GENERIC CRITERIA IN FLOWERING PLANTS 



319 



It is interesting to note that Hubbard started on this investigation because Lepturus, 

 as then deUmited, contained a mixture of temperate and lowland tropical species, so that 

 phytogeography in this instance gave a clue to an unsatisfactory taxonomic situation. 



The Gramineae, on account of their economic importance as well as their intrinsic 

 interest, have been studied in more detail that most families of flowering plants. For the 

 reasons just indicated, taxonomists have been driven to search for generic criteria which 

 can be used in addition to the customary reproductive ones. This search has been remark- 

 ably successful and, since there is no reason to believe that the family is unique in this 

 respect, it seems worth mentioning the kind of characters which have been found useful. 

 There is in fact every reason to suppose that vegetative characters can be profitably used, 

 with due caution, in most if not all families. 



Duval Jouve and later Prat were the first to make a thorough study of the systematic 

 value of epidermal characters and other anatomical features in the Gramineae, and so 

 to provide a number of new generic and specific criteria which have since been used to 

 good effect. The shape and orientation of silica c-ells, the type of hair or papilla, the shape 

 of the stomata, and the general pattern of the epidermis all help in characterising not only 

 genera but tribes and subfamilies. The distribution of green tissues in the leaves and 

 the occurrence of simple or compound starch grains in the endosperm are also characters 

 of major taxonomic importance. In groups where the vegetative as well as the reproductive 

 parts are extremely reduced and all conform to a single pattern in their gross morphology 

 microscopic characters of the epidermis and internal tissues may prove of great value. 



Salicornia, which is said to be cosmopolitan on coastal and inland saline soils, may 

 well be like one of the form genera of palaeobotanists. A preliminary examination of the 

 epidermis of S. dolichostachya, S. perennis and the Mediterranean species, S. macrostachya, 

 shows differences which may be significant. 



The value of cytology in helping to define the limits of genera and subgenera and 

 sometimes, as in Ranunculaceae (Gregory, 1941), in grouping genera into tribes, is well 

 known. The value of the basic number as a generic criterion varies considerably in different 

 families. In the Cruciferae, Manton (1932) has shown that cytology on the whole confirms 

 the classification of traditional taxonomy, a conclusion which must greatly strengthen 

 our faith in the reality of genera, at least in this family. She was also able to provide 

 guidance about the separation or union of six pairs of genera about which varied views 

 had been held by taxonomists and to point out a number of other genera in which systematic 

 revision or further cytological work was desirable. 



A point of particular interest is that the taxonomic changes made by Schulz in his 

 revision of the family in Das Pflanzenreich are supported by the cytological evidence. The 

 changes that Schulz made were mainly the result of his use of the number and orientation 

 of the nectaries and the presence or absence of branched hairs as generic criteria, and it 

 is striking that criteria as diverse as these comparatively trivial ones and the basic chromo- 

 some numbers should be in such good agreement. 



In the Chenopodiaceae on the other hand the basic number 9 appears to be of universal 

 occurrence, except for Spinacia with x = 6, and is therefore of no help in settling generic 

 limits. In the Gramineae there is a variety of basic numbers which help mainly in the 

 delimitation of tribes or subfamilies. Chromosome numbers have however confirmed the 

 correctness of the separation of Puccinellia from Glyceria. the basic number being 7 in 

 the former and 10 in the latter, and have led to a more thorough morphological examination 

 of these genera with the result that they are now placed in different tribes. The occurrence 

 of the basic number 10 in Catabrosa, another genus until now classified in the tribe 

 Festuceae has stimulated a re-investigation of its morphology which, as might be expected, 

 has revealed a close similarity to Glyceria. 



