HABITS OF THE HERRING GULL——STRONG. 505 
breeding places. Unfortunately we know nothing concerning the 
relationships of individuals in one colony to those of another. We 
have no information concerning whether gulls breeding at one of the 
Strawberry Islands, for instance, have interbred in recent years 
with Gravel Island gulls. Furthermore, we have no data, as yet, 
concerning the existence of definable pure lines in the morphological 
characters of herring gulls. 
The other type of variability in behavior is not associated with 
pure line inheritance. It represents simple chance variations from 
the average type of behavior which are to be expected just as we 
find variations in morphological characters. 
In the section of this paper which deals with the general behavior 
of juvenal gulls I mentioned variability in the behavior of some 
newly hatched gulls. A single bird in one nest showed no terror, 
whereas two nestlings of essentially the same age in another nest 
were in great distress from fright over the presence of human 
intruders. I can think of no reason for considering that either the 
quiet bird or the frantic pair were more intelligent or that either 
form of behavior was adaptive. Nor have we reason to believe that 
it was a case of pure line differentiation. It seems quite possible 
that chance variations in the metabolic states of the birds or possibly 
in their nervous organization were responsible for the difference in 
reaction to our presence. It is conceivable that the reactions would 
have been either similar or reversed if we had approached the nests 
a few hours later. 
A large amount of variation has been noted in the choice of mate- 
rials used in constructing the nest. Evidently the herring gull 
uses what is available with a preference for finer and softer mate- 
rials. It seems probable to me that such nest building, which is 
evidently mostly instinctive or stereotyped, is not absolutely without 
the elements of intelligence. There must be adaptation of special 
materials which may be found, to use and location. Though the 
general form, size, and location of the nest are characters of the 
species, the variations which fit the nest to its special location, for 
instance, are no more stereotyped than various acts of man which 
are called intelligent. 
As a consequence of persistent nest robbing, gulls at certain 
breeding places have been reported as taking on a tree-nesting habit. 
I know of no evidence worth considering for believing that the 
recent ancestors of such birds were tree nesters, and we have every 
reason for considering the inherited choice of location for most her- 
ring gulls as on the ground. 
We have already seen in this paper that even the structure of the 
nest may be modified in adaptation to the location in a tree. More 
skill is shown in weaving the nest, according to reports, so that it 
