118 Observations on the Habits o/Oscinella frit 



factors. Humidity will have but little influence in the case of the frit-fly 

 as the immature stages are passed in very close proximity to the plant 

 surface, where the humidity is high. 



It was expected that the correlation between prevalence and meteoro- 

 logical conditions would be more apparent than is indicated on the charts. 

 There would appear to be some relation between high prevalence and 

 high temperature, and emergence seems to follow periods of rainfall. 

 It is obvious however that insufficient data have been obtained and more 

 are being collected in an endeavour to settle the relationship. 



Graham-Smith (13) found that common Muscid flies were very sensitive 

 to meteorological changes, extremes in either direction killing off the 

 flies in large numbers. It has been pointed out above that on certain 

 days, namely July 14th, July 27th and August 7th the flies appeared to 

 be practically absent from the field, although conditions for collection 

 were not markedly unfavourable. It will be shown that, under control, 

 the flies may live for as long a period as 66 days (p. 124), both in spring 

 and summer, an interval much in excess of that between any two 

 maxima shown by the curves. The only conclusion is that longevity in 

 the field must be greatly curtailed by the action of undetermined en- 

 vironmental factors. 



The influence of relative preference and parasitism, on the prevalence 

 of the fly on any one crop, has yet to be determined. 



Addendum. 



The greater abundance of the fly in 1919 is very striking. In the 

 period July 5th to October 19th, 8319 flies were collected as against 

 3597 flies between April 28th and August 12th, 1920 (3657 flies being 

 collected in July 1919, but only 1391 in July 1920, in an equal number of 

 sweeps). 



The results of comparative sweeping on plots A, B and C {vide 

 Table I) seem to indicate that the fly normally remains close to the 

 locality of its emergence. This view is supported by the following figures 

 showing the percentages of unattacked plants on the various plots in 

 1919 and 1920. 



1919: A 17-0 1920: A and B 30-0 



B 71-3 F 49-2 



C 75-0 



B and C (1919) were drilled on March 28th and A (1919) on May 2nd 

 with the variety "Abundance." 



