322 Colchicine 



Atter revising the methods for making tetraj3loids and choosing much 

 larger samples, 50 commercial varieties of red clover, new tetraploids 

 superior to the first, were developed. 



In Scandinavian coimtries notable progress has been made with 

 red clover, T. pratense. Twenty-eight chromosomes does not appear 

 to exceed the optimal nimiber. The yield of forage is also indepen- 

 dent of seed production. The seed setting becomes important for 

 propagation purposes but not yield of forage. At least five major 

 tetraploid varieties have been tested over several areas in Denmark, 

 Norway, and Sweden. The results are encouraging as a method for 

 improving red clover by jjolyploidy.^i'^- ^^' ■'-• --^ It is of interest that 

 the new tetraploids in rctl clover do not necessarily come from the 

 best diploid strains. Only by testing the tetraploids can their true 

 value be judged. 



In addition to gigas features valued for forage production, the 

 earlier and more rajjid growth in the second year was better than in 

 diploids. Undoubtedly, the tendency toward a perennial habit in 

 polyploids would seem to be correlated with this trait. Susceptibility 

 to insects and diseases are a weakness in most strains, diploids as well 

 as tetraploids, but there were some red clover tetraploids with ex- 

 cellent insect and disease resistance. One red clover strain, Sv. 054, 

 from a diploid \aricty Merkur had good yielding capacity and re- 

 sistance to the nematode, clover eel. 



Diploid alsike clover, T. hyhridinii, made tetraploid, showed 

 promise at once, giving consistent increases in forage from 15 to 25 

 per cent. For overwintering capacity the alsike clover was good from 

 the start. --'J Continued successful performance stimulated a change 

 to breeding on the tetraploid level. VV^ithout doubt, these two tetra- 

 jjloid clovers have made satisfactory performance. 



A third species, T. repens (white clover) , was not successful, biu 

 as this is a natural tetraploid, 32 chromosomes, finther increases pre- 

 siunably took the niniiber to 64, a niunber above the optimum for the 

 species. We must conclude that one cannot draw a general rule for 

 all cloxer breeding (ct. Chapter 1 1, Ref. No. 4) . 



The tetraploid Melilotus suffered from a reduced fertility and 

 was not as promising for practical purposes, althotigh there were 

 enough differences in fertility among eight plants of tetraploids to 

 make jjrogress in selecting toward higher fertility. '^i Crosses and 

 selections demonstrated that higher levels of self-fertility coidtl be 

 obtained. If interspecific hybridization could be effected, the com- 

 bined germplasm would open another avenue for analysis. 



Polyploidy has been olitaincd in MecUcago satixia, M. media, M. 

 lupuUna, and M. denticulata.-''' Vigorous strains appeared among 

 these polyploids; however, the usual reductions in seed setting were 



