432 TOWER, DARWINISM. 
chance positions or to adaptive qualifications possessed by the 
organisms concerned. This information was vital to the entire pro- 
blem of selection. There is abundant evidence that artificial selection 
does not carry modification of a particular stock much, if any, beyond 
the natural range of variability of that stock and that highly selected 
races commonly revert to the average condition of the parental 
group when the selective tension is removed or lessened. The 
collective experiences of this kind have lead experimental evo- 
lutionists to the opinion that Natural Selection does not act as an 
Originating force but as a conserving force. AS DE VRIES states, it 
acts as a sieve. It is increasingly evident that the action of artificial 
selection is mainly that of a sorting out, purification and isolation 
of divergent conditions existing naturally in the original group. 
This is a very different process from that supposed to operate in 
Natural Selection, where the sorting that takes place is based 
upon adaptive survival capacities present in the heterogeneous 
array of progeny that are produced in every generation. The 
initial problem, therefore, was to obtain by impersonal test more 
accurate information concerning this assumed difference in survival : 
capacity. A second problem, but one not considered in this paper, 
was to obtain evidence concerning the role of altered, hostile and 
new conditions relative to survival capacity and also with regard to 
what survived. That is, under varying conditions in experiment 
and in nature, does the process of elimination give any indication 
of actions similar to those which, by hypothesis, are the corner 
stone of Natural Selection. 
WITH PLANTS. 
The number of seeds produced by any of the higher plants is 
greatly in excess of the number that can reach maturity and many 
speculations exist as to how the elimination happens. A series of 
experiments was made between. 1905 and 1908 with the seeds of 
Solanum rostratum. In 1904, about two litres of the seeds of this 
plant were collected at Guadalupe near Mexico City. These were 
gathered at random from an area of about four square miles and, 
to avoid any chance of selective influence, entire plants were taken 
when the seeds were ripe but before the burs had opened or had 


