91 



Quercus laurifolia, Fagus grandifolia, Pinus glabra, Ilex opaca, 

 Symplocos tinctoria, and Vaccininm elliottii. 



Now to the new locality. On February 12, 1927, Dr. R. M. 

 Harper and the writer were guided about 3^ mile into the fast- 

 ness of Johnson's Juniper Swamp by Mr. L. R. Carson of Bristol. 

 This swamp, already described and located as 8 miles south of 

 Bristol by Harper (3) is totally unlike the Flat Creek habitat. 

 The waterlogged, peaty substratum is highly acid (pH 4.2-4.5). 

 Fallen logs in all stages of decay criss-cross so that exploration 

 becomes very arduous. The luxuriant mats of a number of 

 species of Hepaticae and Musci attest to the very humid at- 

 mosphere of this hydrophytic forest. Among the trees and 

 shrubs, mostly evergreen, are: Magnolia virginiana, Clijtonia 

 monophylla, Chamaecyparis thyoides, Pinus taeda, P. elliottii, 

 Taxodium imbricarium, Nyssa biflora, Persea piibescens, and now 

 and then on the higher accumulated peat a young Magnolia 

 foetida. To our amazement in this contrasting habitat and 

 among such strange associates, we found the supposedly very 

 "selective" Taxus Floridana. 



This swamp projects the Florida yew's range at least ten miles 

 farther south, significant enough when it is realized that its 

 hitherto known distribution is within an area of only about 

 fifteen miles along the bluffs. Of special interest is the fact that 

 Taxus floridana parallels the peculiar distribution of its northern 

 relative Taxus canadensis in southern Ohio and according to 

 Tansley (5, pp. i66ff, 250) its European relative, Taxus baccata, 

 in England. The latter is abundant on the chalk uplands but 

 occurs also in acid moors. Markle (4) reports T. canadensis 

 as "one of the commonest undergrowth shrubs" in a cedar 

 swamp of Champaign county five miles south of Urbana. Very 

 commonly Taxus canadensis is found in shady mesophytic habi- 

 tats of ranyons and even in the dunes of Michigan. Such habi- 

 tats arc often circumneutral or alkaline in reaction. The plant 

 lists gixen for the Cedar Swamp by Dachnowski (i) and 

 Markle indicate circumneutral peat wnth local acid patches. 

 But as no hydrogen ion data are available of the peat where T. 

 canadensis grows comparison with the Florida juniper swamp 

 is impossible. The sharp distinction between the edaphic factors 

 of the juniper swamp and all other known habitats of the 

 Florida \-ew and the striking parallel shown by its other relatives 



