THE MYCETOPHILID^. 97 



curious structure formed of their own excrements, and carried on 

 their backs, one such structure, resembling an Ancy/us-sheW. This, 

 has been observed by Perris in the larvte of MycetopJiila scatophora, 

 and by Brauer in a species from Brazil. The larvae, as a rule, spin 

 a cocoon, but this is not always the case. Bolitophila, according ta 

 Dufour, has no cocoon ; neither have certain AlycetophilcB. Others 

 {Sciara fuscipes) have an earthy cocoon. The general rule, however, 

 is that some kind of case is formed for the pupae, which are found, 

 as a rule, in fungi.* Some species of Sciara are also met with in 

 cow-dung. The pupa (Fig. 15) is smooth, the angles being rounded^ 

 not sharp, as in the Tipulidse. (The pupas of Sciara are somewhat 

 different, and strongly resemble those of the Cecidomyice.) The larval 

 skin is not retained in this stage, as is done in many cases. The 

 legs are applied to the breast and venter ; the antennae bent round 

 the eyes, and their remaining portion between the legs and wings. In 



Fig. 15. — Pupa of .Sr/rtriz. 



Sciara they may be expanded into a tooth at the base. In the same 

 genus the prothoracic stigmata is placed on an elevation, which may 

 extend into a horn. An air-tube may be seen entering it. Stigmata 

 on abdomen plainly marked. 



There are certain interesting points about the larvae of Myceto- 

 philidcc. One, is their curious habit of congregating into large bodies 

 for migration; the other the brilliant luminous nature of certain 

 larvfe in this family. The former we shall deal with when we go 

 more fully into the genus Sciara, whilst now we will look at the 

 remarkable phenomenon of luminosity shown by certain species in- 

 cluded here. The cases of larvae emitting a brilliant light are few in 

 the Diptera. This phosphorescent character has been observed in 



* Mr. Dale tells me "some larvse spin a web on the outside of a fungus, and 

 turn into pup^ somewhat after the manner of the small Eggar Moth i^A. lanestris). " 

 This would be probably in the genus Sciophila. 



7 



