448 PART IV. CLASSIFICATION. 



members of a whorl, or, more rarely, to a multiplication of the 

 whorls or of their members. If, however, the organs which are 

 absent, but which should typically be present, be indicated in the 

 empirical diagram by dots, it becomes a theoretical diagram. In 

 this way it is possible to arrive at general types on which large 

 numbers of flowers are constructed. Fig. 264, for instance, is the 

 empirical diagram of the flower of the Lily, and it is at the same 

 time the type on which the flower of Grasses (Fig. 265) is con- 

 structed in which certain members are suppressed. 



lu constructing a floral diagram the position of the main axis 

 should be indicated by a dot placed above the diagram : the bract, 

 which would of course be exactly opposite to it, may or may not 

 be indicated : the side of the flower toward the main axis is said 

 to be posterior, and that toward the subtending bract, anterior. 

 A plane which passes through the flower and also through the 

 main stem and the median line of the bract is termed the 'median 



FIG. 264. Floral Diagram Fra. 265. Floral Diagram FIG. 266. Floral Dia- 



of a Lily. of a Grass. gram of a Crucifer ; the 



median stamens are 

 duplicated. 



plane or section of the flower : the plane which cuts the median 

 plane at right angles is the lateral plane or section : and the 

 plane which bisects the angles made by the intersection of the 

 median and lateral planes is the diagonal plane or section : any 

 plane other than these is said to be oblique. By means of these 

 conceptions the position of the parts of a flower may be accurately 

 indicated : thus, in describing the flower of the Cruciferse (Fig. 

 266), the two external sepals lie in the median plane ; the two 

 inner sepals, the two outer stamens, and the two carpels, in the 

 lateral plane ; whilst the petals and the four inner stamens lie in 

 the diagonal planes. 



The number and the relations of the different parts of the flower 

 may be indicated not by diagrams only, but also by formulae in 

 which, as in the diagrams, for the sake of clearness, all the 



