742 IOWA DEPARTMENT OF AGRICULTURE 



Oxtongue seeds (fig. 21, r) are mostly lance shaped, reddish yellow, the 

 surface having dark transverse lines, the margin at the broader end of 

 the seed roughened (as shown at the left of the figure) ; a part of the 

 seeds whitish, curved, the inner curved edge white-hairy (shown at the 

 upper right-hand of the figure) ; common in poorly cleaned red clover and 

 alfalfa seeds imported from Europe; not found in domestic grown seed. 



Hawkweed picris seeds (fig. 21, s) are reddish brown, straight or 

 curved, bearing fine transverse, dark-edged ridges, the faces of the seed 

 having one or two slender grooves lengthwise; frequently found in im- 

 ported red clover and alfalfa seed. 



Hawkweed seeds (fig. 21, t) are small, black, cylindrical, ridged length- 

 wise, pointed at one end, the opposite end bearing a short brush of fine, 

 white bristles; common in grass seed. The seeds of several kinds of 

 hawkweed are similar. One kind is the orange hawkweed, which has 

 proved troublesome in the Northeastern States. 



DETAILS OF MAKING SEED TESTS. 



Procedure. — Certain details of procedure in making seed tests should 

 be followed if tests of seeds of different kinds are to be fairly comparable. 

 The natural course to be followed in testing forage-crop seeds involves, 

 in general, the preparation of the small sample for actual test, its exami- 

 nation, the separation of the crop seed and its impurities, a test of the 

 germinating power of the crop seed, and the determination of the actual 

 value of the seed as compared with pure seed. 



Careful work in making a test is comparatively useless if the sample 

 does not fairly represent the bulk of the seed from which it is taken. 



The responsibility for selecting the small trade sample rests entirely 

 with the dealer who submits it. When seed in bulk, as in a sack, is to 

 be sampled, small amounts of seed should be taken from the top, bottom, 

 sides, and center of the sack. If the sack be emptied and the seed thor- 

 oughly mixed, it is probable that a fairer sample can be taken. 



The test sample.— Since the small sample thus taken or the trade 

 sample will be too large to be tested in its entirety, it must be again 

 subdivided to obtain the test sample. In official tests this all-important 

 subdivision is effected by the use of a mechanical mixer, which takes a 

 little from all parts of the larger quantity. In home testing perhaps 

 there is no better plan than to pour the seed in a symmetrical pile on 

 a flat surface and carfully subdivide it by means of a table knife. A 

 subdivision amounting to a teaspoonful for the clovers and small-seeded 

 grasses, a tablespoonful for the coarse grass seeds, and a considerably 

 larger amount for cereal grains may be accepted for the test. 



The balance previously described having been put in proper condi- 

 tion for use, the total weight of the selected sample is to be taken 

 and recorded in terms of whole and fractional shots. This permits the 

 computation of percentages by ordinary division according to the methods 

 used in percentage. If, however, quantities of seed balancing QV^ or 12% 

 shots are used, the one-sixteenth shot weight represents 1 per cent or one- 

 half of 1 per cent, as heretofore explained. 



