1912} — Bush,— Geographic Origin of Crataegus viridis 83 
Ozark Colony, has anything to do with the great body of Southern 
Virides, which are, as stated above, coastal plants occurring at low 
altitudes, is highly improbable, as it nowhere joins this southern body 
of trees and the individuals composing it show a decided preference 
for high rocky barrens and plains. 
In the first part of this paper I have alluded to the well-known fact 
of plant association and have named about twenty-five trees and 
shrubs of the two hundred and odd southern species that occur in 
southeastern Missouri and which are found with the true Crataegus 
viridis of Linnaeus in the Southern States, and I now call attention 
to the fact that not one of these characteristically southern plants is 
to be found associated with the Western Ozark Colony, but that the 
Crataegus is here associated with a very different set of plants, of 
which Quercus Marylandica, Hicoria villosa, Cornus florida, Viburnum 
rufidulum, Diospyros Virginiana, Fraxinus Americana, Sassafras 
variifolium and Ulmus fulva are the most representative, all plants 
noted for their preference for high, dry and rocky ground. 
Some three years ago I visited Hannibal, Missouri, which is some 
hundred and twenty-five miles north of Saint Louis, the supposed 
northern limit of Crataegus viridis L., and was agreeably surprised to 
find on the high hills and mounds there several species or forms of 
Virides, quite a colony, and later I saw many acres of trees in the river 
bottom opposite Hannibal, in Illinois, near the town named Shepherd. 
Having quite recently visited Quincy, Illinois, twenty miles north 
of Hannibal, Missouri, and one hundred and forty-five miles north 
of Saint Louis, I was further surprised at seeing the hills and mounds 
around that place covered with Virides, many thousands of individuals, 
some of the trees being the largest specimens I had ever seen. Many 
of these trees are on the high hills at Keokuk, Iowa, and the forms no 
doubt occur in all eastern Iowa, and perhaps as far north as southern 
Wisconsin. There are so many specimens at this colony, which I 
shall designate as the Upper Mississippi Colony, perhaps one hun- 
dred times as many as in all southeastern Missouri and northeastern 
Arkansas, that it is undoubtedly the center of greatest abundance, 
and the Virides found at Saint Louis, Missouri, instead of coming up 
the Mississippi Valley, have really straggled down it. 
This Upper Mississippi Colony is so far from the great southern 
body of Virides, having not a single southern plant associated with it, 
that there can be scarcely any doubt that the species composing it are 
