a 

 species set at naught. Wliile recognising most of the species set up 

 by these great workers, one's aim must be to realise their system, 

 and to follow it in the light of later discoveries of forms unknown 

 to thein. At the same time it is inevitable that in all such revisions 

 of large genera, additional sj^ecies must be set up, consisting for 

 the greater part of certain forms which have in error or for con- 

 venience been placed as varieties, but which, if they liad been more- 

 fully considered by the jiioneers of systematic botany, would have 

 been given specific rank. Tlie author, recognising the fallacy of 

 multiplying species unnecessarily, has in doubtful cases allowed' 

 the varietal rank to stand. It is known that species may show varia- 

 tions according to climatic and soil conditions, and it may be con 

 ceded that the variations will be greatest or most likely to occur in 

 the first named, and least in the last-named of the following : — 

 Habit; colour of flower; size, shape and texture of leaf; size and 

 shape of stipules ; arrangement of flowers ; size and -shape of bracts, 

 bracteoles, calyx, ovary and seeds. So that in the case of Pultenaea 

 a variation of ovary, whether stipitate or sessile, villous or glab- 

 rous, shape of calyx, position and shape of bracteoles and bracts, 

 are of more importance in determining a species than the habit, 

 the shape aiul size of the leaves, or the colour or size of the corolla. 

 In other words, if ovary, calyx, bracteoles and bracts are similar, 

 a good deal of variation in habit or in leaf can be allowed wdthin 

 the limits of a species, wh'^reas if habit, leaves, corolla and stipules 

 are similar, a marked difference in the reproductive organs just 

 named will justify the i-etting up of a species. 



In the diagnostic drawings this has been kept in view, and they 

 include besides sketches showing shape and size of leaves and 

 stipules, drawings showing only the shape, size and relative posi- 

 tion of calyx, bracteoles and bracts. No attempt is made to show 

 the marginal curving of the leaf, or the nature of the indumentum. 

 Regarding the corolla, one has only to read in descriptions the- 

 oft-repeated " Standard twice as long as calyx," or " nearly 



11 



