Harlan and Pope: Value of Back-Crosses 



321 



other characters, morphological and 

 physiological, are desired from the 

 Manchuria parent. How feasible it 

 may be to secure by recombination the 

 characters of a parent is difficult to 

 answer. Inheritance studies neces- 

 sarily involve characters that can be 

 measured or evaluated. Most of these 

 are morphological in nature. Occa- 

 sionally a physiological detail of met- 

 abolism becomes evident through 

 chance, as when anthocyan colors are 

 formed. The number of variations in 

 metabolism which are not thus revealed 

 cannot even be guessed at, but must 

 be large. We do know that variations 

 in the protoplasmic complex are inher- 

 ited. That the difference of the cell 

 content of varieties is evidenced in 

 many ways, such as the treatment 

 necessary to coagulate the proteids of 

 wort from barley, the variations in 

 milling quality in wheat, and recently 

 by the specialized forms of rust. Fol- 

 lowing the lead of Dr. Stakman, 37 

 specialized forms of stem rust now 

 are known and the number seems 

 limited only by the facilities for test- 

 ing. These forms are largely differen- 

 tiated by the reaction of different 

 strains of wheat to them. This means 

 that there now are known 37 different 

 heritable protoplasmic complexes of 

 wheat that have to do with rust resist- 

 ance. It is probable that several times 

 that number actually exist. There 

 must also be many heritable variations 

 of cell content not associated with rust 

 resistance but associated with the vigor 

 and yielding power of the plant and 

 with the nature of the proteids in the 

 grain which may be of importance in 

 milling, malting, or other commercial 

 use. 



At any rate, in the mating of two 

 parents there is an indefinite number 

 of heritable physiological characters 

 that will combine in an unknown num- 

 ber of combinations and the nature of 

 these combinations cannot be deter- 

 mined by an inspection of the plants, 

 as is the case with morphological char- 

 acters. We are thus breeding for 



characters we cannot see and whose 

 numbers we do not know. This brings 

 up the question of probabilities. If we 

 are dealing with an unknown number 

 if invisible characters, what is the 

 chance of securing a segregate that has 

 all the Manchuria characters except 

 roughness of awns ? When our patho- 

 logists get under way on the leaf and 

 root fungi detrimental to barley it is 

 probable that they can equal in num- 

 ber the rapidly increasing parasites of 

 wheat. When to the number of differ- 

 ent protoplasmic complexes shown by 

 disease susceptibility is added those 

 differences which must be associated 

 with other heritable qualities the num- 

 ber of non-visible factors may reach 

 into the hundreds. If there were only 

 20, the breeding by the use of large 

 numbers would be impossible, assum- 

 ing that all 20 of the physiological char- 

 acters of Manchuria are superior to 

 those of Lion for conditions in Minne- 

 sota, and that these are inherited inde- 

 pendently. There would be just one 

 chance in 1,048,576 that the 20 desir- 

 able Manchuria characters would be 

 found in any segregate and one chance 

 in four that this segregate would be 

 smooth-awned if found. 



It obviously would be impossible to 

 grow such a generation or to discover 

 the desired plant if grown. On the 

 other hand, in back-crossing, if the 20 

 characters are inherited independently, 

 there is a rapid elimination of those 

 coming from the Lion parent. With 

 each recross the Lion "blood" is re- 

 duced one-half ; in five back matings 

 only one-sixty-fourth of the blood is 

 not ]\Ianchuria and that one-sixty- 

 fourth is heterozygous with only one 

 chance in 128 of any one factor being 

 finally other than Manchuria. That 

 is, if there were 20 independently in- 

 herited factors the expectancy would 

 be for 108 out of 128 plants to be en- 

 tirely homozygous for Manchuria char- 

 acters and the remaining 20 to be 

 heterozygous for only a single char- 

 acter. It should be possible here to 

 select the plant desired. 



