27 
the following year the species would eat up the whole country, are 
followed by scarcity of the species for some years, or even its apparent 
absence in the district Avhere it was so abundant. Some accurate 
observations and records on these two points are much wanted. Such 
occurrences are mentioned in conversation and forgotten, and no 
record is made unless the species be a very rare one, or some extreme 
aberration is taken, and even then precisely what interests us just now 
is apt to be left .out. 
Assuming then, that our data are all right, though very probably 
they are not entirely so, it is interesting to speculate how they came 
about, and what is the connection between them. 
Lepidoptera lay large numbers of eggs, but still very different 
numbers in different species, ranging from say 50 to 1500 (A. betularia 
lays about the latter number and if. pyrina perhaps nearer 2000) by 
one $ . Taking 200 as a fair average to work with, and allowing half 
to the S and half to the $ , we find, of course, that if all the eggs 
laid became imagines, the species would increase a hundredfold in 
each generation, say annually. This means, that taking one brood 
and one season with another, 99 out of every 100 perish, not necessarily 
before reaching the imago state, but before continuing the species. 
This means a very severe process of pruning away every individual 
that does not come up to a certain standard of perfection in a great 
number of very various qualities. The agencies that inflict this loss 
are no doubt equally various, some taking a large, some a small toll, 
but amongst them destroying 99 per cent, of individuals. 
Now in a given area, a flower pot, a garden, a field, or a whole 
country, our hypothetical species has—let us say—1000 individuals ; 
next y r ear it would have 100,000 but for the fact that 99,000 are 
destroyed. Now in one particular year we find instead of the usual 
1000 there are actually 100,000 or 1,000,000 individuals. How could 
this happen ? it means that for two or three years certain of the 
pruning agencies were much less active than usual, and instead of 
taking 99 out of every 100, they took only, say, 90 ; if this happened 
for two successive years, then our normal 1000 would be 100,000, or 
if for three, it would be 1,000,000. 
Of course, if the 99, instead of becoming 90, became some smaller 
number, then this increase would be still more rapid. It is probable 
that each pruning agency takes not exactly so many, but a certain 
percentage, so that the failure of one would be made up to some 
extent by the greater activity of another. Therefore we may assume 
that such phenomenal increases as we are contemplating are probably 
due to the failure simultaneously of two or more destructive forces. 
It is sometimes stated, or assumed, that such destruction is not 
selective, but fortuitous, and that the individuals that perish are as 
good as those that survive. This can hardly be so; the insect 
destroyed by an ichneumon may be as well protected against birds as 
one that the ichneumon leaves, and one that is killed by frost may 
be as well protected against birds or ichneumons as any survivors, but 
it is surely certain that whatever danger proves fatal to an individual, 
is a danger against which that individual is less protected than its 
fellows. 
It follows then, that when, say, ten individuals survive instead of 
xix. 
