288 Mr Bate^^oii's Revisioiis qf MendeF^ Theori/ of Heredity 



Mendel con.sidercd that tho unit eharacter coiild oiily be detennined b}- ex- 

 periment; oii the one haud groups of apparently distinct charactcrs might be 

 iriherited as one unit (for example seed-coat colour and flower colour in ceitain peas) 

 whilc on the other hand au apparently single eharacter, such as the colour of some 

 flowers, niight be duc to the siniultaneous presence of several unit characters, each 

 capable of being inherited separately frora the others. 



Mr Bateson's Revisioiis of Mendel*. 



1. The nature of the gametes in hi/h-ids. — Mendel says that the gametes, 

 produccd by a hybrid Zygote of Constitution Aa, are not themselves hybiiil, but 

 are of two kiuds, both equally numerous; one kind contaius only elements Ä, the 

 other containing only elements a. Now when similar gametes, produced in this 

 way, unite to form a Zygote, what will iiappcn ? The individuals will behave like 

 pure-bred individuals of the race represented by their gametes. This is all that 

 Mendel ever asserts of them ; when he says that their characters remain constant, 

 it is clear that he means a coustaucy of the same order as that observed witliin the 

 limits of a pure-bred species becausc he treats the species as themselves constant, 

 and so neglects individual Variation within tlic limits of a single race. Mr Bateson 

 hovvever, alone among modern naturalists, roundly declares that " the pure 

 dominant and pui'c recessive members of each generation are not merely like but 

 identical with the pure parents" (II. p. 12); so that one of the effects attributed 

 by him to cross-breeding is the disappearance of those differences which normally 

 exist between members of the same species. He asserts that by cross-breeding a 

 group of individuals niay be produced, including one ancestor and some of its 

 grandchildren, great grandchildren and more remote posterity (" members of each 

 generation ") which are " not merely like but identical with " each other. This 

 Statement is supported by no scrap of evidence ; it is flatly contradicted by 

 experiments which Mr Bateson himself adduces as "following Meudel's law 

 with considerable accuracy" (o.g., the experiments with Datitra, (II. pp. 21 — 32)); 

 nevertheless, it is madc the basis of the imposing dogma that auy attempt to 

 express the characters of the hybrids described in terms of all their various 

 ancestors is henceforth futile ; the suggostion that the discrepancies between 

 Mendel's own results and those of subsequent observers who have endeavoured 

 to repeat his observations may be accounted for by differences in the anccstry of 

 the peas used is regarded by Mr Bateson as an attempt to mask a dilficulty one 

 dare not face (I. p. 200). 



2. The essentials of Mendel's theory. — Having made the strikiug addition to 

 Mendel's Statements, described in the last paragraph, Mr Bateson procceds to 



• The Statements here discussed are cUiefly contained in the following works, which will be referred 

 to as I. and II. : 



l.—MendeVs Principlet of Heredity : a Defence, by W. Bateson. 8vo. Cambridge, 1902. 

 n. — lieporta to the Evolution Committee of tlie Royal Society: 1. Experiments undcrtaken by 

 W. Bateson and Miss E. R. Saunders. Koy. Soc, l".i02. 



