138 CORRELATED VARIATION. 



lC3 l,> \ 



countenances this idea; but with the umbellifera3 it u 

 no means, as Dr. Hooker informs me, the species with the 

 densest heads which most frequently differ in their inner 

 and outer flowers. It might have been thought that the 

 development of the ray-petals, by drawing nourishment 

 from the reproductive organs causes their abortion; but 

 this can hardly be the sole cause, for in some composita? 

 the seeds of the outer and inner florets difl'er, without any 

 difference in the corolla. Possibly these several differ- 

 ences may be connected with the different flow of nutri- 

 ment toward the central and external flowers. We know, 

 at least, that with irregular flowers those nearest to the 

 axis are most subject to peloria, that is to become abnormally 

 symmetrical. I may add, as an instance of this fact, and 

 as a striking case of correlation, that in many pelargoniums 

 the two upx)er petals in the central flower of the truss often 

 lose their patches of darker color; and when this occurb, 

 the adherent nectary is quite aborted, the central flower 

 thus becoming peloric or regular. When the color is 

 absent from only one of the two upper petals, the nectary 

 is not quite aborted but is much shortened. 



With respect to the development of the corolla, Spren- 

 geFs idea that the ray-florets serve to attract insects, whose 

 agency is highly advantageous, or necessary for the fertili- 

 zation of tliese plants, is highly probable; and if so, nat- 

 ural selection may have come into play. But with respect 

 to the seeds, it seems impossible that their differences in 

 shape, which are not always correlated with any difference 

 in the corolla, can be in any way beneficial; yet in the um- 

 belliferge these differences are of such apparent importance 

 — the seeds being sometimes orthospermous in the exterior 

 flowers and co^lospermous in the central flowers — that the 

 elder De Candolle founded his main divisions in the order 

 on such characters. Hence modifications of structure, 

 viewed by systematists as of high value, may be wholly 

 due to the laws of variation and correlation, without 

 being, as far as we can judge, of the slightest service to the 

 species. 



We may often falsely attribute to correlated variation 

 structures which are common to whole groups of species, 

 and which in truth are simply due to inhericance; for an 

 ancient progenitor may have acquired through natural 



