where the traps in Massachusetts were located? The answer probably in- 
volves a combination of circumstances each of which plays a part of un- 
known importance, Although a satisfactory explanation is not possible 
at the moment, a discussion of some of the factors is pertinent. But, 
before going into a discussion of these it would be well to examine 
briefly the remainder of the Massachusetts bandings. 
Indirect recoveries of birds banded during the fall and taken 
in subsequent fall periods show 47 percent and 52 percent taken within 
a 50-mile radius of the banding stations at Newburyport and Cape Cod 
respectively, and the percent of southward recoveries is about the same 
as the first-season recoveries. These percentages of local recovery 
are not so high as those for some coastal areas to the south such as 
in Maryland and Long Island but seem to conform in a general way to 
bandings both on the ccast and northern inland areas where large bodies 
of water are found and where a wintering populatim usually exists. 
The indirect recoveries to the north of Newburyport show 
strong affiliations with the Maritime Provinces, Labrador, and Quebec, 
(fig. 6). These total 82 (28 percent) amd Quebes alone accounts for 
33 (11 pereent). The indirects (total 772) to the north from Cape Cod 
fall bandings show similar ties with eastern Canada except that New- 
foundland (2 percent) and Nova Scotia (14 percent) are involved to a 
greater degree. Eastern Canada accounts for 259 indirect recoveries 
from fall bandings on the Cape. 
With winter bandings one would expect a very limited recovery 
south of the Newburyport station or the one on Cape Cod, since most of 
these birds were on the wintering ground when banded. Sixty percent of 
all fall recoveries for winter bandings at Newburyport (total 287) were 
taken in the immediate area of the banding station. This is some 12 — 
percent higher than with indirect fall bandings and 13 percent lower 
than for direct first-season recoveries. The southward recoveries (10 
south of New England) were only a little over 3 percent. Recoveries 
(146) from spring bandings (March-May) at Newburyport, during the 
period when most black ducks are on the move, show a close association 
of migrants at Newburyport with populations accustomed to wintering at 
least as fa> south as New Jersey. Twenty percent of these recoveries 
were taken south of New England with approximately 12 percent in south- 
eastern New Jersey. Fifty-one percent were taken in the banding area 
in subsequent years, indicating that many of the spring~banded birds 
were not migrants from the south but actually individuals which had 
wintered at Newburyport. 
Winter and spring bandings on Cape Cod (totaling 679 recover- 
ies) give much the same results as fall bandings. Over half the reported 
recoveries were taken locally and there was only a dribble to the south 
of New England. The pattern of recoveries in eastern Canada from these 
bandings is similar to that of the fall-banded birds. Recoveries of 
Newburyport birds indicate stronger ties with the mainland than do the 
recoveries of the Cape Cod birds. 
=12. 

