\4^ Life amf Lcttfrx of Fro tin's (Jaltnit 



hypothesis; he was beginning to l>e doubtful of tlie transmission of amuiied 

 cnaractei-s, but he hud yet to take the 8t«p of a full and complete denial. 

 Thus aft^er admitting that a few charactere are iic(juired and transmitted by 

 dogs', he writes: 



"With iiu'ii tlioy nri' fowcr still: iriilcc<i it is (titUcult to p>imt out any one tt) thi' noccptjvnce 

 of which srtiiie objection nmy not b«" ofliTttl. lioth M. de Candolle and Ur Carponter liavo spoken 

 of the i<H<»cy and other forms of norvoua disortlers which beyond all doubt afflict the children 

 of tlrunkanlis. Here thon appears an instJince iMised on thousands of observations at lunatic 

 asylums and t-lsowliert' in which an aoijuirtvl habit of drunkenness, which ruins the will and 

 nerves of the parent, appears to l)e transmitted hereditarily to the child. For my own part I 

 hesitate in drawing this conclusion, because there is a simpler rcasou." (p. 351.) 



The 'simpler reason' given by Galton is the toxic action of fluids satu- 

 rated with alcohol on the reproductive elements of both se.xes, and in the 

 case of the mother on the unborn child. Perhaps the 'simplest reason' goes 

 beyond Gal ton's! Now-a-days only asylum officers return 'Alcoholism' as a 

 causal category of insanity. Those who have studied the pedigrees of the 

 alcoholic know that nervous disorders are relatively as frequent in the as- 

 cendants as in the descendants; and the true explanation was given by 

 De Quincey. when he said; "My friends thought drink drove me to insanity, 

 but I knew that insanity drove me to drink." If we start with the drunkard, 

 we find only too frequently mentally defective offspring. But if we start 

 with complete pedigrees, we find stocks, whose members tend to nervous 

 disorders, alcoholism, insanity and suicide. However, tialton's conclusion 

 was a right one: 



"We must not rely on the above-mentioned facts [alcoholism in parent and nervous 

 disorder in o&pring] as evidence of a once acquired habit being hereditarily traosmitted." 



The passage is, however, of great historical interest as it shows how 

 Galton's theory of the distinction between the reproductive and the body 

 cells was forcing him steadily to the conclusion that to modify heredity you 

 must modify the germ cells, or at least an early stage of the embryo. Men 

 like De Candolle, Spencer and Carpenter, nay, even Darwin himself, believed 

 in the transmission of acquired characters; Galton, as far as I am aware, was 

 the first to doubt it, and to assert that in the embryo of the oftspring there is 

 nothing represented that was not in the embryos of its parents. However crude 

 now-a-days we may consider that statement to be, it really covers the con- 

 tinuity of the germ-plasm and the denial of the transmission of acquired 

 characters. On the main j)oint of difference lietween De Candolle and Galton, 

 the present writer also finds Galton's criticism, as abstracted above, wholly 

 justified. De Candolle's work, as he himself tells us, had been planned 

 when he was a young man. His main idea seems to have then been to trace 

 scientific fichievement as arising from environmental causes — educiitional, 



' I am not sure that Galton wius mil influenced by the aversion of Huggins' dog 'Kepler' to 

 butchers' shops, an av<;rsion shared by his relatives, and which some may assume to have 

 originally been 'ac<|uire<l.' There is a letter from Galton t<j Huggins on the snhject of 'Kepler' 

 dated Feb. 15, 187.3, i.e- just before the ForlnighUy paper. See our p. C6, sewjnd footnote. 



