Miscellanea 
435 
was made in the more lengthy manner of drawing 1508 times a card out of a pack containing 
20 cards, numbered in duplicate 0, 1, 2, 3, 4, 5, 6, 7, 8, 9. The card was returned, and the pack 
shuffled and the 1508 draws recorded. .Each successive four gave a number which was treated 
exactly as in the previous four experiments. The results were : 
Theory 
Experiment 
Cancer 
pi 
313 
31(5 
315 
p 2 ... 
29 
26 
20 
P3 ••■ 
2 
3 
6 
Pi ... 
0 
0 
1 
x 2 
approx. 6 
approx. 258 
It will be seen that the ^ 2 of experiment is essentially within the bounds of probability 
{P=% say), while the cancer-data give a value of x 1 which is 43 times as large. For the 
previous set of experiments the ^ 2 for the third is about 25, and for the other three about 
18 — larger values than occur in the shuffling experiment — but of a quite different order to 
the cancer value. Many arguments — all houses not being of same age, some houses pulled 
down, and so forth — might be used to account for the multiple cancer cases, but I think 
these data certainly justify a fuller inquiry into the whole question. They provide some 
evidence, of more value than mere impression, that the hypothesis of " cancer " houses is worthy 
of a fuller consideration. 
V. Hybridisation of Canaries. 
Note on the Communication by C. L. W. Noorduyn to the Members of the Genetics 
Congress held in Paris from September 18 — 23, 1911. 
By A. RUDOLF GALLOWAY, M.B., CM., M.A. 
At the Genetics Congress held at Paris in September, 1911, Mr C. L. W. Noorduyn, of 
Groningen, quoting Biometrika, Vol. VII., Nos. 1 and 2, gave from my paper on " Canary breed- 
ing" the "Table of Canary Hybrids bred since 1891, arranged to show Plumage Colour." 
This table indicates the rarity of lightly variegated and " clear " hybrids, and Mr Noorduyn 
regretted that I did not specify the " five or six hybrids " out of the total of 526 that were not 
bred from <J wild bird x $ canary, but from the reverse mating of $ canary x $ wild bird. 
He believed the proportion of lightly variegated hybrids to be greater from $ canary x ? wild 
bird than from the reciprocal mating. 
In view of the prominence so kindly given to this table by Mr Noorduyn, it is necessary for 
me to say that the information desired is really included in the paper, but refers to five or six 
exceptional matings, not to five or six individual birds as stated by Mr Noorduyn. 
55—2 
