18 CIRCULAR 926, U. S. DEPARTMENT OF AGRICULTURE 



and were asked to smell vials of the insecticides so that they might 

 be identified if they caused much off-flavor. With the treated 

 samples duplicate samples from the untreated plots were pre- 

 sented, one labeled as a control and the other coded. The judges 

 were asked to score each coded sample with numbers ranging from 

 7, for no off-flavor, to 1, for much off-flavor. The lima beans were 

 boiled 20 minutes and drained before being served. The tomatoes 

 were sliced and served raw. The potatoes were peeled, quartered, 

 steamed 25 minutes, mixed with one-sixth their weight of water, 

 beaten 3 minutes in an electric mixer, and served. 



A statistical analysis of the scores for lima beans and tomatoes 

 showed no differences significant at the 5-percent level. 



Since the scores for the potato samples showed differences, they 

 were appraised a second time. An analysis of variance of the 

 pooled data from the two appraisals gave the following results : 



Aldrin 5.89 



Chlordane 5.91 



Toxaphene 5.75 



Untreated control 6.40 



Least significant difference: 



At P = 0.05 0.36 



At P = 0.01 0.48 



The interaction of treatments with judges was not significant at 

 P = 0.05. Thus all samples from plants taken from treated plots 

 were found to have more off -flavor than the controls, under the 

 conditions of the experiment, the differences being significant at 

 the 1-percent level of probability. No judge indicated that the 

 flavors were similar to those of any of the insecticides. 



These findings do not mean that the average consumer, or even 

 one having an unusually acute sense of taste, would necessarily 

 detect an off-flavor in treated potatoes as they would ordinarily 

 be served at the table. They do show, however, that certain people 

 after training were able to detect differences in flavor when they 

 directly compared the treated samples with the control, under 

 highly favorable conditions. The judges also reported that all 

 samples, including controls, had an earthy or bitter taste. Off- 

 flavors due to insecticides might be more conspicuous in potatoes 

 with a blander natural flavor. 



SUMMARY 



Studies to determine the effect of annual soil applications of 

 toxaphene and chlordane at 20 pounds, aldrin at 4 pounds, and 

 ethylene dibromide at 3 gallons per acre on vegetable crops were 

 begun at Oxnard, Calif., in March 1949. Each of these treatments 

 and an untreated check were tested in six randomized blocks on 

 15 kinds of vegetables for 3 years. 



The growth and yield of the following crops were not affected 

 by any of these treatments: Tomatoes, peppers, cabbage, cauli- 

 flower, spinach, lettuce, beets, peas, sweetpotatoes, and the Ven- 

 tura lima beans. 



Following the second or third application of toxaphene there 



