DOUBLE FLOWERS. 



479 



tion of the behaviour of these ever-sporting forms becomes clear if 

 they are tested by crossing with non-double-throwing singles. If in 

 such matings the true-breeding single is used as the female, and the 

 double-throwing single as the male parent, the resulting cross-breds 

 are all found to give a proportion of doubles — amounting in the typical 

 case to an average of one in four; all, therefore, must have inherited 

 from their double-throwing father the particular constitution which 

 enables an individual to throw double offspring. As explained above, 

 we have reason to think that the capacity of any single-flowered Stock 

 plant to produce double offspring depends upon the absence of certain 

 factors requisite for the production of a normal (i.e. single) flower 

 from one of the two germ cells which united to produce the individual 

 in question — that is to say, either from the contributing male or from 

 the contributing female element. In the case in question it is the 

 male element which lacks them. If these essential factors are lacking 

 in both the contributing germ cells, the resulting individual is a 

 double ; if they are present in both, the individual is a single breeding 

 true to singleness; but if they are brought in on one side of the 

 pedigree, and not on the other, the individual is a single, some of 

 whose reproductive cells will contain these factors while some will 

 not. The last-mentioned type of individual will consequently give 

 rise to a mixed offspring of singles and doubles. From the results of 

 the mating non-double-throwing single as the female parent x ever- 

 sporting single as the male we may then conclude that all the pollen 

 of an ever-sporting single lacks the factors essential to singleness, or, 

 as we may paraphrase it, carries doubleness. 



If, on the other hand, the cross is made in the reverse form, 

 the double-throwing-single being employed as the female parent, and 

 the pure-breeding form as the male, only slightly more than half the 

 resulting cross -breds are found to yield a mixed offspring of singles 

 and doubles, while the remainder behave as true-breeding singles. 

 If we are right in the supposition that a double results when both the 

 contributing germ cells lack certain factors, then, since we have seeji 

 that these factors are absent from the pollen of the ever-sporting single, 

 it follows that the proportion of doubles occurring among the offspring 

 of cross-bred singles of this class represents the proportion of the 

 ovules in the mother-plant, which lacks these factors, or, in other 

 words, carry doubleness. This proportion, as stated above, is found 

 on the average to be rather more than half, while the remainder- 

 rather less than half — carry singleness. In the double-throwing 

 singles, then, we have a plant whose pollen all carries doubleness, 

 but whose ovules are mixed, about half carrying singleness and half 

 doubleness. 



This conception of the constitution ol the double-thrO'Wing single 

 type provides an explanation of the ever-sporting habit. In this way 

 we are able to understand why all the successive generations of single 

 offspring of any ever-sporting individual will themselves be ever- 

 sporting, for every single individual in the pedigree has the same 



