24 Seed Wheat. 



This set of experiments is interesting chiefly as the initial set of the whole series, and 

 as the one that gave direction to all the following sets. The difficulty of securing 

 anything like exactness in the grading of the samples planted gave rise to the adoption 

 of the sieves with accurately constructed millimetre meshes,* and to the methods of 



Fig. 28. Diagram to represent the method of planting the rows. The first three rows, 2 links 

 apart with plants 1 link apart in the rows, are of the 3 '25 grade. The third row is incomplete 

 because of insufficient seed, and hence only the portion a to 6 could be compared with the 

 corresponding portion of the first row of the 3 '00 grade plants. Most of the fractional rows 

 reported in the tables arose in this manner. 



grading described below. The present samples of seed were hand-picked, and the grade 

 numbers inserted in the table are only approximate, as they were derived afterwards 

 by comparison with graded samples. The land was newly cleared and very patchy, and 

 in this respect not in very good condition for wheat, or for experiment purposes. The 

 season was one of the best ever experienced on the Wagga Farm. The spring rains at 

 the time of sowing were abundant, and throughout the season the conditions were 

 favourable to the growth of wheat. The crops of the district were unusually good. It 

 was the year of the flood of the Murrumbidgee River. 



The object of the experiments was to try good plump grains alongside rust-shrivelled 

 ones, the latter, of course, being at the same time smaller than the former. In the 

 four cases where the smaller seeds are marked shrivelled, they were so from the effects 

 of rust, having been selected for the purpose of these trials from rusty plants the 

 previous season. 



Summing up, we may say that the plump seed gave the best return of grain in four 

 cases out of five, and that the average excess was 18 '52 per cent, taking the lower yield 

 as the basis of computation, while in the single case in which the shrivelled seed 

 exceeded the yield of the plump, the excess was 15'19 per cent. The straw was not 

 weighed : had it been weighed it probably would have been shown, as in subsequent 

 years, that the total marketable product was greatest from the large and plump seeds in 

 practically all cases. This opinion was formed when it was too late to examine the 

 straw, and this was what led to weighing the straw in subsequent years. 



The number of trials was too small to be completely satisfactory. 



How the various sizes of seed were prepared for the second and subsequent trials. 



The method adopted to obtain seed for the comparative tests gave seven grades, each 

 grade being composed of seeds very much alike in size, and, consequently, also in weight. 

 These are the seven grades so frequently mentioned in the first part of this report, and 

 illustrated in Figs. 4 to 10. The seeds might have been selected by weight instead of 

 by size, or they might have been selected by a process taking account of both size and 

 weight. The method adopted that of sieving is, however, nearly the same as that by 

 means of which seed-wheat should be prepared on the large scale, and this fact was 

 what decided the method of selection. 



The following three points are worthy of note in connection with the process of selec- 

 tion by means of the sieves here used. They have a definite bearing on the subject, and 

 should be studied carefully by the reader who wishes to understand the details of the 

 investigation. 



(1.) As the various transverse diameters of wheat grains are unequal, the grains will stop, 

 or not stop, on a mesh of a certain size, according as they happen to present one or the other 

 diameter towards the mesh. This would lead to unequal grading if a sample of grain were 

 simply passed once through the sieve. In order to get gradings that are as nearly as 

 possible comparative, it is, therefore, necessary to continue passing the sample through the 



* These sieves are shown and explained in Figs. 2 and 3. 



