E. R. Saunders 1-1:7 



two factors (indicated by H and 70 which are necessary to produce 

 hoariness. A further point, of special interest, is that the hoary 

 character due to the presence together of H and K is not manifested 

 unless the colour factors C and R are also ijresent ; that is to say, a plant 

 may contain H and K, but if the flower is white (or cream) oiving to the 

 absence of C, or R, or both, it will not be hoary^ This curious inter- 

 dependence between the hoary character and presence of sap colour in 

 the flower is not confined to the ordinary garden varieties of Stocks. 

 In M. sinuata one likewise meets with a hoary type and a glabrous 

 variety (var. oyensis) in which loss of hoariness goes with loss of flower 

 colour. The hoary type has pale pui-ple flowers ; in the glabrous plant 

 the flowers are white. The two are recorded as growing associated 

 together on the lie d'Yeu oft" the coast of La Vendee^ 



We may express the factorial composition of the hoary and glabrous 

 types as folhnvs : 



(1) Hoary types will contain G, R, 77 and K. 



(2) Coloured glabrous strains may contain CRH, GRK or only GR. 

 (R) White or cream glabrous strains may contain any of the 



following single factors or groups of factors, C, 7?, 77, 7i, UK, GH, GK, 

 GliK, RH, RK, RHK; or they may lack all. 



Among glabrous types then, if homozygous, we shall have to consider 

 3 possible sap-coloured forms and 12 possible whites and creams. If 

 interbreeding occurs giving rise to heterozygotes we shall expect a larger 

 number still, as many in fact as there are combinations of these factors 

 which do not involve the presence of all four, i.e. a total of 20 among 

 sap-colours and 4-5 among the white and creams. Similarly in the 

 hoary group, to the one homozygous form we should add 15 possible 

 heterozygotes. The precise constitution of all these different forms has 

 been set forth in detail in an earlier paper and need not be repeated 



1 The combination of hoariness and absence of any sap colour does however occur. 

 We are familiar with it in the case of the white foi-m of incana and in hoary white 

 Brompton Stocks, but in these cases the white tlower does not result from the abaeiicf of 

 C or R. All the four factors C, R, H and K can be shown to be present; the absence of 

 colour here must be ascribed to some other cause of the nature of inhibition. In 

 Bromptons and some annuals the inhibition appears to be absolute, but in hwana a 

 noticeable tiiigeing of the flowers on fading affords further proof that the colour factors 

 are not actually wanting. Whereas then in the (jhihroiiit whites one or both of the 

 necessaiy factors for colour are absent, in hoanj whites they are present but non-effective. 

 It is only to the class of deficieut whites (as opposed to inhibition whites) that the state- 

 ment made in Punnett's Mendelism (3rd ed. p. .50) holds good, viz. "that in families 

 where coloured and white Stocks occur the whites are always glabrous." 



•- See Curtis's Botiinical Magazine, Vol. cxxvi. Tab. 7703, 1900. 



10—2 



