E. R. Sattnders 



157 



the earlier experiments'. At that time the fact that the adult condition 

 was not assumed by the partially hoary plants in the early seedling 

 stage, and in the case of the quarter-hoary plants not until quite late 

 on, was not fully appreciated. Moreover the experiments were carried 

 out on a considerable scale, and limits of space compelled one to discard 

 large numbers of individuals in the seedling stage. It is owing to this 

 circuiustance without doubt that the quarter-hoary condition was only 

 recorded in a single individual. For the same reason it is more than 

 probable that in many cases a number of half-hoary plants discarded at 

 an early stage will have been wrongly classified as glabrous. If these 

 points are borne in mind together with the explanation offered above 

 (see note, p. 148) of the false appearance of parthenogenesis in one or 

 two cases, the results in which some discrepancy appears to exist 

 between observation and expectation, based on the conclusions formu- 

 lated above, all fall into line, and th(^ whole series of earlier and later 

 experiments taken together form a concordant whole which can be 

 simply accounted for on the supposition of certain inter-relations 

 between various pairs of factors as set out in the present account. For 

 convenience of reference these earlier results are restated here together 

 with the expectation which the later work now enables us to give for 

 the various matings which have so far been carried out. 



Expectation in the case, of various mnlintjs of a halflioary type, trhetlier a 



white by inhibition or havimj coloured sap, ivith fully honry and ijlabroas 



types. 



Expectation in Fg 



(«) 



Mating 

 \H X Hoary 



IH X CRHK 

 ^H X Glabrous 

 ^HxCEH 

 ^HxCRK 

 IHxGR 

 i H X RHK \ 

 \HxGHK <, 



Expectation 

 inf, 



all // 



all // 

 all i H 

 all ^H 



&\\H 



Hoary 

 12 



m 



36 

 or 36 



Individuals con- 

 taining JJK. 

 Appearance not 

 yet ascertained 



Half- 



lioary 



Quarter- 

 lioary 



(ilabrous 



^HxCR 



\H 



if J proves not to be effi'ctive in the absence of CR- 



— — . 1 2 1 



or — — 3 6 7 



if J proves not to be effective in the absence of CR'^ 

 ' Reps. Evol. Ctee. 



- The considerable ilifference in the expectation as to glalirous plants in the two cases 

 should make it easy to decide this point when once a full count has been obtained in ^2- 



