October 21, 1920] 



NATURE 



255 



Our Conceptions of the Processes of Heredity. 



By Miss E. R. Saunders, F.L.S. 



II. 



'T'HE behaviour of the sex-chromosomes as here out- 

 ■*■ lined suffices to account for the occurrence of 

 sex-linlied inheritance, but the relations found to hold 

 between one sex-linked character and another need 

 further explanation. If a cross is made involving 

 two se.x-linkcd characters, the F, females_when tested 

 by a double recessive male are found lo produce the 

 expected four classes of gametes, but not in equal 

 proportions, or in the same proportions in the case 

 of different pairs of sex-linked characters. Partial 

 linkage (coupling) occurs of the kind which has 

 already been described for the stock and the sweet 

 pea. The parental combinations predominate, the re- 

 combinations (" cross-overs ") comprise the smaller 

 categories. The strength of the linkage varies, how- 

 ever, for different characters, but is found to be con- 

 stant for any given pair. Since the sex-linked 

 factors are by hypothesis carried in the sex-chromo- 

 somes, a clean separation of homologous members at 

 meiosis should result in the characters which were 

 associated in the parents remaining strictly in the 

 same combination in each succeeding generation. The 

 fact that this is not the case has led Morgan to con- 

 clude that an interchange of chromosome material 

 must take place at this phase among- a proportion of 

 the gametes, and that the percentage of these " cross- 

 overs " will depend on the distance apart of the loci 

 of the factors concerned. This phenomenon of link- 

 age may also be exhibited by pairs of characters which 

 show no sex-linkage in their inheritance. The factors 

 involved in these latter cases must presumably, there- 

 fore, be disposed in one of the chromosomes which is 

 not the sex-chromosome. 



To this brief sketch of the main points of Morgan's \ 

 chromosome theory must be added mention of the I 

 extremely interesting relation which lends strong i 

 support to his view, and the significance of which 

 seems scarcely to admit of question, viz. that in 

 Drosophila ampelophila there are four pairs of 

 chromosomes, and that the linkage relations of the 

 hundred and more characters investigated Indicate 

 that they form four distinct groups. It is scarcely' 

 possible to suppose that the one fact is not directly 

 connected with the other. The interesting discovery 

 of Bridges (Journ. Exper. Zool. vol. xv., 1913) that j 

 the app<;nrance of certain unexpected categories among i 

 Drosophila ollspring, where females of a particular j 

 strain were used, coincided with the presence in these 

 females of an additional chromosome adds another ! 

 link in the chain of evidence. On examination it was I 

 found that in these females the X chromosome pair j 

 occasionally failed to separate at the reduction divi- ' 

 sion, and, consequently, that the two XX chromosomes 

 sometimes both remained in the egg, and sometimes i 

 both passed out into the polar body. Hence there 

 arose from fertilisation of the XX egps some indi- 

 vidual? containing three sex-chromosomes, with the 

 resulting upset of the expectation in regard to sex- ] 

 limitation of characters which was observed. ' 



It, however, remains a curious annmalv that In '. 

 the cross-bred Drosophila male no rori-esponding I 

 crossing-over of linked characters, whether associated 

 with the sex-rharacter or not, has vet been observed. 

 His gametes carry onlv the sam« factorial rombina- 

 tions which he received from his parents. For this 



• rrom ihe nptnl^c A.lrfrf .. of ih, ProMeni of S»rtion K (Ratan*) ' 

 <J«Hv»r«l ,( iharardiff mMling of lh« Biilnh AtMcialion on AucuM /t ' 

 onlinpcd from p. 377. ., •: ■ 



NO. 2660, VOL. 106] 



contrast in the behaviour between the sexes there is 

 at present no explanation. The reverse condition has 

 been described by Tanaka (Journ. Coll. Agr., Sapporo, 

 Japan, 1913-14) in the silkworm. Here interchange 

 takes place in the male, but not in the female. 



It must, then, be acknowltKlged-that Morgan's inter- 

 pretation of the cytological evidence has much in its 

 favour. The striking parallel between the behaviour 

 of the chromosomes and the distributional relations 

 of Mendelian allelomorphs is obvious. The existence 

 in Drosophila ampelophila of four pairs of chromo- 

 somes and of four sets of linked characters can 

 scarcely be mere coincidence. The employment 

 of the smaller physical unit in accounting for 

 the reshuffling of characters in their transmission 

 commends itself in principle. The necessity for 

 postulating the occurrence of some orderly irregularity 

 in the hereditary process in order to explain the 

 phenomenon of partial linkage is, it will be seen, 

 inherent alike in both theories. When, however, we 

 come to examine the general applicability of Morgan's 

 theory, we are confronted with a considerable body 

 of facts among plants which we find difficult to recon- 

 cile with the requirement that factorial segregation is 

 accomplished by means of the reduction division. An 

 instance in which this is particularly clearly indicated 

 is _ that of the sulphur-white stock. I have chosen 

 this example because here we have to do with two 

 characters which are distinguished with the utmost 

 sharpness, viz. plastid colour and flower form. The 

 peculiar behaviour of this strain is due to the fact 

 that not only are the two factors for flower form 

 (singleness and doubleness) differently distributed to 

 the male and female sides of the individual, as in all 

 double-throwing stocks, but the factor controlling 

 plastid colour likewise shows linkage with the sex- 

 nature of the germ-cells. As a result every individual, 

 even though self-fertilised, yields a mixed offspring, 

 consisting chiefly of single whites and double creams, 

 but including a small percentage of double whites. 

 So far as the ovules are concerned, the mode of in- 

 heritance can be accounted for on either theory. 

 According to the reduplication hypothesis, the factors 

 XY, producing singleness, and W, giving white 

 plastids, are partially coupled so as to give the 

 gametic ratio on the female side 7 WX Y : i WXy : 

 f wxY : Twxy (or possibly 15 : i : 1 : 15).' On the 

 chromosome scheme the factorial group WXY must 

 be assumed to be disposed in one mcnil>er of the bivalent 

 chromosome formed at meiosis, the corresponding 

 recessive allelomorphs wxy in the other. If the three 

 factors be supposed to be arranged in the chromosome 

 in alphabetical order, and if, on separation, a break 

 take place betwcy»n the loci of the two factors for 

 flower form (as shown), so as to give " cross-overs " 

 of Y and y In about 12 per cent, of the gametes, the 

 occurrence of such "cross-overs " would fulfil the 

 required conditions. But the case of the pollen presents 

 a distinct difficulty on this latter view. The sulphur- 

 white stock is distinguisho<l from both the Drosophila 

 and the Abraxas type by the fact that none of the male 

 germs carry either of the dominant characters. In 

 olnce of the XX— XY form of sex-linke<l Inheritance 

 in the former type and the WZ- -7,7. in the latter, we 

 should need to regard this form as ronstitutini? a new 

 class, which we might represent as DR — RR, thug 

 indicating that both members of the bivalent chromo- 



> The leiien X and V an uvtd h'r* lo dcnoM parliculA faclon, an>, aa 

 in Morian'ii *rh«me, the en'ira wx-chromotomt. 



