JO May, 1909.] U'/wat Improvement Committee. 285 



straw and is storm-proof. But this last-mentioned quality should not be 

 developed coo strongly, otherwise a loss is made in harvesting because of 

 the grain not being readilv separated from the chaff. Some wheats, other- 

 wise good, are rejected for this reason. 



The varietv of high gluten-content is most useful for blending with 

 starchv wheats, in order to produce bread wholesome and easily masticated. 

 The dough of the flour of wheat of high gluten-content is more difficult to 

 knead, but this difficulty disappears where the work is done with machines. 



Semi-Durums. 



Among the semi-durums bred at the College are a number of bald ones^ 

 derived principallv by crossing Medeah, a durimi wheat, with a bald 

 variet\- of common tvpe. The strong influence of the Medeah is marked 

 in almost all the progeny, especially as regards the colouring. It is in this 

 wav that the bald ^Nledeah varieties have l^een produced, several having been 

 grown here for a numter of years. 



The variety Purple Straw x Medeah is represented in Plate 2 ; (2) 

 gives a side view of the spikelets arranged on both sides of the rachis. 

 It will be noticed that the rachis shown in Fig. 4. is very hairy compared 

 with that of Federation and some of the other wheats illu-strated. Fig. 9 

 gives a view of the same variety turned at right angles showing the rows 

 of spikelets. Fig. i presents the underside of the rows of spikelets and 

 shows how they are attached to the rachis. Fig. 8 illustrates the setting 

 of the grains in the spike or ear (in order to .show this two of the glumes 

 have iDeen remo\ed from each spikelet). It hap'pens that in this specimen 

 all of the florets are fertile, except one or \.\\o at the apex. Twentv-three 

 grains are depicted, with the rounded or bosom part turned outwards. In 

 the next figure, Xo. 3, a .side view of the florets is seen; also a view of 

 the line of grain with the crease side uppermost. Fig. 7 shows a line of 

 grain similar to those in Fig. 8, but on the right-hand side mav be .seen 

 the central and iliminutive florets with their attachment to the rachilla, 

 Avhile below are other florets. Fig. 6 is a good illustration of a dissected 

 ear showing the rachillse along which the florets are arranged alter- 

 nately. The grain is not jambed into a corner, as is commonly thought, 

 but takes a definite arrangement on the rachillae, just as the latter are 

 arranged on the rachis. This arrangement prevents the grain in an ear 

 which is apparently \'ery crowded in the spikelets from iTecoming mis- 

 shapen. In Fig. 6 the terminal florets of each .spikelet are shown. They 

 either are not fertile, or being fertile have not developed anv grain in them. 

 Fig. 4 is a representation of the rachis, or axis to which the rachillae 

 and florets are attached. The one under notice is very hairy, both the 

 basal rachis hairs of the spikelets and the lateral rachis hairs being long. 

 In the Durums I have just examined, these hairs .seem to be long, whilst in 

 the Spelts and Emmer they are fewer, and in some instances almost absent. 

 The zig-zag form of the rachis is shown in the different examples depicted. 

 From the shoulder-like parts of the rachis, the florets and rachillae spring; 

 also the tufts of what are termed basal hairs. The functions of these 

 hairs and of the lateral rachis hairs are not evident. It may be that they 

 pieserve a still atmosphere and check evaporation, and thus enable the 

 ear to loetter withstand a drought. 



By carefully dissecting the series of spikelets on one side of an ear, 

 and laying out regularly the parts as detached, it will be noticed that 

 a fairly large percentage of the florets are infertile or have no grains in 

 them. ('See Fig. 5.) 



