F. W. Dry 89 



These differences in abundance of the Midge of the different broods 

 and in different localities stand out clearly enough. In the following 

 section I try to explain them. 



V. Explanation of results. 



(a) Conditions determining the percentage of plants attacked by flies 

 of Brood I of the Swede Midge in a given field. 



Apart from conditions affecting the whole of any particular Swede 

 Midge District in any particular year the following conditions are 

 important in determining the percentage of plants attacked by Brood I 

 of the Midge in any given field. 



(1) Date of sowing. 



Early sown fields are attacked worse than late sown fields. This 

 is shown in Tables 2 and 3. As already explained singling is doubtless 

 in part responsible for the figures collected in June sown fields being 

 lower than those from May sown fields. But this explanation will not 

 account for the complete absence of attacked plants from many June 

 sown fields, which I frequently found to be the case, for swede hoers are 

 almost all unaware of the existence of the Midge. The Midge does not 

 infect the plants until they get into rough leaf. The flies of the first 

 brood appear in June, a few sometimes in July. Consequently early 

 sown fields are exposed longer to attack than late sown fields. 



(2) Distance from a swede field of the previous year, and 



(3) Size of field. 



In 1913 and 1914 I found that numerous small plots of early 

 sown swedes next to swede fields of the previous year had percentages 

 of infected plants high compared with the percentages in other fields 

 in the same district. The following are examples of such 1913 fields: 



% CL Highest other % CL in district 



A. District IX G ^ 



B. ,, 11 ^3 



C. „ 30 J 



D. District VII 19 \ 



E. 24 J 8 



The graph, Fig. 5, is of interest in this connection. In this graph 

 the % CL is plotted for different rows, of which the numbers are given, 

 in a 1913 swede field next to a 1912 swede field. The left side of the 

 graph represents the side of the 1913 field near to the 1912 field, the 

 right side the side of the field remote from that field. Similar graphs 



