156 Knajnceed Gall-flu 



earlier in the season ; it is also possible that where several larvae, say 

 eight to ten are feeding together in one flower-head, there is not sufficient 

 nourishment to enable them all to reach a large size. 



The larvae are ellipsoidal in shape ; the anterior end is bluntly 

 pointed and the posterior end sharply truncate, and the greatest width 

 is across the anterior third of the body (Fig. 9). With the exception 

 of a very few extremely small hairs on the posterior surface of the last 

 segment, the body of the larva is completely smooth and free from 

 bristles or spines ; Bezzi (1913) in his very useful account of the Try- 

 paneidae states that " the under surface [of Trypaneid larvae] bears 

 transverse rows of small black spines directed backwards," and further, 

 that " the anal end is somewhat impressed, contoured by a variable 

 number of fleshy points or tubercles, some of which bear also chitinous 

 spines." Neither of these characters is present in the fully grown 

 larvae of this species, nor according to Mik in those of JJ . cardui ; in 

 his description of the larvae of this species he says they showed " keine 

 Spur von Dornchen " and " die Larve ist also vollig kahl, glatt " ; as 

 far as can be judged from the descriptions of the larvae of U. cuspidata 

 and U . quadrifasciata given by Goureau and Dufour, the larvae of these 

 species possess neither spines nor anal tubercles. The larvae of these 

 four species differ therefore from the majority of Trypaneid larvae in 

 these two features, which, according to Bezzi, are possessed by Try- 

 paneid larvae in general. 



In colour the fully grown larva is pale creamy white, except at the 

 ends ; the two anterior segments are reddish brown, and the posterior 

 segment is pale yellow round its anterior border, merging into a very 

 dark chestnut-brown, almost black, colour on its posterior surface ; 

 the two areas surrounding the spiracles are lighter in colour. 



The segmentation of the larva is well defined except at the anterior 

 end ; Bezzi states that the number of segments in Trypaneid larvae 

 generally is usually fourteen. I have been able to distinguish only 

 thirteen segments in these larvae, and Goureau regarded the larvae 

 of U . cuspidata as probably possessing twelve segments, excluding the 

 head ; there may be a fusion of two segments in the cephalic region 

 of these Urophora larvae. On the ventral surface of the middle segments 

 there are indications of creeping-pads (Kriechschwielen). 



The mouth appears as a slit whose long axis is dorso-ventral ; the 

 antennae and sense-organs are difficult to make out on fully grown 

 larvae, in fact I have never been able to make them out satisfactorily ; 

 probably they are retracted into the body together with the cephalo- 



