RED-CLOVEIt SEED PRODUCTION. 



15 



as often as three times a day. Many uncovered heads were partly 

 destroyed by the grasshoppers, and this undoubtedly accounts for 

 the small seed yield of the uncovered heads, since bumblebees were 

 plentiful. 



HEADS COVERED AND NOT POLLINATED. 



Another experiment was conducted in order to determine whether 

 clover heads kept covered during their entire blooming period and 

 not pollinated could set seed. 



Plants having at least six heads which would come into bloom at 

 approximately the same time were selected for this work. Fifty 



Fig. 5. — General view of the field in which the clover work was conducted in 1912 at Ames, Iowa. The 

 stakes represent plants selected for individual pollination work. The cages in the background were 

 used to test the efficiency of different" insects as pollinators of red clover. 



plants at Ames and 25 at Altoona were selected hi 1911 and 27 

 plants at Ames in 1912. The average seed yields per head are shown 

 in Table III. 



Table III. — Average seed yields of clover heads which were covered with tarlatan and not 



pollinated. 



Location, year, and number of plants. 



Heads covered with tarlatan. 



A 



B 



C 



D 



E 



P 



Ames, 1911: 50 plants 



0.1 



.16 

 



0.11 

 .4 

 .1 



0.15 

 .35 

 







.04 

 



0.16 

 .26 

 .02 







Altoona, 1911: 25 plants 



.2 



Ames, 1912: 27 plants 











Average, 102 plants 



.OS 



.17 



.15 



.009 



.14 



.04 







To the results presented in Table III may be added the results 

 given in column E of the summary of Table II, where 145 heads were 



