Plants in their Relation to Others . 
497 
a shifting of the ecological barriers, or the erection of such. Lastly, a fourth 
great reservation is expressed in what is said about barriers. The very 
great majority of species, for example, will be absolutely cut off from 
spreading by meeting a sea of any serious width, and the comparatively few 
species that may at times be able to cross may be easily appreciated by 
a study of Guppy’s monumental labours on the subject. A broad river may 
act as a barrier to many, and probably will usually tend to delay spread, at 
any rate. The same is true of a high range of mountains, especially if they 
reach the snows, but here the barrier also as a rule becomes partly ecological, 
inasmuch as the climates on the two sides usually differ, to say nothing of 
the different climate at high levels. Other ecological boundaries than 
climatic changes are usually, in my opinion, too narrow to stop completely 
the passage of species, though they may deflect, distort, or delay it ; but 
the wording of the rule is framed so as to cover all possibilities. And 
finally, in modern times, as every one knows, man has done more than any of 
nature’s agencies in the distribution of species about the globe, partly by 
intentional or accidental transport, partly by the great clearances, involving 
change of conditions, that he has made. 
In studying the local distribution of species in New Zealand, where it 
was more exactly worked out than in Ceylon, I worked with the aid of 
a prediction that I made from a consideration of the bearings of Age and 
Area upon the subject (16, p. 442 ), that the number of endemic species in 
any genus would rise gradually to a maximum at or near the point or 
region where the genus entered New Zealand in the first place. The 
diagram here reproduced shows the way in which this occurs. A single 
wide (w) is supposed to enter New Zealand at the centre, and to follow 
