October, 1910.] 



307 



Edible Products. 



assumed that strength and lowness of 

 yield are correlated with one another. If 

 this view be correct the combination 

 of heavy yield with strength is an im- 

 possible one. At present little evidence 

 can be brought forward from one side 

 or the other, though it is worth noting 

 that in some few districts in England 

 Red Fife crops as well as Square Head's 

 Master, Such fresh evidence as can be 

 brought forward at this stage points, 

 however, to the incorrectness of the 

 general view, and seems to show that a 

 heavy crop of good quality is by no 

 means an impossibility. 



The best proofs of its possibility or 

 otherwise would he afforded by a detailed 

 study of the inheritance of yielding 

 capacity, a matter on which it must be 

 admitted we know little at present;. 

 That it is a unit character is perhaps 

 indicated by the fact that some varieties 

 are consistently heavier yielders than 

 others even under a wide range of varia- 

 tion in the conditions. For instance, 

 Square Head's Master has, on this 

 account, gradually driven such varieties 

 as Red Lammas, Chiddam,Talavera, &c, 

 practically out of existence. Further, 

 tbe cultivation of a long series of hybrids 

 between heavy and comparatively low 

 yielding wheats seems to point to 

 segregation of these features. Exact 

 statistics, however, are very difficult to 

 obtain owing to the wide range of 

 fluctuating variability in this character 

 and the difficulty of growing plants 

 under sufficiently uniform conditions to 

 eliminate this. Even when the outer 

 rows of an F.2 culture are neglected as 

 consisting of obviously favoured plants, 

 gaps, due to failures in germination or 

 the attacks of mice, &c, give neigh- 

 bouring plants a greater root range and 

 better opportunities for development 

 than others. In the absence of such in- 

 formation one has to fall back on the 

 yields of the plots grown from the 

 F.2 generation and then on the crops of 

 succeeding years, basing conclusions as 

 far as possible on plots of sufficient 

 acreage to giye trustworthy returns. 

 For this purpose the Fife hybrids men- 

 tioned previously, are fairly suitable, as 

 under the conditions under which these 

 experiments were made Fife barely 

 yields twenty bushels to the acre, whilst 

 Rough Chaff may be expected to give a 

 good average yield of thirty-two bushels. 



In making the selections for further 

 cultivation these strong types, promis- 

 ing to give the best yield, were deliber- 

 ately chosen. Some forty of these, 

 which have been tested in plots varying 

 from one-quarter to three acres in 

 extent, have given in each case yields of 



the same order as the parent Rough 

 Chaff and over 50 per cent, greater than 

 Red Fife on the same farm. On other 

 soils some grown on the large scale have 

 produced crops of forty two to forty- 

 four bushels, but in these cases the 

 cropping capacity of Rough Chaff is un- 

 known, though Fife is known to be a 

 failure as regards yield. The evidence 

 for the segregation for high and low 

 yields is by no means final, but it is 

 sufficient to show that high yields of 

 good quality are not unobtainable. 



The question of heavy yields per acre 

 is intimately connected with the power 

 of resisting the various diseases to which 

 the Wheat crop is liable, as no plant 

 crippled by the attacks of a parasite can 

 be expected to yield its full quantity of 

 grain. It is a well known fact that if a 

 large number of varieties of any plant 

 grown under the same conditions are 

 exposed to the same chances of infection 

 they show marked differences in the 

 extent to which they become attacked 

 by various parasites. This is well shown 

 in the case of wheats and the various 

 rusts which live upon them. In fact, it 

 has now become part of the routine 

 work of many experimental stations to 

 collect and grow as many varieties as 

 possible, with the view of selecting the 

 most immune types for local cultivation, 

 In our earlier tests several varieties were 

 found showing an extraordinary power 

 of resisting the attacks of the common 

 yellow rust, Puccinia glumarum. Even 

 in years when the rust attack has been 

 at its worst they have shown only merest 

 traces of infection. Such immune varie- 

 ties were at once crossed both with 

 moderately and with extremely sus- 

 ceptible varieties to determine whether 

 the power of resisting disease would 

 prove a unit character. In each case 

 the hybrid plant proved susceptible 

 to yellow rust, whilst its offspring con- 

 sisted of immune and susceptible forms 

 in the proportion of one of the former 

 to three of the latter. In the many 

 cases examined the segregation has 

 proved to be exceedingly sharp. The 

 property of resisting the attacks of 

 yellow rust is thus shown to be a Mende- 

 lian recessive, and consequently all 

 extracted immnnes should breed true to 

 this feature in succeeding generations. 

 This point has now been tested many 

 times, with concordant results in all 

 cases. Further, the experiments have 

 shown that immunity is independent of 

 any recognisable morphological charac- 

 ters. Thus in the case of yellow rust 

 there appears to be no valid reason why 

 the plant breeder should not mitigate 

 the evils of its attacks by using this 



