J. T. Wadswoeth 155 



the larvae of U. soIslitidJis derive the greater portion of their food from 

 the sap liberated by rupture of the cells which line the larval chamber. 

 If this assumption is correct, then the absence of frass might be ex- 

 plained, as in this case practically all the food would be absorbed. 



The evidence obtained leads to the conclusion that there is only 

 one brood produced each year. Pearly in September, 1912, I obtained 

 young larvae from flower-heads gathered at Prestatyn and then thought 

 that possibly two broods were produced in one season ; as I have never 

 found pupae or empty puparia during the months of August, September, 

 and October, the conclusion arrived at is that the young larvae obtained 

 in September, were the progeny of late-emerged flies and that there 

 is only one brood produced each year in this part of the country. Goureau 

 states that in France U. cuspidata gives rise to two broods in favourable 

 seasons. 



The life-history of Urophora solstitialis may be briefly summarised 

 as follows : 



Egg stage. 8-12 days, from end of June, during July and early 

 August. 



Larval stages. Feeding period three months or less ; from early 

 July to early October. 



Hibernating stage of larva. About seven to seven and a half months, 

 from early October to mid-May. 



Pupal stage. About 4-5 weeks. From mid-May to end of June. 



Imago. Length of life uncertain, probably at least a month. (I have 

 kept them alive for three weeks.) They emerge from mid-June to end 

 of July and early August. 



There is a certain amount of overlapping in the times given above 

 owing to the extended period of emergence of the imagines. 



Description of the mature larva. 



There is considerable variation in the size of the fully fed larvae ; 

 thus, three examples taken from one gall in November measured 

 2-8 X 1-5 ; 3-5 X 1-7 ; 4x2 mm. respectively ; the average size of 

 five large larvae was 4-4 x 2 mm. 



Naturally the variation in size is also exhibited by the pupae as will 

 be seen from the measurements of these given further on. A probable 

 explanation of this variation is that the smaller individuals develop 

 from eggs laid by late-emerged flies and consequently the feeding period 

 is much shorter than in the case of those which develop from eggs laid 



Ann. Biol. i. U 



