Marianas Mallard-— YAMASHINA 
123 
because Anas oustaleti resembles Anas platyr- 
hynchos, especially in the color of the speculum. 
However, this assumption seems hardly to fit the 
case, because it overlooks the remarkable simi¬ 
larity of the poecilorhyncha type to Anas poeci- 
lorhyncha superciliosa. Furthermore, the specu¬ 
lum of Anas oustaleti does not always resemble 
that of Anas platyrhynchos, being sometimes 
dark green without conspicuous white edgings, 
and hence more like that of Anas poecilorhyn¬ 
cha superciliosa. 
On the other hand, it is well known that 
Anas platyrhynchos can easily be crossed with 
Anas poecilorhyncha (including superciliosa ), 
producing offspring very similar to the platy¬ 
rhynchos type of Anas oustaleti. In the male of 
F 1 offspring of this cross in nuptial plumage 
the head is green except on the sides, where 
buff feathers are mixed plentifully, with a dark 
stripe through the eye. The mantle is as that of 
Anas poecilorhyncha, but breast, sides of body, 
wing, and tail are similar to those of Anas 
platyrhynchos, the central tail feathers curling 
backward. The female of this cross is inter¬ 
mediate between the parental species, showing 
the dark stripes on the cheeks found only in 
Anas poecilorhyncha. A particularly interest¬ 
ing fact is that the speculum of Anas platyrhyn¬ 
chos acts as dominant over that of Anas poe¬ 
cilorhyncha. The platyrhynchos type speculum 
in Anas oustaleti might be derived in this way, 
and the occasional occurrence of the poecilo¬ 
rhyncha type speculum may be explained by 
the formation of a homozygous condition gov¬ 
erning the recessive gene of the poecilorhyncha 
type speculum. From these evidences we may 
deduce that Anas oustaleti originated from 
hybridization between a local race of Anas 
poecilorhyncha, probably formerly resident on 
the Marianas, and Anas platyrhynchos stock 
which straggled there occasionally from the 
north. The- scarcity of the platyrhynchos type in 
the southern islands strengthens this supposi¬ 
tion. The opportunity for hybridization should 
occur more rarely in the south, and thus more 
frequent back-crossing of the hybrid with the 
indigenous Anas poecilorhyncha on Tinian and 
Guam explains the superabundance there of the 
poecilorhyncha type. As the hybridization 
should have taken place more frequently to the 
north in Saipan, the ratio of occurrence of the 
platyrhynchos type is logically higher there. 
There is no evidence of previous occurrence 
of Anas poecilorhyncha (including superciliosa) 
in the Marianas Islands, but it seems logical, 
because Anas poecilorhyncha is found in the 
neighboring Caroline and Palau Islands, as well 
as in the Philippines and Japan. Neither is 
Anas platyrhynchos recorded from the Marianas, 
but its winter range reaches in the east to the 
Hawaiian Islands and in the west to Borneo 
and South India, and it winters frequently just 
north of the Marianas in the Bonin and Volcano 
Islands. Therefore, it is not unreasonable to 
suppose that Anas platyrhynchos has occasion¬ 
ally straggled to the Marianas, remained and 
produced hybrids with an Anas poecilorhyncha 
stock which was formerly indigenous there. 2 
According to recent cytogenetical investiga¬ 
tions, the interspecies hybrid shows more or less 
sterility due to a dissimilarity of parental chro¬ 
mosome constitutions. This sterility, however, 
can be reduced by pure breeding after back- 
crossing, which forms new homozygotes hav¬ 
ing a chromosome constitution different from 
either of the parental species. The possibility 
of species formation in this manner was sug¬ 
gested by Danforth and Sandnes (1939). In the 
Marianas Islands, the habitat of this duck is 
restricted to very small ponds or lagoons found 
in each island. Consequently, the population is 
very small, and pure blood lines may be carried 
in each population occupying a certain pond or 
lagoon. The phenomenon suggested by Dan¬ 
forth and Sandnes (1939) seems to be realized 
by these circumstances. 
2 1 believe Anas wyvilliana of the Hawaiian Islands 
and Anas laysanensis of Laysan Island originated from 
Anas platyrhynchos stock alone, because no dichro¬ 
matic phases have been reported in either species and 
there is no evidence that a second species ever occurred 
in these islands. 
