The Eehavriar of the Chromasomes as 8tadie<l diroagii Linkage. 259 



as a bearer of the genes involved are certainly extremely rare: and this. 

 in view of the stronff positive evidence pointine toward the chromosomeH. 

 teada me to adopt, as a valuable workine hypothesis, the view that all 

 Xendeiian genes, at least, are contained in the chromosomes. 



•Reduplication": Somatic Sesregation. 



A theory of linkage entirely different from the one foQowed in this 

 paper has been a»iopted by Bateso^ and others. The essence of this 

 interpretation is that seffrpsrarioa occurs at an early period, before matur- 

 ation, and that the differences in the numbers of gametes bearing dif- 

 ferent combinations of genes are brought about by an unaiual amount 

 of ceil multiplication after segregation. This idea seems to have origin- 

 al»! from a desire to explain the phenomena on the basis of stome 

 system of dicho&imy. En support of the latter conception it was argni-d 

 Öiat the gametic ratios observed in cases of linkage fall into a series, 

 the terms of which are 3:1, 7:1. 15 : I. 31 : I. etc. Numerous ratins 

 whicJi do not find any place in this series are now known isee Teo':v 

 113; a, BillEt 'laa. and many of the cases described in this paper). 

 Cor.r.no I'li) has re<!ently shown that it is doubtful if many of Bateso^i's 

 own cases may really be justly placed in it: and BATESO:!f (*13i has 

 himself :idmitted that there seems to be no reason why other ratios 

 should not occur. The tiieory of reduplication rests, then, primarily upon 

 the ;is.sumptii>n of somatic segregation. This conception goes bat^k to 

 the RiJCS-WsiSMLiJrN"" hypothesis of qualitative nuclear division as an 

 ftxpLination of differeutiatiou — a hypothesis ably criticized by de Ysies 

 I'SOi and now generally given up by students of morphogenesis. Batesox 

 and PorxETT i 'I I c i have appealed to cases where pollen and ovules of 

 the same plant transmit ilifferent chara<!ter3 (e. g.. Oenothera^ Petunia, 

 stockai as showing the occurrence of somatic segregation. That this is 

 far from a necessary conclusion is shown by the explanations ~ . . 

 by DE-VaiEa I'lli for Oenothera and by (j<JiiD8<:'FrviTr>T ('13) f< . 

 Et seems to the writer that aU these cases are most simply and satis- 

 factorily exphiined in a manner somewhat similar tii that i)f de Ymss. 

 aameiy. by assuming the occurrence of genes which destroy all ovules 

 (in some of the (Oenothera leases i or all pollen (in stocks and other 

 Oenothera cases) that receive them, and that are linked to the genes 

 the inheritance of which is being studieil. This hypothesis is borne out. 

 in the case of ijenothera. by »tEEäts' i'09) observation that many of the 

 pollen grains and ovules degenerate. Et at least has the advantage that 



Id» 



