DISCUSSION. 523 



Mr. Baker offered some very interesting remarks on the distri- 

 bution of the various groups of Eucalypts. He supported the 

 Specific-Contour method, and believed that it would yield valuable 

 results as to the origin of the Eucalypt-flora. 



Mr. Tillyard, in reply, dealt briefly with each of the foregoing 

 criticisms. He pointed out that the question was not one of the 

 origin of species, but of the distribution of species ; and, therefore, 

 he did not consider that the question of mutation entered into the 

 discussion at all, even apart from the question of its intrinsic value 

 as a theory, which was not acceptable to everyone. In reply to 

 Mr. Fletcher, he exhibited a large coloured rainfall-map of Aus- 

 tralia, and showed how the ectogenic contours corresponded very 

 closely to the monsoonal isohyets, while the entogenic contours 

 came fairly close to the antarctic isohyets. Superabundance of 

 rainfall, as in Western Tasmania, might, however, be a factor 

 against distribution. The correlation between rainfall and distri- 

 bution was of a secondary nature ; climate did not determine what 

 groups should inhabit a region, but it did determine the form 

 of the contours of the groups that had reached the region. 

 In reply to Mr. Waterhouse, he showed that he had 

 made ample provision for showing lacunae, and instanced 

 the case of a subtropical group, which might spread all 

 round a central mountain-range, but fail to ascend to the highest 

 levels. He also showed, in reply to Mr. Mackinnon, how trans- 

 parencies could be used with the Specific Contour method, the con- 

 tour being drawn on the transparency, and the underlying map 

 showing the geology, rainfall- or temperature-variation of the 

 region, as the case might require. He agreed with Mr. Maiden's 

 criticism, but claimed that it was time that every collector should 

 realise the value of every single record, even of the commonest 

 species. His method would emphasise this fact. As regards in- 

 definiteness in species, it was one of the merits of the contour- 

 method, that both splitter and lumper would produce almost iden- 

 tically the same contour, since a species A, subdivided into geo- 

 graphical races, A 15 A 2 , A 3 , . . . . etc., would count only one, 

 except for the rare case in which two races might overlap. Reply- 



