The moisture was 12 per cent, and the retentive capacity 64 per cent, of the dry 

 weight. Such a condition was favorable for a liigh percentage of formaldehyde 

 injury and when the onions came up the injury was apparent. The percentage of 

 chemical loss was determined on June 1 and the yield data were taken on August 30. 

 The data, presented in Table VIII (p. 29) show that the highest gain was obtained 

 by the use of the 1-50-5000 formula and the next by the 1-64-3000 formula. The 

 injury caused by the 1-50-3000 formula was so severe that there was an ultimate 

 loss even though smut was controlled very effectively. On such a soil the 1-5O-5000 

 formula is undoubtedly the best. 



Conclusions from all of the Formaldehyde Experiments 



After five years of experimentation, the writers came to some fairly definite con- 

 clusions in regard to the control of onion smut with formaldehyde. These conclu- 

 sions, which will now be briefly stated, are supported by the data which have just 

 been presented. 



1. Smut can be controlled in any field in the Connecticut Valley and during any 

 season by the use of a formaldehyde solution. 



2. Extreme dilutions such as one part of formaldehyde in 128 parts of water, 

 involving the use of a large amount of water, extra labor and inconvenience, are not 

 necessary under the conditions which prevail here. 



3. Within fairly wide limits, the control of smut is not dependent on the dilution, 

 but on the actual amount of formaldehyde applied per unit distance of row. Thus, 

 the concentration is not hmited by lack of control, but by danger of injury to the 

 seeds. 



4. Any formula for the apphcation of formaldehyde which is strong enough to 

 control smut, also causes a certain percentage of the seeds to fail to germinate. This 

 injury is especially noticed when the soil is very dry at planting time. 



5. Formaldehyde injury varies inversely with the moisture content of the soil, 

 and directly as the concentration of the solution and the amount of such solution 

 applied per unit length of row. 



6. The grower may reduce the loss from formaldehyde to a very small percentage 

 and at the same time get better control of smut by changing his formula of applica- 

 tion according to the moisture conditions of the soil on the day when the seed is 

 planted. 



7. From the standpoints of (a) profitable (but not necessarily maximum) pre- 

 vention of smut, (b) minimizing the labor of drawing water, (c) ease of mi.xing the 

 solution, (d) maximum reduction of the weight of water which must be carried on 

 the drill, and (e) avoidance of severe chemical injury, we recommend the following 

 as the most practical method of application : 



// the soil is very dry, use the 1-50-5000 for inula ; if fairly moist, the 1-50-4000, 

 and if wet, the 1-50-3000 formxda. 



Or this might be expressed to the grower thus : 



Put one gallon of formaldehyde into a 50-gallon barrel and fill to the top with water. 

 If soil is very dry, apply at rate of one barrel to one acre of onions {IS inches apart). If 

 the soil is medium moist, apply l}/i barrels, and if wet and heavy, 1% barrels per acre. 



8. On a dry soil the amount of seed per acre should be increased. 



The Formaldehyde Tank for the Onion Drill 



The formaldehyde solution is applied to the row from a tank attached to the 

 seeder. Various kinds of tanks and methods of regulating the flow of the solution 

 from them have been used and described by previous investigators (5:161). 



In order that the control of smut be effective and injury reduced to the minimum 

 it is important (1) that the rate of flow be uniform and (2) that the operator have 

 some means of knowing just how much solution he is applying. These two require- 

 ments were doubly important for our experimental work. 



In order that the data obtained might be rehable, a machine was needed which 

 would distribute evenly and with a fairly high degree of accuracy any desired quan- 

 tity of solution on a stated length of row. For this purpose the regulating appara- 

 tus on all of the machines which have been described before was found to be worth- 



20 



