152 



R. H. RICHENS 



I 



Anglian and Kentish populations of U. carpinifolia. It is obviously premature, therefore, 

 to speculate as to its early history, but it may prove useful to list the following facts, 

 which will have to be taken into consideration when adequate distributional data have 

 been obtained. In the first place, place names based on elm or wice are rare in the Midlands, 

 and agricultural settlement appears to have been sparse till well after Roman times (cf. 

 Darby, 1936). It therefore seems probable that elms were relatively infrequent in this 

 region in early times as compared with East Anglia. However, the absence of U. procera 

 from East Anglia and eastern Kent suggests that this species did not migrate into England 

 from the east. U. procera does not appear to be known as an indigenous tree on the 

 continent. U. procera var. australis (Henry) Rehder from southern Europe is very remote 

 from the typical variety and should almost certainly be excluded from the species. U. 

 procera occurs in Spain, and samples from Aranjuez and Mondonedo in the Cambridge 

 Herbarium are indistinguishable from English material. It seems probable that the 

 Spanish trees were, as stated by Evelyn (1679), introduced into Spain from this country. 



It would therefore appear that the origin of U. procera must be sought somewhere 

 in the English Midlands. Possibly it arose as a single local clone and became widely 

 spread on account of such desirable characters as straight bole and late defoliation. Its 

 relation to U. carpinifolia is problematical. Transitional forms, notably lib and the 

 glabrous members of Ila, occur, and, of these, samples 86b and 95 of lib are from far 

 beyond the eastern limit of U. procera. It may be, then, that U. procera arose from some 

 such type of U. carpinifolia, the scabrous character deriving either from a persistence of 

 the juvenile condition or from introgression from U. glabra. On the other hand, it is 

 quite likely that some intermediates between U. procera and U. carpinifolia are later 

 hybrids. 



The data presented here are not adequate to discuss the history of U. plotii. 

 Summary 



A technique for analysing the taxonomic variation of East Anglian elms is described, 

 based on systematic sampling of the boundary hedges of ancient parishes. Five quantita- 

 tive and one qualitative leaf characters were determined for each sample, and the degree 

 of affinity between samples was ascertained by a graphical method based on an application 

 of the t test. As a result of this approach, eleven taxonomic groups could be distinguished 

 within the genus, together with a number of subgroups. The three major species, U. 

 glabra, U. carpinifolia and U. procera, were fairly satisfactorily distinguishable, but the 

 groups here recognised both cut across and transcend the segregates of U. carpinifolia 

 recognised in the new Flora of the British Isles. It seems probable that U. carpinifolia 

 is composed of a large number of distinguishable clones, and it is doubtful whether any 

 useful purpose is served by applying Latin names to assemblages of these. A method 

 of characterising the clones and clonal groups numerically is described. 



U. glabra occurs sparsely in the area studied; U. carpinifolia is widespread; while 

 U. procera hardly occurs east of a line running from Peterborough to Waltham. Many 

 of the recognised groups and subgroups of U. carpinifolia have characteristic geographical 

 distributions. 



The evidence suggests that U. carpinifolia was introduced from the east as a forage 

 plant, sometime in the neolithic. Bronze or Iron ages. U. procera appears to have arisen 

 somewhere in the Midlands, whence it is spreading eastward. 



It is considered likely that an intensive study of the distribution and relationships 

 of the individual clones of U. carpinifolia would throw an interesting light on the history 

 of agricultural settlement in England. 



