PRESIDENTIAL ADDRESS. 57 
attempted differential descriptions. of species, a simple ex- 
elusive and inclusive definition was considered sufficient. A 
short deseription, which included all the necessary individu- 
als and excluded all others, was almost all that was required. 
Now much more complete morphological data—and the ful- 
ler the better—are demanded. As yet, however, there is 
little tendency to include as specific traits in the descrip- 
tions other facts than those of morphology. Baker and 
Smith, before this Society, have shown the importance of 
the presence or relative absence of certain chemical bodies 
in the separation of our species of Eucalyptus, Melaleuca, 
Callitris, etc. Mental characteristics, based as they must be 
on cellular, molecular, or perhaps chemical differences, 
~ should also, I think, receive full consideration. The com- 
mon Australian Willy-wagtail or Shepherd’s Companion 
(Rhipidura motacilloides), has universally the habit of sett- 
ling on the backs of sheep, cattle and horses whilst these are 
feeding, and of circling round them from this perch to catch 
the flies disturbed during grazing. The bird is a friendly 
one, and little afraid of man. Its cousins, Rh. albiscapa 
and Fh. rufifrons, have not, as far as I am aware, the habit 
of settling on the backs of grazing animals, though they 
also are relatively tame. This peculiar habit of the Shep- 
herd’s Companion is a true specific trait, equally entitled to 
inclusion in the description as the morphological details of 
colouring or the character of the pigmentation of the eggs. 
I would go further and say that if I came across a race of 
Shepherd’s Companions which, in the presence of frequent 
opportunities, universally failed to manifest this habit, I 
would be induced to scan the members of this group closely 
to see whether such a difference in behaviour was not accom- 
panied by some other departure, perhaps morphological, 
perhaps in the structure of the nest, which would entitle 
one to distinguish the race as a distinct variety. It may 
here be noted that closely related species, so close as to be at 
