1925] PALMER, IS QUERCUS ARKANSANA A HYBRID? 197 



come in contact along this border line, although the former is not found 

 on the sandy soil where Q. arkansana is abundant, nor the latter on the 

 more sterile uplands. Quercus nigra belongs strictly to the lowlands 

 flora and is not found with Q. arkansana nor in its environment. 



The fine loamy sand (the Nackitoch sand of the U. S. soil survey map) 

 upon which Q. arkansana grows here is not entirely lacking in lime, as is 

 indicated by the presence of such more or less lime-loving plants as Quercus 

 Muhlenbergii, Ulmus fulva, Rhamnus caroliniana, Bumelia lanuginosa 

 and several species of Crataegus of the Molles group. Samples of this 

 soil tested by Dr. E. T. Wherry, of the Bureau of Soils U. S. Dept. of 

 Agriculture, were pronounced by him to be "minimacid" or almost 

 neutral. The calcium element is, of course, derived from the marly 

 fossiliferous layers. This soil is not extensively exposed but outcrops at 

 intermediate levels between the lowlands and the top of the plateau, and 

 in the ravines and along the courses of small streams traversing the latter. 

 The limits of the plants peculiar to this zone are therefore defined by the 

 comparatively sterile clays and gravels above and by the black stiff 

 soils of the river floodplains below. It should be remarked that within 

 these restricted sandy areas Quercus arkansana is nearly always present 

 and is often very abundant; indeed in some places it is not only the com- 

 monest Oak, but actually constitutes a large part of the forest growth. 

 I have seen and examined hundreds of specimens in such places and they 

 could doubtless be counted by thousands. I am not aware of any hybrid 

 that has succeeded in establishing itself in such numbers, or of becoming 

 dominant over areas from which both parent species are excluded. 



If instead of a hybrid we are, in the case of Quercus arkansana, here 

 dealing with a relic colony of a species nearing extinction and limited by 

 peculiar ecological conditions, we might reasonably expect to find similar 

 relics elsewhere, where these conditions are approximated within its 



probable former range. 



In traveling by automobile from Hot Springs to Texarkana, a few 

 months ago, we passed towards evening through an eroded area bordering 

 a small stream in the western part of Clark County, Arkansas. The general 

 aspect of the region and its flora bore a striking resemblance to certain 

 localities in the Hempstead County hills; my attention was first attracted 

 here by the occurrence in ravines of a large fossil oyster (Exogyra ponderosa) 

 and other fossils found at McNab. The lateness of the hour and an 

 approaching thunder storm prevented an exploration of the locality, but 

 a little farther down along a small sandy stream I had the satisfaction of 

 finding a few small specimens of Quercus arkansana. A few days later 

 on the same trip we passed near the locality in Alabama where Dr. Mohr, 

 and later Dr. Harper, collected the Oak which I believe to be identical 

 with the Arkansas species. Unfortunately I did not realize until too late 

 that we were so near the habitat of this interesting tree, and I did not 

 have the good luck to see it there. While the regions are so widely sepa- 



