218 



Resul ts. With the exceptions of strains 8 and 9, Ravap reduced 

 mite populations to zero with two applications, as shown in Table 63. 

 Mite numbers were reduced significantly (p < 0.0001) in all cases. 



Although there were no significant differences between mite popu- 

 lation means of treatment groups prior to treatment, significant 

 differences between strains did occur (Table 64) . This was due to the 

 wide variation in mite numbers found on strain replications in the four 

 quarters of house 200. Table 65 shows a significant difference between 

 post-treatment group mite means. Note that the mite populations in the 

 untreated group had doubled since the pretreatment count. Also shown is 

 the ranking of mite means by strain. This is essentially a ranking of 

 means from the untreated groups, since the mite populations in the 

 treated groups had been virtually eliminated. Means shown in Table 65 

 were reduced by a factor of 2 since both treatments were considered in 

 their computation. 



In Table 66, pre- and post-treatment mite counts in the control 

 group were added together and the means computed. This table gives some 

 indication of strain resistance to mite populations. The strains, as 

 ranked here, either maintained high mite populations throughout the 

 trial or had rapid increases in population numbers during the trial. 



Upon our arrival in Chipley, Florida, to begin the pretreatment 

 mite count, it was noted that the pyrethrin fly control system previously 

 mentioned was in partial operation on the north side of house 200. 

 Since one group from each treatment was on the north side of the house, 

 the house was divided into east and west halves and both halves were 

 analyzed separately. Tables 67 and 68 show the pretreatment mite counts 

 by treatment and strain for the west and east ends respectively of house 



