42 
ment brought about by the moisture, as shown by Dr. Chapman and 
myself. The action of smoke has been rapid, because it is practically 
a new phase of environment. It also, as an unusual factor, has acted 
on the constitution of the larva, and produced change and variation. 
But it also, like moisture, acts in a secondary way, and also in the 
direction of darkening objects, but its area is more generally limited to 
tree-trunks, fences, walls, &c., in the comparatively near vicinity of its 
production. Indirectly, therefore, the force of “ natural selection ” has 
again cleared out the paler, and protected the darker forms. Hence we 
see, that the artificial, so to speak, environment, has with “ natural 
selection ” produced a much more rapid, and sometimes, especially when 
aided by rain, a more complete change in a shorter period, than has the 
natural environment in an incomparably longer time. 
Mr. Robson further writes:—“In our fens and bogs, our wet moors 
and mosses, we would surely find some evidence in support of this theory, 
if it were true.” Has Mr. Robson so little knowledge of the insects 
from these localities, or so little material in his collection, as to suppose 
that such localities do not give numbers of insects which support this 
theory ? If he will show me a collection of insects made on any of our 
large bogs, I will guarantee to show him any number of species proving 
my point. He complains that I “ argue from a special case, and assume 
that the reply has a general application.” I assume nothing of the kind. 
I certainly argued that there was a large class of pale coloured species 
belonging to very different groups of lepidoptera, which had a very 
similar facies, such facies undoubtedly being entirely governed by 
“natural selection.” Mr. Robson jumps to the conclusion that Macro- 
gaster arundinis, Loelia ccenosa, Lithosia muscerda, Collix sparsata, the 
genus Chilo, have very different habits from the Leucanidce, and yet all 
are “ of ochreous tints, which certainly is the prevailing hue in the species 
that inhabit our swamps and morasses.” I am quite certain that Mr. 
Robson is here arguing on a misapprehension of facts relating to 
the habits of M. arundinis, L. ccenosa, L. muscerda, and the genus 
Chilo, and that he has no conception of the range of variation in 
these species. I believe he has never seen one of these species in a state 
of nature. I have seen them all but L. ccenosa, so I can safely leave the 
entomological public to judge of the value of our respective opinions. 
Granting that a certain percentage of the Fen (not bog) species are pale 
in tint, what I wrote is still applicable to these as a group. Mr. Robson 
appears to fail to grasp what I have written. His failure to see the fact 
that “ natural selection ” is, in such districts and under the conditions of 
existence of these species, acting in direct opposition to the production 
of melanic forms is unfortunate, but I see no immediate remedy. To say 
that I “ have lost the thread of my argument,” because I have got rather 
out of the line he could understand, is rather amusing. I import another 
factor, viz. “ smoke,” says Mr. Robson, in dealing with the modification 
shown by marsh-frequenting species, near large towns. Certainly! 
I am sorry Mr. Robson cannot compound the three probabilities here 
referred to (moisture, smoke, and natural selection), but it would ap¬ 
parently be waste of time to explain the matter further. 
Mr. Robson unwittingly proves in his reference to L. pollens, my 
compounded notion with regard to this species. (1) L. pollens, when 
it inhabits very wet places, should have a tendency to be dark; (2) 
L. pollens, in marshy places, rests on grasses and reeds, therefore “ natu- 
