118 
IOWA. ACADEMY OP SCIENCE 
and lessening numbers, the amount of seed-production may fall below 
the minimum necessary for the perpetuation of the species. 
Dr. W. 0. Focke, of Bremen, first identified the Oenotheras near Liv- 
erpool, England, as belonging to 0. Lamar chiana. Charles Bailey, in 
a more recent account of this vegetation (1907a, b) concludes that their 
introduction probably came froA sweepings of grain-ships and docks and 
in grain for poultry from America. It seems more probable, however, 
that they originated as escapes from English gardens at a very early 
date. 
In concluding this examination of historical records it should be said 
that I have endeavoured to present the documents and other evidence 
from which my inferences and conclusions liave been drawn, in such a 
Avay that the reader who examines the evidence can judge for himself 
of the justice of the conclusions deduced. I have not been biased in favor 
of any theory of the origin and history of 0. Lamar cldana. I have 
shown that a form very closely resembling 0. JLamarckiana, except in 
certain rosette characters, was originally wild in Virginia, but it has 
never seemed to me that the question whether 0. Lamar ckiana has been 
hybridized or not is of great significance in connection with the inter- 
pretation of the mutation phenomena in these open pollinated forms, 
which must have experienced crossing in nature before their introduc- 
tion into gardens. It is, however, a matter of much importance to de- 
termine that a form at least closely similar to 0. Lamar ckiana was the 
first Oenothera introduced into cultivation. 
In nature, the individuals of all open-pollinated species are hybrids, 
in the sense that many more or less diverse elements have contribuated 
to their ancestry. In making cultures from wild open-pollinated forms 
I have been impressed with the variability of the first generation in 
cultivation in comparison with forms which have been selfed for a 
number of generations. It is of course necessary, in breeding, to select 
certain individuals for later generations, and if these are self -pollinated 
the resulting races are sure to show increasing uniformity in later gen- 
erations. If space for cultures permitted that every individual could 
thus become the starting point for a race, it would be found that each 
such individual would originate a race showing slight peculiarities. In 
the last analysis, as Jennings^ has .remarked, the differences between 
races would be found to go down to the limits of observation and meas- 
urement. The occasional appearance of mutants, or marked departures 
from the type which breed true, is of course another matter. 
iJennings, H. S. Experimental evidence on the effectiveness of selection. 
Amer. Nat. J,!, : 136-145. 1910. 
