DOUBLE FLOWERS. 



473 



stamens in the characteristic position, also opposite to the petals 

 proper. The view commonly held is that the additional petaloid struc- 

 tures are produced by tangential splitting of the stamens, but the 

 cause of the curious reversal of the surfaces needs some further 

 explanation. Fomerly a more fully double type was in general culti- 

 vation, in which a third set of petaloid structures fill the centre of 

 the flower, and only these central segments have the surfaces reversed. 

 We have another instance in the Wallflower (Cheiranthus Cheiri) of a 

 plant in which two distinct types of double occur, viz. the old-fashioned 

 English form only capable of propagation by vegetative methods, in 

 which a mass of petaloid structures occupy the whole centre of the 

 flower, stamens and carpels being both absent: and the more recent 

 double strains of the German growers in which the flower is not so 

 fully double (fig. 171, e). In these latter plants stamens and carpels 

 are usually normal and the doubhng results apparently from the 

 splitting of the petals, which give rise to a larger or smaller number 

 of additional petaloid structures borne on the four projecting angles 

 formed by the enlarged top of the axis (fig. 171, d). 



But the aspect of the morphological question of greatest interest 

 to the breeder is the practical one — To what extent does the condition 

 of doubleness involve a condition of sterility? As we have seen, the 

 doubling may be of such a nature as to bring about complete loss of 

 function of both reproductive organs. We have the case already 

 mentioned of the old-fashioned double Wallflower; to the same class 

 belong the doubles of Arabis albida, Cardamine pratensis, Abutilon 

 (of more recent introduction). Stocks {Matthiola) (fig. 175), and many 

 others. On the other hand, the doubling may be such as to affect the 

 organs of one sex only, while those of the other remain functional. We 

 have a good example of a case in which it is the female organ which 

 becomes sterile in the double, while the male organs still function in 

 the Petunia, which produces more than the normal number of stamens 

 and an abundance of good pollen, but has the ovary so malformed as, 

 in almost all cases, to be incapable of being fertilized (see fig. 174). 

 Among some highly cultivated strains double individuals are 

 occasionally found from which some seed can be obtained, but these 

 cases appear to be extremely rare; moreover, this capacity to yield 

 seed may not hold good for all the flowers on such individuals, nor, 

 apparently, for all their offspring. Except in these rare cases, there- 

 fore, it is only possible to breed from the singles. Moreover, since 

 the anthers of the doubles are generally deeply buried among the 

 compact mass of supernumerary petals, the propagation of doubles 

 can practically only be ensured by fertilizing singles by hand, with 

 the pollen of doubles, and seed supplied commercially is obtained 

 in this way. 



The converse case where the male organs become sterile while the 

 female are practically unaffected is exhibited by the Sweet William 

 {Dianthus barbatus), where we find double flowers almost always 

 destitute of pollen, but with a functional ovary (fig. 172). This is not 



