peters: moisture requirements of seeds. 



T.\BLE i. Results of tests with navy beans. 



33 



Numbers 59 and 60 are particularly interesting as they 

 show germination of both seeds with amounts of water sup- 

 plied well below the wilting coefficient of the sand. Number 

 61 unfortunately had a dead seed. As a further check in this 

 series, the beans were weighed when selected, again when 

 the test was complete, and were then dried and the loss of 

 water determined. In the following table "calculated absorp- 

 tion" is based upon the results shown in table 1 above. The 

 actual loss of weight is in every case below the calculated ab- 

 sorption, even though it includes the water originally pres- 

 ent in the seeds. This either indicates that germination can 

 take place with less water than the amounts indicated there, 

 or illustrates the difficulty of making transfers without the 

 loss of water, probably the latter, although corn 4 compared 

 with corn 8 in table 1, given originally 5 per cent and 10 per 

 cent of water in the sand, seem to bear out the former idea, 

 since the absorption was 4 per cent and 48 per cent, respec- 

 tively. 



The final series upon which a report can be made was run 

 with wheat, ten grains to the test. Results follow : 



3 — Sci. Bui. — 860 



