290 Mr Batesoiis Revmom qf MendeVs Theory of Ilcredity 



is that the rostilting liybriils, thoiish their " character may be anything," will at 

 least be unitürm ; this uniformity is iudeed considered to diftcreutiate Mendeliau 

 hybrids from those in which the ancestral characters are blended (II. p. 153). 



4. The similarity of reciprocal crosses. — Mendel says that in all his experiments 

 "it is perfectly immaterial whether the dominant character belong to the seed- 

 "bearer or to the pollen-pareiit ; the form of the hybrid remains identical in both 

 " cases " (I. p. .50) ; aud this, at leixst, seems " essential," because the law of segre- 

 gatiou which he formulates to describe his results depends upon it. Mr Bateson 

 however propounds a theory by which "the great fact of gametic purity " may 

 apply to cases in which reciprocal crosses are unlike. He coiiceivcs a species in 

 which the male gametes coutain only one elemeut A of a Mendelian pair, the 

 female gametes containing only the complementary member « of the pair. When 

 a male and female of such a species breed together, the fertilised Zygote is of 

 composition Aa ; but the gametes produced by such a Zygote will contain only Ä 

 elements if the zygote body be a male, only a elements if it be a female. Each 

 individual will therefore completely supprcss half the heritage it receives from its 

 pareiits during the course of its developmeut. This novel and striking conccption 

 is not only completely foreign to Mendel but to all other naturalists, except 

 Mr Bateson and his collcague Miss Saunders (II. p. 132). 



The Horse and the Ass among animals, and certain species of Digitalis among 

 plants, are adduced as examples of species between which the reciprocal crosses 

 are unlike. If Mr Bateson's remarkable hypothesis be applied, it follows that in 

 either the Horse or the Ass, and in some species of Digitalis, there are characters 

 which cannot be transmitted from the female through a son to grand-daughters, 

 others which cannot be inherited bj- grandsons from the maternal grandfather. 

 With so rieh a störe of new and startling theorics to propound, Mr Bateson 

 naturally has no time to illustrate each of them in detail; but it would be 

 interesting if he would spare a momeut to teil us what these charactere are in one 

 of the cases he mentions. 



ö. The hehaviour of " Compound characters" on c^-ossing. — The facts on which 

 Mendel bases his theory of Compound characters are thesc : A ])urpIe-flowercd 

 bean (Phaseohts multißorus) was crossed with a white-flowered bean (7V(. nanus); 

 from the account given we may assume that the hybrids produced were uniform 

 in character, and we are told that whiteness of flower was completely recossive*. 

 These hybrids, when self-fertili.sed, gave a whole series of flower colours; among 

 31 plants obtained, one had white flowers, the others had flowers of various shades 

 from purple-rcd to pale violet. The oxpcrimcnt was not continucd. The ex- 

 planatioii offered is simple, and exhibits the phenomena observed as cxactly 

 parallel with those seen when Peas, of races which differ in two pairs of Mendelian 

 characters, are crossed. Taking the well-known Statements concerning the result 



• I consider only Mendel's stntementä of flower-colour; the rsccs difforod, of course, in other 

 characters, but the discussion of these is not immediately relevant. 



