278 Mr. C. J. Gahan on larvx of Tridenotoma cMldrcni, 



characters in common with the larvae of PythidiB, Pjjro- 

 chroidtX and (UdemcridcV, and the chief differences, which 

 are to be found in the form of the 8th and 9th segments of 

 the abdomen, are of only secondary importance and such 

 as might be found to occur between different genera of the 

 same family. The presence of scabrous areas on the ter- 

 gites and sternites of several segments in Tridenotoma is 

 a character wanting in the larvae of PytliidcV and Pyro- 

 chro'id/i', but seems to have a correspondence with the 

 scabrous elevations or tubercles occurring on certain of the 

 segments in the larvse of OEdemcridiv. 



On the whole, then, it seems to me that the Tridcnoto- 

 midm, in a phylogenetic system such as Ganglbaucr's, 

 should be placed not at the end, but at, or very near, the 

 beginning of the Heteromerous series. Although the 

 general appearance of the imago seems very unlike that 

 of the Fyrocliroidie or GEdemerido} or the other families 

 allied to these, yet it must be remembered that in one 

 very important structural character, namely the open 

 anterior coxal cavities, the TridcnotomidcV agree with these 

 families and differ from the Tencbrionid/v. 



2. The larva of Mdittomma insidarc, Fairm. -' 



Mdittomma insularc, Fairm., is a species o? ZymexyHd/B 

 that was first described in 1893 (" Bull. Soc. Ent. Fr." 1893, 

 p. cccxxiii). It occurs in the Seychelles Islands, and 

 differs considerably from other species of the genus McliU- 

 omma in one character which appears to have been over- 

 looked by the original describer ; the eyes, instead of 

 being widely separated from one another on the ventral 

 side of the head, are very closely a23proximated and almost 

 contiguous. Some larvffi of this sjDecies that were found 

 living in the stem of the coco-nut palm in the Seychelles, 

 were sent a few years ago to the Director of the Imperial 

 Institute, and by him were presented to the British 

 Museum, with a request for their identification. They 

 were accompanied by examples of the perfect insect. 

 Although no pupoe have been received, there can be no 

 doubt whatever that the larvae belong to the species indi- 

 cated. These larva3 are of different sizes, varying in length 

 from about 10 to 30 millim. ; and present a very strong 

 family resemblance to the other known larvae of Lymcxy- 

 lidfB, the only noteworthy difference consisting in the 



