~ 
L PERCY SLADEN TRUST EXPEDITION. 
resemblance. In reality the contrary is the case. When we examine further, we find 
a considerable number of sets of widely separated localities with the same or closely 
allied forms inhabiting them, and for this some explanatory cause must be sought out. 
Each of the larger groups of animals, and in the case of vertebrates each of the 
subgroups, found in a locality may point to a new geographical relationship for that 
locality. The explanation for this must be sought not only by a careful study of 
the structure and bionomics of many generations of many living forms, so as to obtain 
a clear understanding of the modes by which they have evolved, but also by examining 
the past types themselves. We are compelled to believe that all existing forms have 
arisen by some process of evolution from primitive parents living in a past age, and we 
require to know its laws in order to appreciate the reason why so many living organisms 
have such peculiar distributions. 
The question naturally arises, how many generations of a hypothetical species, genus, 
or family can remain the same in any two localities—such, for instance, as Ceylon and 
Madagascar? Is it capable of existing in the two places for an indefinite time in the 
same form? Can it in the two places evolve along parallel lines so as to remain 
identical after the lapse of ages? Or, what is equally important in distribution, 
can two organisms, of different genera, families, or groups, undergo such a change, 
by what is termed convergent evolution, that they must be placed by systematists in 
the same genus, family, or group? While we can give no definite answer to these 
questions, our view is that both parallelism and convergence do take, and have taken, 
place to a far greater extent than is usually supposed. We believe that these two 
phenomena go far to explain many of the similarities between the organisms of different 
lands. For instance, does the occurrence of -Zpyornis in Madagascar and Dinornis in 
New Zealand imply a close and recent connection between these two islands, or do they 
represent a case of convergent or parallel evolution from different ancestors or from 
some older and more widely distributed Ratite or Carinate? Or, again, are we to 
postulate the recent union of our three southern continents in order to account for the 
distribution of the ostrich, rhea, and emu? We believe not. On the other hand, 
we do find that the resemblances between the organisms of certain localities are too 
close to admit of any explanation founded on convergent evolution. In these cases the 
previous existence of a land-connection becomes a necessary corollary. In examining a 
single locality we may also find that we must demand different land-connections, in 
order to account for the distribution of different groups of organisms, though we must 
always bear in mind the fact that all groups of organisms did not appear at the same 
geological period. We next require the geological dates of these land-connections. 
In this respect isolated forms of once dominant and widespread groups are often of 
peculiar interest. And, lastly, we should study more particularly such deposits as form 
the land in order to trace out and localise the lines of connection. 
The study of animal distribution postulates the existence in past times of four great 
land-connections. ‘These occurred between Asia and North America, cid Behring 
Straits (in Pliocene times), between Australia and Asia across Malaysia (Cretaceous), 
between South America and Africa (at the commencement of the Tertiary), and between 
South Africa and India by Madagascar (possibly as late as the Eocene period). With 
