788 



Journal of Agricultural Research 



Vol. VI, No. 20 



FIELD TESTS AND CHAMBER GERMINATION TESTS CONTINUED FOR ONE YEAR 



The field tests begun in May, 191 3, were kept under observation for 

 nearly 13 months. In making these tests 500 seeds of each lot were 

 spaced 2 inches apart in rows 4 inches apart in well-prepared beds which 

 had been steam sterilized to kill weed seeds. 



Observations were made frequently throughout the summer and as 

 late as the middle of October. The plants were occasionally thinned to 

 prevent crowding and to facilitate the observations. Early in November 

 the beds were covered with cheesecloth on wooden frames to protect 

 them from contamination by other seeds. 



No observations were made after the cheesecloth covers were put in 

 place until January 2 and 3 at the close of a period of warm, rainy 

 weather. At this time there were a large number of new seedlings in 

 the beds which were planted with lots of seed containing large percent- 

 ages of impermeable seeds and a few in beds which were planted with 

 lots of seed containing small percentages of impermeable seeds. Many 

 of these seedlings had been heaved out by preceding freezes and there 

 were evidences that some seedlings had been destroyed by insects. All 

 of the seedlings which appeared in January and were not otherwise 

 destroyed were killed by subsequent freezing and thawing. 



By the 23d of March healthy clover seedlings of all kinds had appeared 

 in abundance in protected places in the vicinity of the sterilized beds. 

 On this day the cheesecloth covers were permanently removed. 



Table XIII. — Seedling production by impermeable clover and alfalfa seeds in the field 

 in 12% months compared with germination in a germinating chamber 



" These seedlings appearing in midwinter were killed by later freezing. 



6 Produced 1914 stand of plants. 



e About three-fifths of the white-clover bed became so covered with the growth of plants produced in 

 1913 that observations the following winter and spring had to be confined to the other two-fifths of the 

 bed. From these observations the percentages for the whole bed were calculated. 



No new seedlings appeared during February, which was very cold, 

 New growth began late in March and new seedlings appeared in increasing 

 numbers from this time to about the middle of April and more slowly 



