Criteria for Judging Polyploidy 365 



clreds of cases. -^ The sizes can be judged by volumetric measure- 

 ment, weights, or length and ^vidth measurements. As a sorting 

 method for choosing the tetrai:)loid rye plants in the treated genera- 

 tion, size of seed is a reliable feature.-" The grain weights of letra- 

 ploid rye were distinctly separated from dii)loids. Table 15.1 shows 

 the increase based on thousand-grain weights for diploid and tetra- 

 ploids. A mean weight of 30. 'M ^vas obtained for diploid and 46.50 

 for tetraploid.''- 



Increasing the size of seed has been used as an argument to im- 

 prove the crop yield of diploids through polyploidy. The fallacy lies 

 in the fact that the seeds of tetraploids may be larger and heavier, 

 but the reduced number of seeds per plant prevents complete use of 

 the increase. Reduced fertility in autoploids is the most common 

 cause of decreased yield in number of seeds. Decreased seed produc- 

 tion in watermelon brought out this relation. A comparison of ten 

 fruits, diploid and tetraploid, showed avarages of 290.0 and 92.7 per 

 fruit, respectively.-i Since cultivation was similar and the varieties 

 were strictly comparable, the reduction was directly correlated with 

 tetraploidy. For reasons discussed in the previous chapter, triploids 

 are without seeds. 



Amphiploids do not show the same consistent increase in seed 

 weight or size compared with the parental species. A comparison be- 

 tween amphiploids and parental types was made among species of 

 Broinits of the Gramineae. On the basis of weight for 200 seeds, the 

 amphijiloid increased as much as 75 per cent, while other increases 

 w^ere not more than 17 per cent^^ (Table 15.2). Genetic factors and 

 the contributions by each parent have a greater influence than merely 

 doubling the number of chromosomes. 



A given kind of plant may regularly show specific marks among 

 the tetraploid seeds. Close inspection of the tetraploid seed of water- 

 melon showed that fissures developed in the seed coat upon drying.-^ 

 A rupture of the outer layers of ovules creates this condition. These 

 marks as well as size of seed are good criteria for making preliminary 

 sorting of the tetraploid. Another distinction was the thickness of 

 "triploid" seeds and tetraploid. Seed from tetraploid fruit pollinated 

 by di])loids are called "triploid" seed and are thinner than the seed 

 from tetraploid fruits pollinated by tetraploids. -^ Other marks such 

 as coarseness, special spines, ridges, and color differences, once noted 

 can be reliably used as guides in selection auiong treated plants and 

 the tetraploid generations.^' i^- ^^- ^^'' "' ^^ 



Fruits of tetraploids are not necessarily larger than those of dip- 

 loids. Nevertheless, distinguishing marks can be found among teira- 

 jDloid fruits. The external marking, shape, and attachment to plant 

 are some of the features that have been used. Parthenocarpic fruits, 

 such as ihe triploid, may be somewhat triangular. -^ llie Iksin por- 



