K. Pearson 
71 
Messrs Lewis and Embleton iindeformed members give rise only to undeformed 
offspring. It is not possible to say whetljer the deformity really is latent or not, 
there have been no cousin marriages to form even a rough test of latency. 
The absence of cousin marriages of any kind compels us to consider " lobster- 
claw " as a dominant character, and after the ancestress I. 2 of my pedigree, who 
may in Mendelian terminology possibly have been a homozygote, all the descendants 
must be looked upon as heteiozygotcs. On this assumption all normal individuals 
are recessive, and the fact that normal individuals from tainted stocks do not have 
deformed offspring meets with a ready explanation. 
It is this noteworthy point not ordy in the present family, but in Lewis and 
Embleton's family, in Drinkwater's Brachydactylous Family and in Nettleship's 
Night Blind Family, which is the main-stay in these cases of the Mendelian 
theory. It is perfectly true that it would flow from other hypotheses, for example 
from the assumption that the gamete was not pure, but that dominance flowed 
from a numerical preponderance of allogenic determinants*. Such a theory could 
only be tested by inbreeding. For the sake of science, therefore, if not for local 
well-being, it is desirable that marriages between the normal members of these 
stocks should occur, and this even in sufficient numbers to test whether there 
are really latent tendencies to the deformity in normal members of the stock. 
Shoidd this not prove to be the case, the non-marriage of the deformed members 
would be sufficient to check the spread of the deformity. Obviously a determi- 
nantal theory not based upon the pure gamete would further account for the 
occasional and rare appearance of the deformity in a child of normal parents. 
Mendelism must assert in a case like the present where the character must be 
dominant, that such an appearance is a mutation or sport. 
(3) Since the deformity must be looked upon as dominant in the Mendelian 
sense, it follows that on the average half the off'spring of a deformed parent ought 
to be deformed. We have data for such families in three generations, and the 
following results flow from them : 
N = Normal 
D = Deformed 
Families 
Totals 
I 
N. D. 
2 
N. D. 
S 
N. D. 
J^ 
N. D. 
5 
N. D. 
N. D. 
II. Generation 
III. Generation 
IV. Generation 
4 4 
.5 5 
0 3 
0 6 
0 1 
0 1 
2 5 
3 0 
4 4 
5 11 
5 10 
Totals 
14 2.5 
See the paper in this nuinlier of Biomctrika on p. 80. 
