THE A WA KENT NO OF THE SEED 



33 



merit were to be repeated and the bottles filled more\ 

 nearly to the right degree at the start. We may vary 

 the experiment by dispensing with the sand and simply 

 filling up the bottles to various heights with soaked 

 seeds. 



The fact that so much air is necessary seems to 

 indicate that some part of the air is changed (or 



29. Seeds on wet sand with different amounts of air at their disposal. 



"used up") by the seeds. Let us see if this is so. 

 After two or three days, cautiously remove the cork 

 from one of the bottles (containing about five -sixths 

 air) and immediately lower a lighted match into it; 

 or, better still, take a tall bottle or jar, place a layer 

 (an inch or so deep) of seeds in the bottom, stopper 

 tightly, allow it to stand a day or so and test with 

 lighted match. If the match goes out, it indicates 

 that the oxygen of the air has united with some other 



