76 THE FIRST BOOK OF FARMING 



For large seeds take two plates (see Fig. 39) 

 and a piece of cloth as wide as the bottom of the 

 plate and twice as long. Count out fifty or one 

 hundred seeds from a package, wet the cloth and 

 wring it out. Place one end of the cloth on the 

 plate, place the seeds on the cloth and fold the other 

 end of the cloth over them. On a slip of paper 

 mark the number of seeds and date, and place on 

 the edge of the plate. Now cover the whole with 

 another plate, or with a pane of glass to keep from 

 drying. Set the plate of seeds in a warm room 

 and examine occasionally for several days. If the 

 cloth tends to dry, moisten it from time to time. 

 As the seeds sprout take them out and keep a rec- 

 ord of them. Or leave them in the plate and after 

 four or five days count those that have sprouted. 

 This will give the proportion of good seeds in the 

 packages. 



For small seeds fold the cloth first and place the 

 seeds on top of it. 



Another good tester for small seeds is made by 

 running about an inch of freshly mixed plaster of 

 Paris into a small dish or pan and moulding flat" 

 cavities in the surface by setting bottles into it. 

 The dish or pan and bottles should be slightly 

 greased to prevent the plaster sticking to them. 

 When the cast has hardened it should be turned 

 out of the mould and set in a large dish or pan. 

 One hundred small seeds are then counted out and 

 put into one of the cavities, others are put into the 

 other cavities. Water is then poured into the pan 



