2 Bews. — Some General Principles of Plant Distribution as 
among the sedimentary formations which form most of the framework 
of the land. Hooker, however (27), was much more inclined than 
either Darwin or Wallace to find a way out of difficulties by supposing 
great geographical changes. In explaining the connexions between 
the isolated southern floras, he says, ‘ The most conspicuous characters 
that extra-tropical Africa presents in common with Australia are 
the abundance of the species of the following orders : Proteaceae, Com- 
positae, Irideae, Haemodoraceae, Buettneriaceae, PolygaleUe, Restiaceae, 
Ericaceae, Epacrideae, Decandrous Papilionaceae,Rutaceae, Thymelaeaceae, 
Santalaceae, Anthospermous Rubiaceae. . . . With regard to the natural 
orders enumerated, their genera are almost unexceptionally different in the 
two countries. . . . The many bonds of affinity between the three floras — 
the Australian, Antarctic, and South African — indicate that these may all 
have been members of one great vegetation, which may once have covered 
as large a southern area as the European does a northern. To what 
portion of the globe the maximum development of this southern flora 
is to be assigned it is vain at present to speculate, but the geographical 
changes that have resulted in its dismemberment into isolated groups over 
the Southern Ocean must have been great indeed.’ 
In contrast with this, I quote one more paragraph from Thiselton- 
Dyer (20, p. 303 ):' Darwin objected to “ continental extensions ” on geological 
grounds, but he also objected to Lyell that they do not “ account for all the 
phenomena of distribution on islands, such, for example, as the absence of 
Acacias and Banksias in New Zealand ”. He agreed with de Candolle 
that “ it is poor work putting together the merely possible means of distri- 
bution ”. But he also agreed that they were the only practicable door of 
escape from multiple origins. If they would not work, then “ every one who 
believes in single centres will have to admit continental extensions ”, and 
that he regarded as a mere counsel of despair, “ to make continents as easily 
as a cook makes pancakes 
Coming now to Thiselton-Dyer’s own views and those of Guppy, who 
has contributed probably more 'than any one else to the detailed study of 
plant migration in recent times (24-26), we find that these authors are agreed 
that a theory of southward migration is the key to the interpretation of the 
geographical distribution of plants. Even with regard to the Glossopteris 
flora, Thiselton-Dyer says, ‘ I confess it would not surprise me if fresh dis- 
coveries in the distribution of the Glossopteris flora were to point to the 
possibility of its also having migrated southwards from a centre of origin in 
the northern hemisphere ’. Guppy’s earlier views were similar to those of 
Wallace. He sought to explain all the phenomena of distribution by 
migration. Later he was inclined to hesitate. As Thiselton-Dyer says, 
‘ Guppy’s heart failed him when he had to deal with the isolated case of 
Agathis , which alone seemed inexplicable by known means of transport 
