aa lt t . yi 
January 13, a 
NATURE 
41 

=A sistent criterion by which comparables may be re- 
cognised is absolutely essential to the application of 
the method. Nevertheless no information offered 
reduces the difficulty materially. We are told that 
only forms in the same “circle of affinity” are to 
deliberately. 
—— 
be taken—a definition which is plainly left vague 
How this is to be construed we are 
never precisely told. The species to be compared 
must be more or less alike in their modes or at least 
in their facilities of dispersal—a property we have 
commonly no means of estimating in any trustworthy 
or quantitative way. Unless I have misunderstood 
the chain of reasoning, its validity is severely strained 
at this point. 
The author is shy of special illustrative examples 
and they need not be essential to an argument 
dealing solely with general propositions, but in a 
chapter contributed by Prof. Small we are provided 
with an illustration on the largest scale. There we 
are given to understand that the natural order 
Composite is a “circle of affinity” to which the 
method of Age and Area can be properly applied. 
If a group so polymorphic and heterogeneous as the 
Composite constitute a “circle of affinity,” the 
members of which can be compared for these purposes, 
where are we to stop? The tribes of Composite are 
arranged in a genealogical tree upon which the pre- 
sumed point of origin of each is marked, and we are 
told that the order of evolution as given on the tree, 
which has been constructed from anatomical data, 
agrees substantially with the numerical estimate of 
the areas occupied by each tribe. Needless to say, 
numerous eminent botanists have arranged the tribes 
in almost as many other ways, probably with equal 
propriety. These speculative genealogical trees, once 
fashionable, are, I had supposed, discredited. All 
that they can attempt is the display of a logical order 
of interrelationship based on the modifications of the 
special set of organs selected as a criterion; for the 
Composite this order will differ with each set of organs 
chosen. In support of Prof. Small’s arrangement he 
gives an imposing tabulation of the geological levels 
in which each tribe is believed to have arisen. Not 
until the text of Prof. Small’s previous papers is 
consulted does a reader discover that this tabulation 
is almost wholly conjectural. In a well-written and 
judicious chapter by Mrs. Reid, who discusses what 
paleobotany can produce in support of Age and 
Area, we find no such confident pronouncements. The 
inclusion of the chapter on the Composites reflects 
more credit on Dr. Willis’s candour than on his 
scientific judgment. The propositions made in the 
name of the theory there stand forth with a neglect 
of caution which Dr. Willis himself seldom exhibits. 
NO. 2776, VOL. 111] 
For the reasons given, the theory of Age and Area, 
except in so far as it is truistical, is as yet of doubtful 
value, and unless amended to meet the difficulties 
specified it cannot be applied with any confidence. I 
suspect that certain predictions respecting the flora 
of the islands near New Zealand, which, though made 
in advance, as we are frequently reminded, were 
fulfilled, did not involve any feat of which common 
sense would have been incapable. 
Dr. Willis is a great advocate of the theory of 
mutation in its crudest form. The speculation now 
presented to us as Age and Area is a development 
of an idea which came to him when he reflected on 
the fact that in Ceylon several endemic species are 
limited to small areas, though sometimes associated 
with related species of wider distribution. The theory 
of mutation of de Vries appeared at about the same 
time, and Dr. Willis asked himself whether the wonder- 
ful ‘‘ mutations ” which had been reported in Gnothera 
might not exemplify the process by which the Ceylonese 
endemics had been begotten by the “ wides,” as he 
calls them. Endemics had previously been held to be 
largely relics. In the new light they become “in 
the vast majority’ novelties, about to spread with 
the lapse of time in widening circles. On any theory 
of evolution endemics must be in part novelties and 
in part relics ; but why, apart from the theory of Age 
and Area, we should believe that endemics are in such 
great majority novelties I do not clearly understand, 
for though we know little of origins we are certain 
that myriads of species have become extinct. It is 
surely contrary to all expectation that the process of 
extinction should be in general so rapid, and the final 
endemic phase so short that the number of species 
in that final stage of existence should be insignificant. 
The supposition implies the optimistic but embarrass- 
ing corollary that a species, once established, is in no 
great danger of extermination unless some catastrophic 
or lethal change occur in the conditions of life. 
Cupressus macrocarpa is admitted to be in danger 
because, as we are told, the Monterey peninsula is 
drying up. This is used as the stock illustration of 
the mode in which authentic extinction should occur. 
As it serves three times in this capacity, bearing 
perhaps an undue burden in the argument, we may 
infer that examples of extinction through predicable 
secular change are not plentiful. Unless, indeed, the 
change can be traced directly or indirectly to human 
action, the cause whether of gain or loss of territory 
is apt to be a mere matter of surmise, for though 
losses are so familiar we must not forget that there 
are also mysterious gains—even in our own area,’ 
Who shall say what gave Capros aper its chance ? 
A doubtful British species in the time of Couch, it 
BI 
