HEREDITARY TRANSMISSION OF COLOUR IN CROSS-BREEDING. 211 
variety, and so the fundamental principle of Mendel’s law breaks down, 
and with it his whole theory. Mendel’s experiments were made entirely 
with plants, but I am convinced that similar experiments in the breeding 
of animals will lead to the same results. These experiments have already 
been made upon rats, mice, and snails, and the investigators have come 
to the conclusion therefrom that Mendel’s law is applicable also to the 
crossing and pairing of these animals, a view which, in my opinion, is not 
justified by the facts. 
_ Albino rats and mice, when crossed with the original type, produce 
exclusively hybrids in entire correspondence with the wild type, and there- 
fore are in agreement with Mendel’s law. The reason of this is that the 
possession of pigment is dominant over the want of it. The first genera- 
tion from the cross of the wild type with the albino being entirely like the 
wild type, one would assume that when members of that generation were 
paired together their progeny would follow exactly the characteristics 
of the parents. But this was not the case; they were albinos and 
common coloured individuals in the proportion of one to three—a result 
in agreement with Mendel’s law. This, however, seems to me to be only 
a proof that the force of reversion can be stronger than that of heredity. 
To put Mendel’s theory to the test, one should use varieties in which 
pigment is present, and which have been bred pure for a considerable time, 
and not albinos ; and then one would find that different results would be 
obtained. Take, for instance, two contemporaneous varieties such as 
Nature sometimes presents us with, e.g. albino and cinnamon-coloured 
starlings. Here we have an equal power of hereditary transmission and 
an equal latent tendency in each to revert to the original type. By 
pairing such varieties together the two equal hereditary forces counteract 
each other, leaving the field open to the latent characteristic, with the 
consequence that the offspring take the appearance of common starlings. 
Cinnamon canaries have been bred pure for a shorter time than yellows. 
The pairing of cinnamons with yellows will not therefore result in 
perfect reversion, but the offspring will be variegated greens, i.e. green 
mixed with yellow. From this pairing we get cinnamon-marked young 
(always hens), but only when the father is a cinnamon, the reason of 
which is that the preponderance of the hereditary force is on the father’s 
side. An albino starling paired with one which had some white feathers 
produced young entirely like common starlings. In verifying Mendel’s 
law, it is necessary therefore in the first place to consider for how long 
the varieties to be experimented upon have been bred pure, in order to 
form an accurate idea of the respective forces of heredity on the one hand, 
and of atavism, or the tendency to revert to the wild type, on the other. 
In modern varieties the force of heredity is weak in comparison with 
the tendency to reversion, and consequently in breeding we find that the 
latter force holds the chief sway. 
Last year I received from Herr Bocker, the son of the well-known 
ornithologist at Wetzlar, two young wild cock canaries caught in Teneriffe. 
I crossed them this year (1906) with several different varieties of the 
domestic canary, viz. cinnamons, green variegated, and yellows. The 
young from the cross with the cinnamon were all in their nest feather 
entirely resembling the wild type, as I expected would be the case. Those 
2 
OZ 
