Walter Frank Raphael Weldon. 1800—1906. 
39 
was that his confidence in the generality of the Mendelian hypotheses was much 
shaken. He found that Mendel's views were not consonant with the results 
formulated in a number of papers he had been led to abstract, and that the 
definite categories used by some Mendelian writers did not correspond to really 
well-defined classes in the characters themselves. It was a certain looseness of 
logic, a want of clear definition and scale, an absence of any insight into how far 
the numbers reached really prove what they are stated to prove, that moved 
Weldon when he came to deal with Mendelian work. And his attitude has been 
largely justified. The simplicity of Mendel's Mendelism has been gradually replaced 
by a complexity as great as that of any description hitherto suggested of hereditary 
relationships. This complexity allows of far greater elasticity in the deduction of 
statistical ratios, but the man in the street can no longer express a judgment 
upon whether the theory really accounts for the facts, and the actual statistical 
testing of the numbers obtained, as well as the logical development of the theory, 
will soon be feasible only to mathematical power of a high oi'der. The old categories 
are, as Weldon indicated, being found insufficient, narrower classifications are being 
taken, and irregular dominance, imperfect recessiveness, the correlation of attributes, 
the latency of ancestral characters, and more complex determinantal theories are 
becoming the order of the day. If Weldon's papers " Mendel's Laws of Alternative 
Inheritance in Peas " (28), " On the Ambiguity of Mendel's Categories " (29), and 
" Mr Bateson's Revisions of Mendel's Theory of Heredity " (30), be read with a due 
regard to the dates of their appearance, it will be seen that they served, and that 
they continue to serve, a veiy useful purpose : they enforce the need for more 
cautious statement, for more careful classification, and for greater acquaintance 
with the nature of the inferences which are logically, i.e. mathematically, justifiable 
on the basis of given statistical data. The need will become the more urgent if the 
complexity of Mendelian formulae increases at the present rate. 
To those who accept the biometric standpoint, that in the main evolution has 
not taken place by leaps, but by continuous selection of the favourable variation 
from the distribution of the offspring round the ancestrally fixed type, each 
selection modifying pro rata that type, there must be a manifest want in 
Mendelian theories of inheritance. Reproduction from this standpoint can only 
shake the kaleidoscope of existing alternatives ; it can bring nothing new into the 
field. To complete a Mendelian theory we must apparently associate it for the 
purposes of evolution with some hypothesis of " mutations." The chief upholder of 
such an hypothesis has been de Vries, and Weldon's article on " Professor de Vries 
on the Origin of Species " (32) was the outcome of his consideration of this matter. 
During the ye.'irs 1902 to 1903 an elaborate attempt was made to grow the 
numerous sub-races of JDraba vertia, with the idea that they might throw light on 
mutations. The project failed, largely owing to difficulties in the artificial cultivation 
of some of the species. But for a time all other interests paled before Draha verna. 
"Where are you going at Easter? Stone wall country is very good, and if you find a place 
with delightful old stone villages and pretty chui'ches, Draba verna will be there ! Come into 
