THE SYRPHIDAE OF OHIO 21 



anterior ones are moderately elevated, borne on a pair of horn-like prom- 

 inences which are capable of considerable extension but are usually 

 rather closely retracted. The posterior respiratory appendage is a most 

 remarkable and highly specialized organ which enables the larva to feed 

 at various depths beneath the water without coming to the surface for its 

 aerial respiration. The spiracles are situated distally on an elongated, 

 tube-like appendage which is extensile and retractile in a telescopic man- 

 ner. It is composed of three segments of different calibre, each double 

 in nature, enclosing two trachea, but fused mesad and always parellel, 

 never forked. These sections disappear one within the other when the 

 appendage shortens. 



The tip of the tube bears the spiracles and is ornamented with 

 structures which serve to keep it from being submerged, and are probably 

 homologous with the inter-spiracular ornamentation of the aphidophagous 

 type. 



A peculiarity of the larva of Eristalis spp. , which may or may not 

 be true of all "rat-tailed larvae" is the presence about the anal opening of 

 a group of soft, retractile, radiating flabellae or long flexible finger-like 

 ])rocesses, (Fig. 137). They may be entirely retracted and are at intervals 

 rapidly unfolded. Buckton suggests that they may have a renal function. 

 The larvae of Microdon are perhaps the most peculiar of all Syrph- 

 idae although they seem less aberrant from the aphidophagous form 

 than do the "rat-tailed" larvae. They are elliptical in outline, convex 

 above, flattened ventrally to make a flexible creeping-.sole. The integu- 

 ment is very tough dorsally, and this part of the body is often ornament- 

 ed with a reticulate pattern of papillae or setae. The margin of the 

 body around the creeping-sole is ornamented with a fringe-like arrange- 

 ment of spiny, hair-like processes. (Can these be migrated homologues 

 of the segmental spines?) According to Wheeler— "usually no traces 

 of segmentation are to be observed, but in some adult larvae of M. tristis 

 just before pupation, and after their upper surfaces had been dried by 

 exposure to the air, I have been able to discern in certain lights a dis- 

 tinct division of the body into seven or eight sub-equal .segments." 

 This is about the number of segments which is usually clearly visible in 

 both the "rat-tailed" and aphidophagous forms. 



The mouth parts in the larvae examined ma>- be understood by ref- 

 erence to PI. IX, Fig. 181. It will be seen that they are not directly 

 comparable to those of aphidophagous larvae. The structure at d. 

 is probably equivalent to the lower jaw of aphidophagous ones. It has ni 

 addition a flap-like piece at each side which gives the jaw a triple ap- 



