264 SOILS AND PLANT LIFE 



appear in the table, compare it with one as nearly the 

 same size as possible, estimating the number that should 

 be equal to one per cent. 



Determine in this way the percentage of each kind of 

 weed seed, and estimate the percentage of dirt and chaff. 



It is necessary next to make a germination test of the 

 sample. Procure two pie plates or dinner plates and a piece 

 of clean, boiled muslin four or five inches wide and about 

 eighteen inches long. Dip the muslin in water and wring 

 it out. Then double it once, lay one end in one of the 

 plates, scatter one hundred average seeds on it, fold over 

 the muslin again so as to cover the seeds. Lay the other 

 plate upside down on this one. The muslin should not 

 protrude from between them; and they should fit as 

 closely as possible so that little moisture will escape. 



The plates should now be set away while the seeds 

 germinate. However, the upper one should be lifted off 

 every day or two to make sure that the musUn has not 

 become dry. If it has done so, add water drop by drop 

 until it is well moistened but not saturated. 



At the end of a week, open the cloth and examine the 

 seeds. You will doubtless find that most of them have 

 germinated, but that some which look perfectly sound and 

 good have not done so, while still others are shriveled or 

 discolored, showing that they are incapable of germina- 

 tion. The second class, those which appear good but have 

 not germinated, are known as hard seeds. It is generally 

 considered that one half of them would probably ger- 

 minate in the soil, but that the remainder would not do 

 so. In determining, then, the percentage of germinable, 

 or viable, seeds, we count those that have actually ger- 

 minated and add to this nmnber one half of the hard seeds, 

 and the remaining seeds are regarded as non-germinable. 



Conclusion. — Make a copy of the following form in 



