Station Farm Experiment, 1919 



The experiment on the Cowls farm was duphcated as regards treatments, but 

 on a smaller scale, on a badly infested plot on the Station farm. 



The check rows gradually dwindled throughout the summer while the stand on 

 the treated rows remained excellent. No differences between the results obtained 

 by the use of any of the three formulas could be observed. Through a mistake, the 

 onions of this plot were harvested in the absence of the writers and final data on 

 yield were not obtained. Unquestionably, however, they would have paralleled the 

 yield data for the Cowls farm. 



Greenhouse Experiments, 1919-20 



During the winter a greenhouse bench 4 x 16 feet was filled with soil from a 

 badly infested field. Three crops of onions, raised in succession, were treated with' 

 formaldehyde of different formulas. Commercial formaldehyde was applied at the 

 rate of one pint to 2400, 2800, 3000 and 3200 feet of row. Dilutions from 1-8 up to 

 1-128 were used. Without presenting all the data, we may state the tentative 

 conclusions at which we arrived as a basis for the field experiments of the following 

 year: 



1. The percentage of control was not appreciably greater when one pint of 

 commercial formaldehyde was apphed to 2400 or 2800 feet of row than when applied 

 to 3000 or 3200 feet. 



2. Concentrated formulas such as 1-48-2400, 1-32-2800 and stronger cause 

 serious injury and cannot be used. 



3. The most important indication of the series was that within wide hmits, the 

 amount of dilution is not a factor of any great importance as jar as control of smut is 

 concerned. 



C. A. Clark Farm Experiment, 1920 



This experiment was on land in Sunderland where onions had been grown for 40 

 or 50 years. The soil, however, was not as thoroughly infested as that on the Cowls 

 farm used during the previous season. The part selected for experiment included 

 30 rows, 267 feet long, 12 inches apart, seeded at the rate of 7 pounds per acre. The 

 plots were seeded May 3, the soil at that time being wet and heavy after a backward 



inaccurate because it makes no distinction between those which are badly affected and will die and those 

 which are slightly affected and may produce good onions. Also we have found that it is impossible to dis- 

 tinguish all the diseased plants because many of the lesions are below the surface of the soil. 



2. By counting the percentage of smutted plants among those which are standing at some definite 

 period in the early summer, but after the cotyledon stage. This was done in the Cowls farm experiment just 

 described and is commonly used by other workers on onion smut. This method is worthless because it does 

 not count those which have already died from the disease and disappeared. We have repeatedly noted that 

 seedlings begin to die within three weeks after planting ond the mortality is then heavy for a few weeks. 

 When a young plant dies it shrivels and decomposes so quickly that no trace of it can be found after a very 

 few days. Obviously such plants cannot be taken into consideration in calculating by this method 



3. By removing and counting the plants as fast as they succumb to s7nut. This requires a thorough in- 

 spection about twice a week for a period of some four weeks, after which all diseased plants may be pulled 

 since it is certain that they will not recover. This is certainly the most accurate method, but under field 

 conditions involves so much labor that it is precluded. We have, however, used it in greenhouse experi- 

 ments with satisfaction. 



4. By counting and comparing the number of healthy bulbs in the row at harvest time or shortly before. 

 This does not assume that all onions which have disappeared from the rows during the summer have suc- 

 cumbed to smut, but that all other agencies which would destroy the plants previous to this time would 

 operate ecjually on all the rows. Therefore, the benefit derived by the grower can be determined by direct 

 comparison of the figures thus obtained. From the experimenter's standpoint it is not quite satisfactory 

 because it does not take into consideration the number of plants which were prevented from germinating 

 on account of the use of formaldehyde of a certain formula. In the experiments of the last season we have 

 determined the amount of such injury by comparing the number of plants which came up and then comput- 

 ing the control on the basis of those which remain. This fourth method has been used throughout the ex- 

 periments here as being the most nearly accurate practical method of compaiing the results obtained by a 

 given treatment. 



5 By comparing the yield of onions in pounds or bushels. This method is inaccurate because, as every 

 onion grower has observed, the onions in a thin row (such as they are when thinned by smut) being further 

 apart, are larger than in a thick row; therefore, bulb for bulb, are heavier. This difference, however, rep- 

 resents the gain in weight secured by growing bulbs further apart, and in no case could be interpreted as 

 having any relation to the application of formaldehyde. If larger bulbs are desired, the result could be 

 much more economically obtained by planting a smaller quantity of seed per acre while using the formalde- 

 hyde. This fifth method was used by the writers during the field experiments of 1919 and 1920 and the 

 final weights are included in the tables, but they should not be considered as accurately gauging the results 

 of the experiments. 



15 



