life-historij of Pericoma canescens {Psycho cUdse), 151 
main trachege rise from the prothoracic spiracles to tlie 
anal points without any intermediate lateral spiracles, as 
Perris has already stated, differing from Bouche. I found 
in the larva a pair of glands (ending in a filament at each 
end)^ lying loose among some elongate^, cylindric, white, 
fatty masses in the neighbourhood of the small intestines, 
nearly as in Tipula. These are rudiments of repro- 
ductive organs/^ (Hal.) 
In regard to the position of the spiracles, Haliday, in 
the above quoted passages, agrees with Perris (No. 4), 
and notices the errors of Bouche (No. 2), who describes 
and figures them as peripneustic. But, in the same 
passages, Haliday does not take notice of the structure of 
the posterior end of the larva figured by Perris ; it is 
truncate, and has no tail-like prolongation, and therefore 
cannot be a larva of PsycJwda, in the sense of Haliday. 
It is only one year later (" On some remaining blanks, 
etc.,^^ 1857, No. 11), that Halida}^, now better informed, 
says in a footnote (p. 182), Perris has figured a larva 
which undoubtedly belongs to some other family of 
Diptera.'''' The same doubt had already arisen in 
Erichson^s mind (No. 5), in 1840, the very year of 
Perris's publication, and in his "Bericht^^ he called 
attention to the discrepancy between Bouche and Perris; 
he said, " We can only assume that one of the observers 
had the wrong larva. Besides other not unimportant 
differences, the position of the spiracles in Perris is the 
same as in the larvse of the Mitscidse (Fliegenlarven), 
that is, there is one pair in fronts and one behind. 
Bouche's larva, on the contrary, has a pair of spiracles 
on each segment, just as the other Diptera in the division 
to which Psyclioda belongs.''^ In this case Brichson was 
mistaken ; he was thinking of the Mycetojphilidse and 
Gecidomyidse, which are peripneustic, and forgot that 
among the Nemocera there are several families the larvae 
of which are not peripneustic. It is not easy to guess 
where Perris' s larva belongs. It might be taken for that 
of a small Tipulid, but the larvae of this family are meta- 
pneustic, and do not show such a distinct anterior pair of 
spiracles as is represented in Perris's figure. (I cannot 
compare the figure now, but I possess a tracing which I 
made many years ago.) * 
* My statement in the BerL Ent. Zeits., 1892, p. 462, that the 
larva may belong to some Muscid, was a lapsus memorice 
produced by an indistinct recollection of Erichson's allusion to 
" FHegenlarve." 
