so 
had no opportunity of observing it, or how its relation to the food- 
plant differs from that of T. unicolorana. 
These observations led me to query how far our general idea that 
lepidopterous larvffi are always simply injurious to their foodplants 
was a true one, and how far it was merely a confession of ignorance. 
When one organism lives upon another, whether they be both animals 
or plants, or one of each, a certain settled relative abundance of the 
two usually exist, which we often refer to as the “ balance of nature.” 
It will perhaps be simplest to call two such related organisms host and 
guest, though we might call the predaceous individual, a parasite, a 
carnivore, or various other names, but in fact the relations are so 
various that only very general terms, such perhaps as host and guest, 
will be applicable in all cases, a caterpillar eating a leaf, a lion devouring 
a bullock, a bacillus causing disease, a man reaping wheat, &c., are 
difficult to bring into one simple definition. If such a relation be 
suddenly established, no “balance of nature” has been set up, and it 
is very possible that the guest may very rapidly exterminate the host. 
Man has much to answer for in this way, both in his own proper 
person, and also by introducing European animals into limited 
localities such as oceanic islands. If time, however, can be obtained 
for relative adjustment, then the host will probably be able to preserve 
his existence. This condition is established in practically all the cases 
we are considering. There are probably many ways of establishing 
such a balance, but these occur to me as being perhaps the most 
important. 
A very common one arises out of the fact that the guest in one 
relation is the host in another, and this would settle all difficulties if 
there always were the lesser fleas, and if these went on according to 
the rhyme ad infinitum. This view of the matter reminds us that to 
say host and guest, as if we had only two individuals to deal with, 
gives us no true picture of the complicated relationship of great 
multitudes of organisms, all more or less interdependent, sometimes as 
hosts sometimes as guests. Luckily it is not necessary to further 
consider this aspect of the question before us. The second method of 
adjustment between the host and guest, is by the host developing 
various protective devices, and the third is by making the attack useful 
instead of injurious, and even inviting it. Of the protective devices, 
cryptic arrangements, a hard or spiny covering and others do not, I 
think, probably come into the items we may profitably consider. The 
one that is most related to lepidopterous larvae, is the modification by 
which a plant becomes more or less unpalatable to its guest. This is, 
I think, a device of much greater frequency than we commonly 
suppose, and is in a way related to our subject. Few larva hunters 
can have failed to notice in many cases that a healthy vigorous plant 
is not nearly so usually attacked by larv* as one that is more or less 
out of sorts, a plant not thriving from some inherent individual 
weakness, or because it has been injured, or very frequently a more or 
less isolated plant which is battling against conditions to which its 
fellows have succumbed or have not dared to face. This being so it is 
clear that a healthy normal plant has some quality by which it escapes 
the attacks of the guest. And we may further conclude that the guest 
benefits not the pjant it attacks, but the species to which it belongs* 
by clearing off weaklings and so leaving the race stronger, just as th$ 
