88 Psyche [June 
unknown, while arthemis (lamina) is abundant, and always, so 
far as observation has shown, breeds true. 
Edwards (1877, 1879) drew from his observations the conclusion 
that arthemis (lamina) and proserpina were to be referred to a 
single dimorphic species, flying in company with astyanaz along 
the narrow zone where their ranges overlapped,— indeed dimorphic 
only in that zone,— yet never interbreeding with the other species. 
Mendel’s work lay buried and forgotten; and no one realized that 
this dimorphism might under certain circumstances be a criterion 
of hybridity. The occurrence of other apparently hybrid Basil- 
archias (astyanaz-archippus and arthemis-archippus) has been 
recorded (see Scudder, 1889, and Field, 1904), but its full mean- 
ing seems to have been overlooked. The cumulative significance 
of the various published observations of the genus Basilarchia in 
the eastern United States is contrary to Edwards’s interpretation. 
Our working hypothesis may now be that proserpina is a hybrid 
between arthemis and astyanaz, in which the dark coloring of 
astyanax incompletely dominates the white band of arthemis.t. In 
the narrow belt in which the hybrids commonly occur, these hetero- 
zygous individuals must often breed together, producing offspring 
of which 50 per cent. must resemble the parents (i.e., are hetero- 
zygotes), while 25 per cent. are pure dominants (astyanax) and 
25 per cent. are pure recessives (arthemis).* | Farther north, 
where astyanaz seems not to thrive, but the recessive white-banded 
arthemis holds sway, occasional stray examples of proserpina, 
mating with arthemis, will yield offspring of which 50 per cent. 
will be proserpina and 50 per cent. pure arthemis. In this division 
the Springfield brood probably belongs. South of the zone of 
hybridization, the white band must be almost swamped; for when 
proserpina mates with astyanaz, the offspring will all be dark, and 
half of them will be pure dominants (astyanar). The occasional 
white-banded Basilarchias*® taken on Long Island or in New Jersey, 
or in other places south of the usual range of arthemis, may be 


1Such incomplete dominance is a widely-recognized phenomenon in Mendelian 
inheritance. See Bateson (1909) and Davenport (1910). 
2The name lamina now appears to be superfluous, as we are assuming that 
there is but one form of arthemis. 
Ursula [=astyanagv] var. albofasciata Newcomb (1907). 
