WILLIAM J. FARRER. 17 



SOME CROSS-BREEDING PROBLEMS. 



It is interesting to note that although Mendel's law (now largely the basis 

 on which plant-breeders work) was unknown to Farrer until shortly before 

 his death, he nevertheless recognised the possibility of combining different 

 characteristics in the parents and segregating those particular ones he desired 

 to perpetuate. 



The following passage, which is taken from Farrer's report to the Fourth 

 Rust-in- Wheat Conference, 1896, indicates the general lines on which he 

 proceeded : , 



In order to combine the qualities of earliness ot maturity and resistance to rust in 

 one variety by means of cross-breeding, late rust-resistant and early rust-liable sorts, 

 as I have already pointed out, have to be mated. It will be well to pause for a moment 

 and consider what we ought to expect from the union of types which differ so widely in 

 these two qualities, as well as in others, such as the relative hardness, size, character 

 of the grain, &c. What we generally see in the analagous case of the animal kingdom, 

 with which we are more familiar, is that when parents, which are not closely similiar, 

 are united, if the progeny be numerous, certain individuals inherit some of their 

 characteristics almost entirely from one parent, combined with other characteristics which 

 they have inherited almost entirely from the other parent, whilst as regards a majority 

 of their characteristics they are intermediate in various degrees between both parents ; 

 and when this happens in different degrees and in a different manner with all the 

 progeny, it will be seen how it comes that no two individuals of the same parentage are 

 ever exactly alike, and that the greater the dissimilarity of the parents the greater 

 will be the difference between the offspring of the same union. I will attempt to illustrate 

 briefly what I mean, and for this purpose will make the case as simple as I can, and apply 

 it to the subject we are actually dealing with. 



Suppose I have mated a rust-resistant-late with a rust- liable* -early variety of wheat. 

 The greatest diversity of types will be shown by the offspring which grows from seed of 

 the first generation of the cross from such seed as I am distributing. Suppose we have 

 100 plants growing from such seeds, which are of the same parentage. Out of this 

 number I would expect there might be one or two say one which has inherited in a 

 very high degree, possibly in as high a degree as the parents themselves possessed them, 

 the qualities we are seeking to secure from both parents. A few more five I would 

 expect to inherit high rust -resistant power from one parent, associated with moderate 

 earliness from the other ; and five more to inherit a high degree of earliness with fair 

 rust-resisting power. The remaining eighty-nine I would expect to inherit these 

 qualities in various degrees intermediate between the two parents ; and something of 

 this sort is what I find actually to occur in most cases. The work then, of the person 

 whose business it is to make use of these 100 plants is essentially the work of selecting 

 as many of these eleven plants as promise to fill our requirements, and that work, as I 

 have found out from actual experience, requires for its successful performance a close 

 attention, care, patience, thoroughness, and system. 



Professor R. D. Watt, Professor, of Agriculture, Sydney University, to 

 whom this excerpt was submitted, comments as follows : 



This quotation shows that, although Farrer was at that time in ignorance of Mendel's 

 historic experiments, he was working more or less along Mendelian lines for the main 

 practical lesson of Mendelism is that, if two varieties of any crop, each of which 

 possesses one desirable and one undesirable character, are crossed, there will appear 

 amongst the progeny one or more individuals possessing the two desirable characters, 

 and that some or all of them will breed true to both these desirable characters. The 

 proportions mentioned by Farrer do not agree with Mendel's figures probably because 

 resistance to rust (Puccinia yraminis) is not a simple Mendelian factor in inheritance. 



Two instances of Farrer wheats may be quoted to show how Farrer used something 

 very closely akin to the Mendelian method. Of the many varieties he had at his 

 disposal, a crossbred called Maffra was noted for its early maturity, which was its main 

 asset ; Zealand was one of the best late maturing wheats for hay, and Rymer one of the 

 best late maturing prolific grain yielders. Farrer desired to get a variety of wheat 

 suitable for hay which would mature sufficiently early to enable the farmer to have his 

 hay in the stack before the grain harvest commenced. He therefore crossed Zealand 

 with Maffra, .and amongst the progeny he found a few plants which combined the 

 excellent hay qualities of Zealand with the early maturity of Maffra. From these few 

 plants he saved the grain and sowed it in small plots, found that it bred true, and thus 

 he evolved the variety Firbank, which is still perhaps the best early maturing hay wheat 

 for New South Wales conditions. 



