272 RELATION OF PHYLLOTAXIS TO MECHANICAL LAWS. 



whatever to special external agencies such as insect visitors, hut 

 simply as a structural feature. Natural selection may increase this 

 tendency, but it was always present in the growing-point of the 

 shoot. By stating the converse for " terminal flowers " on centric 

 shoots, one begins to arrive at the truth underlying such gener- 

 alisations as that of De CandoUe that no zygomorphic flower 

 could be terminal, notwithstanding the fact that every flower 

 terminates its own axis.* 



It is also a point of interest that the more general occurrence 

 of eccentric growth in flower shoots, in that it involves the sub- 

 jective appearance of a certain bilateral symmetry, is responsible 

 for the introduction of conventional conceptions of "median 

 planes," etc., which, though transferable to symmetrical phyllo- 

 taxis systems, have absolutely no existence in the more generalised 

 case of asymmetrical construction. 



Because the genetic-spiral of Schimper would not explain the 

 phyllotaxis of a dorsiventral shoot, Sachs was willing to throw 

 over the entire spiral theory, although he was fully aware of the 

 fact that cases of markedly " dorsiventral " phyllotaxis are rela- 

 tively few, occurring usually in growth constructions specialised 

 in other directions; and the elucidation of such "dorsiventral" 

 constructions remains a test case for any suggested hypothesis 

 which claims to interpret the facts of shoot construction and leaf 

 arrangement. The present standpoint is sufficiently clear ; there 

 can be little doubt that the habitual use of the term " dorsiven- 

 tral " introduces a fallacious standpoint ; one incautiously argues 

 that the distinction of a dorsal and ventral surface as in an animal 

 must be a very remarkable specialisation, and so undoubtedly its 

 consequences appear in the adult shoot, though the separation of 

 the two surfaces is not apparent at the growing-point where the 

 construction commences. Once the term " dorsiventrality " is 

 eliminated, or only retained as a harmless metaphor, and it begins 

 to be obvious that all that the phenomena include is an unequal 

 distribution of growth in different directions, the subject becomes 

 much clearer from the point of view of strict morphology ; while 

 the physiological conception of plagiotropism, and whether this is 

 * Cf. Goebel, Organography of Plants, Eng. trans., p. 133. 



