Species

The Species Problem Arises 199

had cultivated and maintained pedigrees.^49 He was of the view that these were
straight mutations forming a single new elementary species at once. In a way, given
the way alloploids occur, he was right, but his idea led fairly directly to the later
views of Goldschmidt, who, like de Vries, felt that species always, or almost always,
arose in sudden saltative leaps.
In the lecture discussing the evening primrose, de Vries further denes the marks
of an elementary species:

Elementary species differ from their nearest allies by progressive changes, that is by

the acquisition of some new character. The derivative species has one unit more than

the parent.^50

This meant that if the new elementary species were crossed with its parental
elementary species, the progeny would be incomplete for that character (he clearly
means Mendelian factor) and would therefore be unnatural.^51 Hence backcrossing
would not occur in the wild or under cultivation.^52 Elementary species thus do not
exhibit subvarieties, for they are the “real type.”^53 De Vries’ concept was fundamen-
tally anti-Darwinian, in the sense that he rejected the idea of there being continu-
ous variation within species on which selection could act in such a way as to form
new species. In fact, he argued for his theory on the grounds that the length of time
required for evolution would be noticeably shorter on his account. At that time, Lord
Kelvin’s arguments against Darwinian evolution—that reasoning from the rate of
cooling of the earth, evolution would need to have happened in tens, not hundreds or
thousands of millions of years—were still current. Rayleigh’s discovery of radioac-
tivity as a source of planetary heat was not announced until about this time (1906),
and it did not immediately lter through to the wider scientic community.^54
Prior to Bateson’s and Poulton’s essays, there was no species problem as such, but
only a species question. The latter is concerned primarily with the origins of species,
how they come to be. The problem arises when we have accounts of species forma-
tion, whether by selection or something else, that do not involve immediate saltation
from one to another or creatio de novo. It is the problem of dening what rank it
might be that species achieve when they become species, and this sets the agenda
for Dobzhansky’s 1935 paper, and the remainder of the twentieth-century debates.
There were a few Darwinians in the period before the Synthesis who discussed
the concept of species. For example, in 1934, J. Arthur Thompson dened species
extensively in the classical terms but with an emphasis on the role played by selec-
tion.^55 Thompson denes it in the context of the human species and races of man, and
gives four criteria: non-trivial difference, true breeding and constancy of characters,

(^49) de Vries 1912, 17, lecture IX.
(^50) de Vries 1912, 253.
(^51) Op. cit., 254.
(^52) Op. cit., 527.
(^53) Op. cit., 127.
(^54) Bowler 1989, 2 0 7.
(^55) Thompson 1934, vol. 2, 1333f.