• 



1889.] BOTANICAL GAZETTE. 135 



tions, and so on. If it be thought that natural selection somehow under- 

 lies all this, the reader is at liberty to substitute the phrase ; but, I must 

 confess, it conveys nothing definite to my mind, while the others 

 undoubtedly do. I do not wish the reader to suppose that my theory is 

 altogether (sic) in opposition to Mr. Darwin's ; for it must not be forgot- 

 ten that he himself laid great stress on the environment as a cause of 

 variability upon which, when once brought about, natural selection could 

 then act." I understand this to mean that he agrees with Mr. Darwin in 

 ascribing effects to environment, but not in regard to natural selection. 

 The fact that H. denies the advantage of cross-fertilization is sufficient 

 ground for saying that he repudiates natural selection as an explanation 

 of floral mechanisms. I regard natural selection, not as a cause of hyper- 

 trophies and atrophies, but as a cause of adaptations, the most important 

 characteristics of organs and organisms. However, I think of na ^ ur ^' 

 selection not so much as a cause as a controller of causes. No doubt H. 

 regards heredity as a cause. But heredity can only insure that a given 

 generation shall resemble its progenitors. Natural selection determines 

 who those progenitors shall be. In regard to other reviewers I quote 

 from J&urn. Bot. xxvi, 313 : " Professor Henslow, for example, is a welt- 

 known upholder of the principle of evolution ; but in the present work he 

 vehemently combats two of the theories which are most closely associated 

 with the great name of Darwin." H. regards a flower as a geologist 

 would regard a hill, i. e., as a resultant of all the forces which have been 

 brought to bear upon it. But organism* resist or avoid the direct ettecfc- 

 of their environment, being active in controlling their conditions, or m 

 adjusting themselves to them. In opposition to this view, 1 hold that 

 many highly specialized flowers, instead of developing to suit their 

 principal visitors, have contracted the parts in front of the receptacle, 

 excluding one set of visitors after another from the landing until tne 

 largest bees could only insert their tongues, e. g., Trifohum pratense ana 



Amphicarpsea monoica. ,. ., n „ r1 „ - t 



However, from the standpoint of pure Lamarckism, if we aciniu 

 that insect contact has a given effect, I hold that the theory will not ac- 

 count for the facts of floral structure. If insects leave the P e ^ nl " 

 altogether, H. claims that the whole perianth atrophies. If. they leaxet n e 

 perianth and light upon the stamens, the perianth atrophies .below and 

 hypertrophies above. In the case of Papilionaceae, etc., I ctaim tnatj w 

 first direct insect contact was equally absent both above and below and 

 had nothing to do with the reduction of one part or with the enlargement 

 of the other. What he calls atrophy-hypertrophy are the t » n g s J° De .?J" 

 plained, and thev can not furnish the explanation. The labellum of orch ids 

 is also against the theory. My view is that it was developed as a ^iimm 



by hawk-moths, which suck without touching the labellum, Ms «" P" 

 as well developed as in Orchis spectabilis, whose labellum forms a landin 

 for humblc-bPfis. What, doks insect contact have to do with the colore 



bracts of Euphorbia and Cornus florida, or with the neutral flowers ot 



Hydrangea and Helian thus annuus? »j,.„„/. ft f tho 



„ I "consider Verbascum a degradation rather than an advance ot the 

 Personales. I call the zvgomorphic type of Personalis ancien £ beoju*e 

 * is the type of the cohort. Campanula Americana « »*« J""W 

 of zygomorphv. The type of the genus Campanula is actmomorphic 

 But in the case of Verbascum the type not only of the great order ot 



