172 BOTANY OF THE LIVING PLANT 



before reaching it. Other more complex arrangements are expressed by 

 divergences of f (Rhododendron) (Fig. 129), and /■j, (Dracaena), etc. These 

 higher divergences go along with shorter internodes, and compact grouping 

 of the leaves ; while their overlapping is avoided by the spiral arrangement. 



Though we may recognise the biological advantage of such arrangements, 

 tlris provides no direct causal explanation. Various theories have been 

 propounded. The old spiral theory assumed an inherent tendency to spiral 

 organisation in plants, and, deductively, attempts were made to read spiral 

 construction into all shoots. Subsequently a theory of contact-pressures 

 was suggested, according to which the spirals resulted from mechanical 

 arrangement of the leaf-primordia upon the axis, comparable to those of 

 marbles in a flat frame. But though such pressures may in certain cases have 

 an effect upon the arrangement of the parts as they mature, they do not 

 explain the initial steps. For when the primordia first appear they are not 

 in contact with one another. In point of fact the exact position of the 

 primordia of leaves upon the axis, and their initial arrangement relative to 

 one another, can at present only be referred to inner causes as yet unlinown. 



Some plants develop with bilateral syinniefry, having anterior and posterior, 

 or right and left sides, which are alike. The flattened shoots of the Prickly 

 Pear, or of PhvUocactus. are examples, also certain Mosses (Fissidens). But 

 the headquarters of this type of s\-mmetry, which is uncommon in Flowering 

 Plants, is in the Marine Al.gae, and a good example is seen in the common 

 Bladder Wrack (Fitciis). 



DORSIVENTRALITY. 



While radial symmetry is thus the rule in upright shoots, those in 

 which the axis is oblique or horizontal usually diverge from the radial in 

 more or less degree. They show an obvious relation to the direction 

 of gravity, light, etc., developing differently on the sides directed 

 upwards and downwards. Such developments are styled dorsiventral. 

 Sometimes the effect appears onl}' in the later development of the 

 axes or appendages ; but in more pronounced cases the initial arrange- 

 ment of the leaves is itself dorsiventral. 



Examples of the former are seen in the lateral branches of manv 

 trees, and conspicuously in many Conifers, such as the Spruce. While 

 the main axis of the " Christmas Tree " is upright and radial, the 

 lateral branches appear flattened. This is due partly to the fact that the 

 lateral branches of higher order have been developed front buds seated 

 right and left on the main branches ; partly to the cur\'ature of the 

 leaves upwards. Nevertheless the construction of these dorsi\-entral 

 branches is essentially radial. If the terminal bud of the main stem 

 be removed, one or more of the lateral buds, wdiich would normally 

 grow into flattened lateral branches, will grow upwards and take its 

 place. With the vertical position, it regains its radial symmetiy. 

 Very much the same may be seen in the Sycamore, or the Horse 



