Recent Advances in the Study of Heredity. 283 
of majus is dominant to that of laciniatum ; i.e., here entirety is 
dominant to dissectedness. In itself it is no explanation to say 
that it is obviously the normal which is dominant in the two cases, 
because if Urtica pilulifera were very rare and had been discovered 
after U. Dodartii it would be the variety, and Dodartii the normal 
But it is possible that the true explanation may underlie this one. 
It certainly is not probable that the two cases of dissectedness are 
manifestation of analogous germinal factors. 
There are, however, further cases in which the result of the 
union of the two characters in a pair seems definitely to contradict 
the theory that the dominant member of the pair always consists in 
the presence of something and the recessive in its absence. In wheat 
the resistance to Rust, a disease caused by the attacks of Puccima 
glumarum has been proved by Biffen to be a recessive character 
and susceptibility to be dominant. In resistant plants the growth 
of the rust-hyphae is checked after they have entered the stomata ; 
and it is probable that the resistance is due to an anti-toxin. So 
that the presence of the anti-toxin is the recessive character and 
its absence is the dominant. To make the theory fit this case we 
we have to suppose that the dominant factor is one which prevents 
the formation of the anti-toxin. 
Again the absence of glands on the leaves in Matthiola inccnta 
is dominant to their presence in M. sinuate. 
But the fact that the theory meets with difficulties in its appli¬ 
cation to these cases should, so far from giving cause for mis¬ 
givings as to its truth, tend to set us at ease on this score. If all 
the facts are consistent with a theory, we may be pretty sure that 
it does not even contain the germ of truth. No theory can 
touch (or be in contact with) all the facts ; so that if a theory 
touches some, it cannot touch others, and there will be hitches and 
difficulties somewhere. If there are to be no hitches, the theory 
must touch none of the facts. Instances of such theories suggest 
themselves readily to the mind. 
