233 



The decrease in egg production beginning in week 5 was attributed to the 

 beginning of an outbreak of fatty liver syndrome (R. H. Harms, personal 

 oommunioation) . Patent symptoms did not appear until several weeks 

 after the experiment ended. 



Analysis of individual strains by a t-test revealed a significant 

 difference in production in one strain due to treatment (Table 75). The 

 treated birds in strain k were producing eggs at a level slightly above 

 the average production level of the 12 strains combined. The production 

 level of the treated strain k hens represented a 3.67? increase in pro- 

 duction (p < 0.0137) over the untreated strain k hens. As can be seen 

 in Table 77, and other tables previously mentioned, strain k hens main- 

 tained low mite populations throughout the trial. Strain k was also 

 one of the few strains to exhibit a natural decrease in mite population 

 at the end of the trial (Table 77). 



In an effort to correlate production data with effects caused by 

 the previously mentioned pyrethrin sprayer, each quarter of house 200 

 was analyzed as a different treatment, i.e. treatments 1 through 4. 

 Treatment means (treatments 1 through h) shown in Table 76 were not 

 significantly different and again, any effects of the pyrethrin sprayer 

 were discounted. 



In an attempt to show a correlation between mite numbers and egg 

 production, a plot of egg production means versus mite means by strain 

 was made (Figure ^5) • According to this graph, there appears to be no 

 correlation. Birds with mite populations of ca. 640 or ca. 2320 mites 

 per bird were both laying at the 81 percentile. 



However, it appears that it is not mite numbers alone that affect 

 production, but a combination of mite numbers and time (Table 77). 



