This leads me to remark that the sentence at the bottom of page [80] is not applicable to my views (143/2. See Falconer at the bottom of page 80: it is the old difficulty--how can variability co-exist with persistence of type? In our copy of the letter the passage is given as occurring on page 60, a slip of the pen for page 80.), though quite applicable to those who attribute modification to the direct action of the conditions of life. An elephant might be more individually variable than any known quadruped (from the effects of the conditions of life or other innate unknown causes), but if these variations did not aid the animal in better resisting all hostile influences, and therefore making it increase in numbers, there would be no tendency to the preservation and accumulation of such variations--i.e. to the formation of a new race. As the proboscidean group seems to be from utterly unknown causes a failing group in many parts of the world, I should not have anticipated the formation of new races.
You make important remarks versus Natural Selection, and you will perhaps be surprised that I do to a large extent agree with you. I could show you many passages, written as strongly as I could in the "Origin," declaring that Natural Selection can do nothing without previous variability; and I have tried to put equally strongly that variability is governed by many laws, mostly quite unknown. My title deceives people, and I wish I had made it rather different. Your phyllotaxis (143/3. Falconer, page 80: "The law of Phyllotaxis...is nearly as constant in its manifestation as any of the physical laws connected with the material world.") will serve as example, for I quite agree that the spiral arrangement of a certain number of whorls of leaves (however that may have primordially arisen, and whether quite as invariable as you state), governs the limits of variability, and therefore governs what Natural Selection can do. Let me explain how it arose that I laid so much stress on Natural Selection, and I still think justly. I came to think from geographical distribution, etc., etc., that species probably change; but for years I was stopped dead by my utter incapability of seeing how every part of each creature (a woodpecker or swallow, for instance) had become adapted to its conditions of life. This seemed to me, and does still seem, the problem to solve; and I think Natural Selection solves it, as artificial selection solves the adaptation of domestic races for man's use. But I suspect that you mean something further,--that there is some unknown law of evolution by which species necessarily change; and if this be so, I cannot agree. This, however, is too large a question even for so unreasonably long a letter as this. Nevertheless, just to explain by mere valueless conjectures how I imagine the teeth of your elephants change, I should look at the change as indirectly resulting from changes in the form of the jaws, or from the development of tusks, or in the case of the primigenius even from correlation with the woolly covering; in all cases Natural Selection checking the variation. If, indeed, an elephant would succeed better by feeding on some new kinds of food, then any variation of any kind in the teeth which favoured their grinding power would be preserved. Now, I can fancy you holding up your hands and crying out what bosh! To return to your concluding sentence: far from being surprised, I look at it as absolutely certain that very much in the "Origin" will be proved rubbish; but I expect and hope that the framework will stand. (143/4. Falconer, page 80: "He [Darwin] has laid the foundations of a great edifice: but he need not be surprised if, in the progress of erection, the superstructure is altered by his successors...")
I had hoped to have called on you on Monday evening, but was quite knocked up. I saw Lyell yesterday morning. He was very curious about your views, and as I had to write to him this morning I could not help telling him a few words on your views.