582 ERNST J. SCHREINER 



results were contrary to research by other workers who found substantial 

 gains of single-cross over synthetic yields. 



Craigmiles, Crowder, and Newton (1965) compared the yield of F-^ 

 bromegrass hybrids (produced by isolating vegetative material of 2 self- 

 incompatible but cross-compatible clones) , 2 experimental 6-clone 

 synthetics, and the commercial variety, Southland. The ¥\ hybrid was the 

 most productive, with an increase of 20 percent over Southland and 15.8 

 percent over the best synthetic tested. The synthetics were not signifi- 

 cantly better than Southland, the highest-yielding variety previously 

 tested. Christie (1967) has recorded serious doubt as to the value of 

 synthetics : 



"Forage crop breeders have become increasingly concerned 

 over the lack of an increase in yield per se as a result of 

 breeding. Improvements have been made in such traits as 

 winter-hardiness, leafiness, disease resistance, and chemical 

 composition, but the increase in yield of dry matter has been 

 disappointing ... .At present, most breeders select clones on 

 the basis of phenotype, evaluate for combining ability, and 

 then use the superior clones as the basis of a synthetic 

 variety. In theory, this seems to be a promising method. 

 Since progress has been disappointing, why don't the results 

 confirm the theory? It is becoming increasingly obvious that 

 one possible weakness in present forage breeding is the assump- 

 tion of random pollination. .. .There are other possible reasons 

 which it is hoped will be reviewed at a later date." 



Synthetic Varieties of Forest Trees 



Selfing is necessary with annual crops, such as maize, to maintain 

 the strains (homozygous lines) used for the parental Syn-0 generation; 

 selfing is not necessary with forage crops or forest trees where the 

 parental Syn-0 plants can be maintained as clones. Synthetics of alfalfa 

 and other perennial forage crops have been developed by intercrossing 

 noninbred plants without control of pollination except for restricting 

 parentage in number and by isolation (Kehr et at. , 1961b) . This proce- 

 dure will apply also to the production of synthetic varieties of forest 

 trees. Also, as with forest trees, the production of synthetics has 

 been favored because of the difficulty of large-scale controlled crossing 

 for the production of hybrids for commercial use. For the same reason, 

 open-pollinated progeny tests, top-cross tests, and polycross tests have 

 been generally utilized to obtain progenies required for tests of 

 combining ability. 



Clonal and seedling seed orchards are the base populations--the 

 Syn-0 generations — for the production of synthetic varieties. The plant- 

 ing stock obtained from seed orchards before they have been progeny tested 

 and rogued for general combining ability should be recognized as the first 

 generation product of mass selection. 



The open-pollinated progenies of the orchards after roguing will be 

 Syn-1 synthetics. Seed-increase to produce Syn-2 foundation seed, and 

 Syn-3 certified or registered seed (as in agricultural crops) is neither 

 practical nor anticipated for forest trees. Seed orchards are generally 

 large enough to meet the forestation requirements of the organizations 

 that establish them. A succession of seed orchards, for example, as in 

 Bingham et at. (1969) Stage A, Stage B, and Stage C orchards, established 



