The Autoploids 321 



fiO per cent fertility, while in the parental diploid fertility was over 

 90 per cent. Yet the hybrid between the subspecies japonica and 

 indira may even drop to 11 per cent when fertility is measured bv 

 seed formation. Sterile F/s, if doubled, immediately raised the seed 

 formation higher than autotetraploids.^'^ As the fertility decreased in 

 a oi\en Fj hybrid, the fertility increased in the corresponding tetra- 

 ploid. That is, the more sterile the diploid F, hybrid, the higher was 

 its restoration of seed fertilitv after doubling.-'"' Pollen sterility ap- 

 proximated the same rides. Thus the disadvantage met by strict auto- 

 tetraploidy seems to be overcome in this type of program. Some real 

 obstacles may yet be encountered in trying to stabilize the polyploid 

 that combines japonica and indica genomes. Further segregation must 

 be studied. 



No quick results can be expected in spite of the apparent solution 

 to the fertility problem, for the tetraploids from hybrids are, like all 

 tetraploids, unselected. Judging from the high midtivalent formation, 

 segregating progenies in F^ and later generations can be expected. 

 This fact may offer exceptional plant breeding opportunities along 

 with serious obstacles. Obviously, these plants and such methods will 

 receive attention in the future as another approach toward plant 

 improvement in rice. 



An c\tensi\'e literature is devoted to autotetraploid barley.-"'^ 

 Some spontaneous An races have been isolated. Also, colchicine has 

 been used by several investigators. Morphological characters that 

 change with polyploidy are well catalogued along with several ex- 

 cellent physiological studies. The progress has been summarized in a 

 comprehensive review, and little more need be added. The practical 

 uses for barley have not come up to those of autotetraploid rye. 



Autotetraploid maize has been followed over a long period, since 

 the earliest strains were made by heat treatment, before colchicine 

 methods were available. Fertility differences cannot be correlated 

 entirely with chromosomal processes at meiosis. The slower growth 

 and reduced fertility are disadvantages of the tetraploid. The dou- 

 bling of monoploids to autodiploids ^vill be developed in another sec- 

 tion. 



Other cereals of economic importance, being natural jK)lyj)loids, 

 require other approaches. The autoploids are inferior to diploids and 

 provide genetic materials only. 



1-^.1—2: Forage, range, and pasture spcdes. Raw polyploids in 

 some species of TrijoUum showed an immediate advantage over the 

 diploid in forage production. ^^-^ The data were obtained from limited 

 scale testing. \\'hen the tetraploids were distributed for larger scale 

 trials, the difficulties not encountered ^\•ith small tests then appeared.^ 



