CHANGES OF SHAPE OTHERWISE CAUSED. 1G5 



another. But any inequality in the rate of growth-on the 

 different sides of the shoot, will destroy this straightness in 

 the lines of growth. If the greatest and least rates of mole- 

 cular increase happens to be on opposite sides, the shoot must 

 assume a curve of single curvature ; but in every other case 

 of unequal molecular increase, a curve of double curvature 

 will result. Now it is a corollary from the instability of the 

 homogeneous, that the rates of growth on all sides of a shoot 

 can never be exactly alike ; and it is to be also inferred from 

 the same general law, that the greatest and least rates of growth 

 will not occur on exactly opposite sides of the shoot, at the 

 same time that equal rates of growth are preserved by the 

 two other sides. Hence, there must almost inevitably arise 

 more or less of twist ; and the appendages of the internodes 

 will so be prevented from occurring perpendicularly one OTer 

 another. 



A deviation of this kind, necessarily initiated by physical 

 causes in conformity with the general laws of evolution, is 

 likely to be made regular and decided by natural selection. 

 For under ordinary circumstances, a plant will profit by hav- 

 ing its axis so twisted as to bring the appended leaves into 

 positions that prevent them from shading one another. And, 

 manifestly, modifications in the forms, sizes, and insertions 

 of the leaves, may, under the same agency, lead to adapted 

 modifications of the twist. We must therefore ascribe this 

 common characteristic of phasnogams, primarily to local differ- 

 ences of nutrition, and secondarily to survival of the fittest. 



It is proper to add that there are some Monocotyledons, 

 as Urania speciosa, in which this character does not occur. 

 What conditions of existence they are that here hold this 

 natural tendency in check, it is not easy to see.* 



* The Natural History Review for July, 1865, contained an article on the doc- 

 trine of morphological composition set forth in the foregoing Chaps. I. to III. In 

 this article, which unites exposition and criticism in a way that is unhappily not 

 common with reviewers, it is suggested that the spiral structure may he caused 

 by natural selection. When this article appeared, the foregoing five pages were 

 Bfanding over in type, as surplus from No. 14, issued in June, 1865. 



