38 VARIATION. 



winter, and the autumn-form in summer, and the difference 

 in result of the development may be explained by assuming 

 that some of the different developmental processes responsible 

 for pigmentation have mutually different temperature-coeffi- 

 cients. We must remember, that the striking difference be- 

 tween the two forms is a result of the fact that no pupae de- 

 velop in intermediate temperatures. The experiments of Dorf- 

 meister have shown how intermediate types are produced in 

 intermediate temperatures. 



There are developmental processes whose result varies con- 

 tinuously in direct relation with an important factor, but there 

 are also processes which do not admit of such a direct response. 

 The mechanism of the response may be such an one, that at a 

 critical point in the relation between the action of one factor 

 and the others, the organism begins to react differently. Some 

 phenoniena, certain processes may by their very nature pro- 

 hibit continuous variation. For instance, meristic phenomena. 

 We do not know many definite facts about the causes for the 

 position of leaves on a stem, but we can see how the very na- 

 ture of the obscure mechanism makes it impossible for a stem 

 to bear its leaves in any way intermediate between two def- 

 inite ones. Thus it appears impossible for a plant which can 

 either bear its leaves in opposite pairs, or in whorls of three, to 

 have a stalk in an intermediate condition. 



In Dahlia arborea, for instance, the better nourished stalks 

 bear their leaves in whorls of three, whereas the weakerstalks 

 and the smaller side-branches have their leaves in opposite 

 pairs. The cause of the difference, the thing which decides 

 that a branch will be of one or of the other kind, is here clearly 

 given by the non-inherited developmental factors. It is easy to 

 grow individuals of this Dahlia which are so ill-developed, that 

 not one stalk is produced with whorls of three leaves. And on 

 the other hand it is possible by appropiate methods which 

 cause the production of thick branches (pruning) to make the 

 lilac and some Fuchsias produce branches with the leaves in 

 whorls of three. 



