_ 
xl 
as M. Valéry Mayet conceives, to insufficient nutriment in the 
primitive stage, when the Colletes-egg has been partially tapped 
by other competitors—the ultimate metamorphosis is protracted 
until the autumn of the following year. 
In Sitaris humeralis, however, such retardation is the general 
rule; it being only in exceptional cases that some of these remain 
scarcely more than a single month in the pseudo-chrysalis state, 
completing their metamorphoses in August and emerging shortly 
after. But they usually hybernate in the former stage; and it is 
only in June of the second year that the interior quasi-larval form 
is separated from the pseudo-puparium, and about five weeks later 
becomes transformed to a true pupa-nymph—the same month, in 
fact, when the adult larva had assumed its corneous tegument in 
the previous year (Fabre, l.c., p. 3839—843). M. Valéry Mayet 
recognises this pupa as “la véritable nymphe” (p. 75); therefore 
the antecedent stage, or ‘‘troisi¢me larve” of Fabre—and not 
his ‘‘ pseudo-chrysalide”—can alone constitute the pseudo-pupa or 
“ nseudo-nymphe.” 
Thus the Sitaris humeralis usually requires two years to com- 
plete its metamorphoses, hybernating the first year in the primitive 
larval condition, and the second in that of the pseudo-chrysalis ; 
whereas the Sitaris Colletes, commencing its operations seven 
months earlier, generally attains maturity within a single year. 
The early transformations of two other species of Meloide have 
also been investigated by M. Jules Lichtenstein, of Montpellier, 
who succeeded in nurturing one of the primitive larvee of Meloé 
cicatricosus on the egg of a Vespa vulgaris placed upon honey in 
a glass tube, and in witnessing its first metamorphosis five days 
later, when it plunged into the honey, but died after feeding 
thereon twelve days. 
This secondary form differed essentially from that of Meloé 
described and figured by Fabre, apparently constituting an inter- 
mediate stage, closely resembling the antecedent larva, but 
destitute of caudal sete, with lacteous head and black eyes (the 
subsequent stage being blind), looking hke a minute salamander 
with its legs distended on the honey. 
Experiments were also tried with the primitive larvee of the 
blister-beetle (Cantharis vesicatoria), which could not be induced 
to feed on the eggs of Vespa or Polistes, nor on simple honey, 
beyond a few feeble attempts; but eventually they accepted the 
