Seed Wheat. 



31 



to the germinating power of the seed, and for several reasons. In the first place, one 

 cannot be certain that plants may not yet appear after the count is made ; some good 

 seeds even are very slow in germinating. Then, again, the plantlet may have been 

 good and yet have failed to reach the surface because it was nipped by some under- 

 ground insect. 



It should be borne in mind that the number of plants that appeared above ground in 

 these experiments was probably greater than would occur in ordinary field practice 

 because of the extra care bestowed upon the planting and culture, this care being neces- 

 sary in order to remove as far as possible all disturbing factors that might cause inequalities 

 in the conditions under which the various seeds were grown. It is also right to again 

 call attention to the fact, or at least what seems to me to be the fact, that these 

 conditions were such as to favour the small seeds more than the large seeds. All this i& 

 a separate matter from the inferiorities of ordinary tailings due to the accumulation in 

 them of diseased seed. All the grains used in these trials were derived from selected 

 healthy plants. This is more fully explained elsewhere. With these few remarks, we 

 may now examine the following table of apparent germination : 



APPARENT Germination of Large, Medium-sized, and Small Grains, as shown by the 

 appearance above ground of Plants three weeks after Sowing, 1896. The figures- 

 represent the number of failures in 200 seeds planted. 



These figures may be easily converted to percentages of failure by dividing by two. 

 It will be noted that nearly three times as many small grains failed as large ones, and that 

 the actual failure of the small grains amounted in this case to about one-sixth. 



A considerable number of weaklings appeared, especially among the plants from small 

 and shrivelled seeds, and these, after languishing for a few weeks, died. The soil was- 

 rather patchy, and in a considerable number of instances plants were rendered sterile by 

 diseases due to unfavourable soil conditions. There was almost no loss through loose 

 smut and bunt. 



So far as comparisons of yield were concerned, the soil losses just spoken of were 

 largely rendered nugatory by the row system. The other losses, such as accidents, smut, 

 and bunt, fell irregularly, and were allowed for, if possible, but these losses were of too 

 trifling an amount to cause any anxiety as to the reliability and usefulness of the 

 calculations. The plots were in careful hands all the time, and I never saw experiments 

 carried out under any better field conditions, if the patchiness be excepted. The experi- 

 ment seed was prepared by myself, and its sowing was under my personal supervision, 

 and all the notes were made by myself from frequent observation. The harvesting and 

 so forth was done under my personal supervision, and I attended personally to all the 

 weighings and calculations. The plots were under the competent supervision of the late 

 Mr. JohnColeman, who took great pains to preserve the plots from the least interference, 

 and when I was absent kept me informed as to progress of the growth. The grains were 

 sown in rows two links apart, and the grains were in all cases sown one link apart in the 

 rows in the manner described in my pamphlet on "Agricultural Experiment Work." 

 There were buffer plants at the ends of the rows in this lot of trials as in all others. 



SUMMARY of Comparisons. (Second Trial.) 



If we take as the yield of each trial the sum of the weights of the grain and of the 

 straw, we find that in all cases the various grades have excelled in their production the 

 smaller grades with which they have been compared, if we except grade 225, and even this. 



