A CONTRIBUTION TO THE CLASSIFICATION OF THE LEPIDOPTERA.

In this place I purpose to state briefly the conclusions that I have reached regarding the phylogeny of the families of the Lepidoptera. These conclusions are the results of an effort to read the record of the action of natural selection as recorded in the wings of these insects. Owing to the limited time at my disposal, but little attention has been given to the evidence presented by other parts of the body; and for the same reason I have been able to study the Tineids, Tortricids, and Pyralids hardly at all. The following classification is, therefore, merely a provisional one; and is put forth chiefly as a record of the results that I have obtained up to this time in applying the method outlined in the preceding pages.

I confidently expect, however, that the principal conclusions stated here will be confirmed by a study of other parts of the body; for in Nature's court the testimony of different witnesses if rightly understood will agree. If any of the conclusions should prove to be incorrect, the fault will be found to lie with the translator and not in the record.

The fullness of the discussion that has already been given of the ways in which wings are modified will warrant considerable condensation in the following outline. I will first indicate the relations of the proposed divisions to each other by means of a table; and will afterwards give fuller characterization of these divisions.

TABLE OF PROPOSED DIVISIONS OF THE LEPIDOPTERA.

A. Suborder JUGATAE.
B. The Macrojugatae Family HEPIALIDAE.
BB. The Microjugatae. . Family MICROPTERYGIDAE.
AA. Suborder FRENATAE.
B. The Microfrenatae.
C. The Tineids. . . Superfamily TINElNA.
CC. The Tortricids. . Superfamily TORTRICINA.

CCC. The Pyralids. . Superfamily PYRALIDINA.

BB. The Macrafrenatae.

C. The Frenulum-conservers.

D. Moths in which the reduction of the anal area of the hind wings precedes the reduction of the anal area of the fore wings. This group is not represented in the North American fauna. Castnia (Fig. 15) will serve as an illustration.

DD. Moths in which the reduction of the anal area of the fore wings precedes the reduction of the anal area of,the hind wings.

F. Moths in which a great reduction of the sub-costal cell of the hind wings is taking place.

G. Moths in which the anal veins of the fore wings anastomose so as to appear to be branched outwardly. (Fig. 25.)

Family MEGALOPYGIDAE.

GG. Moths in which the anal veins do not anastomose in such a way as to appear branched outwardly.

Superfamily ZYGAENINA (in part).

FF. Moths in which the subcostal cell of the hind wings is not greatly reduced.

G. Moths in which the anal veins of the fore wings anastomose so as to appear to be branched outwardly. (Fig. 22.)

Family PSYCHIDAE.

GG. Moths in which the anal veins do not

anastomose in such a way as to appear
branched outwardly.
H. Family COSSIDAE.
HH. Family LIMACODIDAE.

EE. The Specialized Frenulum-conservers.

F. DIOPTIDAE.

FF. The Geometro-Bombycids and the Geometrids.

G. Family NOTODONTIDAE
GG. Family BREPHIDAE.
GGG. Family GEOMETRIDAE.

FFF. The Noctuo-Bombycids and the Noctuids.

G. Family CYMATOPHORIDAE.
GG. Family NOCTUIDAE.

Family LIPARIDAE.
Family AGARISTIDAE.
Family ARCTIIDAE.

FFFF. Isolated Families of Specialized Frenulum-conservers.

G. Family SESIIDAE.
GG. Family THYRIDIDAE.
GGG. Family SPHINGIDAE.
GGGG. Superfamily ZYGAENINA.

CC. The Frenulum-losers.

D. The Frenulum-losing Moths.

E. Moths in which cubitus is apparently three-

branched. Superfamily SATURNIINA.

EE. Moths in which cubitus is apparently four-branched.

F Family DREPANIDAE.
FF. Family LASIOCAMPIDAE.

DD. The Skippers. "Butterflies" in which all of the branches of radius of the fore wings arise from the

discal cell. Family HESPERIDAE. DDD. The Butterflies.--Butterflies in which some of

the branches of radius coalesce beyond the apex of the discal cell.

E. Butterflies in which cubitus of the fore wings is

apparently four-branched. Family PAPILIONIDAE.

Evolution and Taxonomy 99

EE. Butterflies in which cubitus is apparently three-branched.

F. Butterflies exhibiting no tendency to abortion

of the fore legs. Family PIERIDAE. FF. Butterflies exhibiting a marked tendency to abortion of the fore legs.

G. Family LYCAENIDAE.
GG. Family NYMPHALIDAE.

This suborder includes those moths in which the two wings of each side are united by a membranous lobe, the jugum, borne at the base of the inner margin of the fore wings (Fig. 27, j), and in which the anal area of the hind wings is reduced while the radial area is not. The most available recognition character is the similarity in venation of the two pairs of wings; radius being five-branched in the hind wings as well as in the fore wings.

Moths of medium or large size. The mouth-parts are aborted, and correlated with this there persists a comparatively generalized condition of the wings, which is shown by the absence of a jugum plate. The larvae are wood-borers. This division is represented by a single family.

Family HEPIALIDAE.

BB. THE MICROJUGATAE.

Moths of minute size. Mouth mandibulate, with both mandibles and maxillae fitted for mastication. This is doubtless the most generalized form of mouth-parts preserved in this order. Correlated with the presence of functional mouth-parts, these moths show a higher specialization of wing structure than exists in the Hepialidae; there being a plate-like organ at the base of the costa of the hind wings, the jugum plate, and a series of spines; both of which act with the jugum in assuring the synchronous action of the two pairs of