112 Observations on the Habits o/Oscinella frit 



Again the first brood (Generation I) gave rise to adult flies of Genera- 

 tion II after 33-4 days. C'—D, C'~E and D~F marked on Chart II, 

 p. 115, represent these periods respectively. 



Oviposition probably took place soon after 15. 5. 20 and again soon 

 after 17. 6. 20 (the parents being in constant association with plants) 

 as the issue are grouped over two periods, namely, after 50 days (with 

 a maximum of 63 days) and 79 days (with a minimum of 69 days). 

 Therefore the average interval extends to 29 days, which is most prob- 

 ably the interval between the production of separate broods of flies of 

 Generation I. The interval between the emergence of the last fly of the 

 first group and the first fiy of the second group is 6 days. 



Discussion of Data. 

 Significance of Prevalence. 



The term "prevalence" is used to denote the presence of flies in the 

 field without distinction of brood or generation. Prevalence and emer- 

 gence are not synonymous terms, as, for a particular brood or generation, 

 the period of prevalence will exceed the period of emergence, by the 

 longevity. 



In Charts I and II, pp. 114, 115, will be found smoothed^ curves re- 

 presenting the prevalence of frit-fly in the latter part of 1919 and in 1920 

 respectively. Rainfall, maximum shade temperature and minimum grass 

 temperature, also smoothed, are included^. 



The maximum temperature^ in the sun has a direct influence on the 

 apparent prevalence at different hours of the day, as determined by 

 sweeping, but a much less effect than the more even maximum shade 

 temperature on the prevalence from day to day. Usually frit-flies are 



^ The curves have been smoothed as exemplified here. Taking, for example, the 

 readings for June 2-8, 1920 — the adopted means for June 4 and June 6 are the means of 

 readings for the days 2^-6 and 4-6-8 respectively. 



The corresponding means of meteorological data were calculated from the daily readings 

 of the periods 2-6 and 4-8 respectively, because the weather conditions over the whole 

 period would control the total emergence and therefore the prevalence. 



^ I have to thank Dr Rambaut (the Radcliffe Observer) for these figures. This station is 

 separated from the farm by 3^ miles, both having about the same altitude. The error in 

 the farm observations no doubt greatly exceeds the error introduced by utilising the 

 meteorological records of this Station. Self-recording instruments were not available for 

 use on the farm. 



^ Graham-Smith (14) showed that curves for common Muscid flies, caught in traps in 

 the open air, corresponded most clearly to the curve for the maximum temperature in the 

 sun. In this case, however, the number of flies trapped was proportional to the activity 

 of the fly and was not necessarily a measure of the prevalence of the flies at any one time. 



