268 RELATION OF PHYLLOTAXIS TO MECHANICAL LAWS. 



Animal Morphology, again, the conception of an axis and its 

 appendages, though admittedly a convention, is based on the 

 simple fact of observation that in higher animals the body is 

 constructed of one main axis and its limbs, or in Arthropoda, for 

 example, of one main axis and segmental appendages. The fact 

 that the . main axis exhibits dorsiventral symmetry leads to the 

 application of the same theoretical conception to the symmetry 

 of the appendages. JBut the term dorsiventral is an axial designa- 

 tion; the idea of an appendage being dorsiventral in itself is 

 wholly outside the range of the convention. Thus, while the 

 transference of the term dorsiventral as a metaphor to a special 

 case of shoot symmetry may prove helpful, its common application 

 to the case of a bifacial foliage leaf, which is typically an appen- 

 dage borne on a radially symmetrical axis, remains open to grave 

 objection. Similarly, the extension of the term to special floral 

 systems (inflorescences), and even to zygomorphic flowers (Goebel),* 

 may be of very questionable advantage ; since the application of a 

 metaphor to constructions in which it is not otherwise the custom 

 to discuss the dorsal and ventral surfaces, adds little to descriptive 

 power ; while there is no obvious criterion as to which surface 

 is to be regarded as the dorsal and which the ventral, and further 

 conventions require to be introduced. Sachs first pointed out the 

 fact that the so-called dorsiventrality of shoots as a type of 

 morphological construction was correlated with a difference in 

 physiological reaction to stimuli : the change of symmetry being 

 the outward and visible sign, or even the cause {Physiology, pp. 489 

 and 493), of a fundamental change in protoplasmic reaction, and 

 that dorsiventral shoots were plagiotropic. The special case of 

 isobilateral leaf-members here again presented an objection to the 

 use of the wider term bilateral ; while bifacial, on the other 

 hand, scarcely conveys the impression of a dorsiventral shoot 



* According to Goebel, the term dorsiventral imphes an upper and an under 

 side, while the conception of zygomorphy is based on a subordinate feature, the 

 existence of a right and left side ! Both phenomena are obviously equally the 

 consequence of eccentricity, the primary feature being the displacement of the 

 growth-centre, from which all secondary phenomena of " irregularity " or 

 asymmetry naturally follow. 



