SPECIES HYBRID.                                      117

There is no reason to suppose that the subfamily is not present in
the other Pacific islands, but it has not been recorded from them.*
With respect even to the Hawaiian Islands, Perkins (Fauna Haw., 1,
clxxxix) says "Not less than 250 species [of Drosophilinse] must exist
in the islands, and very probably double that number may occur."
The number and variety of Drosophilinse form the most striking peculi¬
arity of the dipterous fauna of these islands. Nearly one-third of the
species of Diptera considered by Perkins to be endemic in the islands
belong to this subfamily. There are a number of large and strikingly
colored species of true Drosophila, as I have convinced myself by an
examination of material in the U. S. National Museum. Besides the
two endemic genera mentioned above, there is a genus described by
Grimshaw under the name Hypenomyia that is perhaps distinct, but
is here tentatively placed as a synonym of Drosophila, in the absence
of a satisfactory description.

XIII. SPECIES HYBRID.

Drosophila melanogaster and D. simulans are the only two species
of Drosophilinse that have been hybridized (see p. 14). The two
species are extremely similar in appearance, and there is every reason
for believing that they are closely related. This fact makes the
hybrids less interesting for study than would be the offspring of two
quite diverse species. In addition, the hybrids so far obtained have
all been completely sterile, so that their genetic behavior can not be
studied. Nevertheless, there are several facts of considerable interest
in connection with the cross.

In general, the hybrids are intermediate in appearance between the
two parent species. This is true of the relative size of the eyes and
width of the cheeks and of the male genitalia. The other differences
between the species, except the length and shape of the egg-filaments,
are too indefinite for exact study; and the hybrid females do not lay
eggs, so that this character can not be observed. The hybrids differ
from both parent species in several respects, as follows:

1.  They are completely sterile, and have poorly developed gonads.

2.  The abdominal plates often are irregular and have narrower and more irregular dark

bands than are usual.

3.  Some of the thoracic bristles are usually missing.

4.  The wings are apt not to unfold properly, or to be somewhat less convex on the anterior

margin than is usual.

5.  The cross-veins are often broken or missing.

The most peculiar thing about these hybrids, however, is the distri¬
bution of the sexes.    The crosses usually result as follows:

D. melanogaster 9  X D. simulans cf = hybrid 9 9 only.

D. simulans  9 X D. melanogaster c? = hybrid cf cf; occasionaUy a few hybrid  9 9.

If the D. melanogaster females bear a Y chromosome (XXY), and
therefore give non-disjunctional exceptions, it  is found  that they

* I had overlooked the fact that Jepson (1917, Ann. Rept. Div. Ent., Dept. Agric. Fiji, 1916,
16) records Drosophila melanogaster from Fiji.