are
collected in April, boiled, wrapped in the epidermis of the agave, sold
on the streets of Mexico, and are eaten with avidity. To all appearances
they are nourishing and palatable, and it is said that connoisseurs
prefer them to oysters or swallows' nests." Hough believes "that the
discovery of the sap-yielding quality of the agave was through search
for these larvae."

In the Nuttall Codex occur numerous representations of insects, some of
which appear to represent butterflies or moths (Pl. 3, figs. 5-8) but
these are quite unidentifiable. That shown in fig. 6 is colored blue in
the original, while the others are of various colors. Possibly the round
markings on the wings in figs. 5, 8, represent the ocelli on the wings
of certain species of moths. In this connection, too, it is interesting
to compare the conventionalized butterfly with its single eye and
pointed antennae from the Aubin manuscript (Pl. 3, fig. 9) with one
drawn on the same plan from the Nuttall Codex (Pl. 3, fig. 8).

MYRIAPODA

Representations of a centipede (probably a species of _Scolopendra_)
occur in the Dresden Codex and in several others examined. That shown in
Pl. 5, fig. 1, from the Vaticanus 3773, is perhaps the least
conventionalized.[303-*] This figure appears partly to encircle a
temple, behind which the major portion of its length is hidden and hence
is not here shown. The bipartite structure coming from the animal's head
doubtless represents the mouthparts, and at its base on either side
arise antennae. The first pair only of legs is shown with a pinching
claw, possibly intended as a conventionalized hand, while the rest are
simple. The plumes decorating the posterior extremity are of course
extraneous and represent the tail of the quetzal or trogon.

In the Dresden Codex, god D constantly appears in connection with a
head-dress from which depends a centipede, greatly reduced and
conventionalized. Two forms of this centipede are shown in Pl. 3, figs.
15, 18. The body appears to consist of four or five segments each with
its pair of ambulatory appendages (though there may not always be the
same number of each) terminated by a circular segment with a
conventionalized three-knobbed structure, apparently corresponding to
the portion that bears the quetzal plume in Pl. 5, fig. 1. The outline
of the head in Pl. 3, fig. 15, is shown in dotted line but by solid line
in fig. 18. One of the antennae appears to be omitted from the former