SEED TESTING . 45 



named. "Rubbish" impurities are of small consequence compared 

 to weed seeds, especially if bad; for instance, a sample of clover 

 99 per cent pure and with 99 per cent germination, would be unsatis- 

 factory if it contained dodder. 



5. What does the seed weigh? Generally heavy seeds within the 

 limits of the species are best. 



6. Are the seeds dry? Well-dried seeds keep best 'and give best 

 results. 



7. W T here did the seeds originate? Seeds from some countries or 

 even localities may be better or poorer than from others. 



8. What percentage will germinate? And at what rate or 

 "strength"? 



9. What percentage are "hard" seeds? Perhaps this will not be 

 considered as serious a question as formerly when the sulphuric 

 acid method (47) comes into more general use. 



FIG. 38 SAND BOX SEED GERMINATOR 

 The wires divide the surface into squares in which the seeds are placed. 



74. Simple conveniences in seed testing include a 

 pocket lens to examine small seeds ; sheets of stout white 

 paper or cardboard on which to spread seeds for examina- 

 tion ; a spatula-like piece of hardwood, bone or celluloid 

 to separate the seeds ; tweezers to pick out small seeds ; 

 a small scales for weighing phials of truly named culti- 

 vated plant and weed seeds with which to compare seeds 

 to be examined ; gummed labels for numbering or nam- 

 ing samples ; blotting paper, strips of flannel, and clean 

 sifted sand for germination tests. 



A small germinator may be made by wetting sand in a 



