354 A. B. RENDLE ON THE USE OF MICROSCOPICAL CHARACTERS 



In tlie highest groups, those comprising the Flowering or Seed 

 Plants, the development of an inflorescence, flower, fruit and seed, 

 has supplied macroscopic characters on which our system 

 of classification has been based. But there is an increasing 

 tendency to use minute characters as supplementary to or in 

 conjunction with the macroscopic; and in the case of limited 

 miaterial the study of minute characters becomes of the greatest 

 value. In the study of fossil plants, w^here the material consists 

 often of specimens showing only stem- structure, the worker has 

 to depend entirely on his microscope for their systematic elucida- 

 tion. I need not remind you of the large amount of recent work 

 in this branch of botany in which the diagnosis of specimens 

 ■depends entirely on internal structure as revealed by the micro- 

 scope, or of the great value of the results obtained as bearing on the 

 phylogeny of the higher plants. I do not propose to go into this 

 in detail. In modern flowering plants there is also considerable 

 -variety in the details of the stem- structure as revealed by the 

 •examination of a section under the microscope, and indications 

 -of the systematic position of the plant in question may be 

 obtained by this means; but there is room for a much more 

 •extensive and careful study on these lines. 



The earlier group of the Flowering Plants, the Gymnosperms, 

 by far the greater number of which are included in the Coniferae 

 Opines, firs, etc.) are distinguished from the later group, the 

 Angiosperms, which includes the great majority of modern trees, 

 shrubs and herbs, by a remarkable uniformity in the structure of 

 the secondary wood. This is due mainly to the absence, as seen 

 in cross-section, of the larger openings which represent the vessels, 

 which are a characteristic feature of the secondary wood of 

 Angiosperms. Among the conifers the wood of the yew shows 

 the greatest uniformity, as it consists of one kind of tissue only, 

 tracheides, the regular arrangement of which is unbroken by the 

 thinner-walled cells (parenchyma) which accompany the tracheides 

 in other conifers, and by the resin-canals which frequently occur 

 Yunning longitudinally between the elements, as in pines, firs, 

 larch and others. Here, there are minute characters which 

 enable us to determine even small fragments of wood as coni- 

 ferous, and among these to discriminate yew. The systematic 

 ■determination of plants by characters supplied by the wood is 

 a matter not merely of academic interest. Valuable information 



