35 [Vol. xl. 
' The: evidence is strongly in favour of the staining 
having taken place prior to the scratching, otherwise the 
stain would have penetrated the scratches ; those on the side 
are in the direction of the large end, towards which they 
terminate and become finer and less deep, and were probably 
caused by the birds shifting the egg to the small end. 
My theory is that what really happens is that, during 
oviposition and while the egg is passing through the cloaca, 
there exists a potential vacuum behind the egg, and this, 
combined with the atmospheric pressure in front, causes the 
partly protruded egg to become stationary or fixed as though 
in a vice as already suggested; or it may well be caused by 
the contraction of the muscles of the uterine and cloacal 
walls, and in order to assist the egg in its further progress 
the bird either drags that part of the egg already exuded on 
the surface of the nest, or the claws are used to further 
assist in the process of exudation. Normally the large end 
leaves the cloaca first, hence the scratches on the large end, 
and the possibility of the small end leaving the cloaca first 
would account for the scratches on the small ends which 
are rare. Honey-Buzzard’s eggs, being almost round, sug- 
gest, too, that they may leave the cloaca at either end, 
consequently this would account for their being scratched at 
either end. 
I am unable to offer any suggestion as to why Caper- 
caillie and Honey-Buzzard eggs are specially subject to 
these scratches ; it is certainly of rare occurrence in the eggs 
of other species. 
The strongest evidence in favour of my theory is based 
on the following points :— 
1. The eggs of T. urogallus are nearly always scratched 
at the large end. 
2. The position and direction of the scratches on a whole 
clutch of eight eggs is practically the same. 
3. The direction of the scratches is mostly longitudinal. 
4, They are finest at the termination of the scratch, 
which varies in direction ‘at the extreme large end, 
where pressure is naturally lessened. 
62 
