wide. Since the golden plovers apparently donot rest on water and hence 
-must make the journey non-stop this has always appeared to be one of 
the most impressive and mysterious cases of migration (Henshaw, 1915). 
But actually we have no banding evidence to tell us that the golden 
plovers from any one point in the summer range tend to fly to a partic- 
ular Pacific Island. Moreover, only the extreme western edge of the 
summer range lies in Alaska; the birds are also foundnesting in a large 
section of Siberia. And in the winter they have been reported not only 
from Pacific Islands but also from New Zealand, Australia, the West 
Indies, Japan, China and India. This range is alarge target,and we have 
no reason to think that the birds are particularly selective about the 
portion of it that they fly to. 
Nevertheless, there is a marked migration of these plovers over 
the Alaska Peninsula and the Aleutian Islands. These birds apparently 
head out over the Pacific and some of them stop on the closest land they 
can reach--the Hawaiian Islands. We have some evidence (not by any 
means conclusive), derived from records of plovers sighted by ships in 
the NE Pacific, which indicates that the course flown by many of these 
birds over the ocean is anything but a straight line between the starting 
points and the goal of the migration. 
In the light of the evidence, it seems that we should focus our 
search for the basis of bird navigation not upon some hypothetical built- 
in physiological instrument capable of setting very precise courses, but 
upon something which could guide a migrating bird in the right general 
direction, plus or minus perhaps 20 to 30 degrees. What environmental 
factors could guide migrants to this degree? Our meagre knowledge of 
birds’ behavior during migration leads to speculation. But a few factors 
are at least worthy of further study. Wind direction, for example, at 
altitudes of 2000 feet or more is less variable than it is at the earth’s 
surface. One investigator (Lowery, 1950) has noted a high correlation 
between the direction of winds aloft and the direction of nocturnal bird 
migrations, observed through a telescope against the illuminated sur- 
face of the moon. Cloud formations also may be stable enough in some 
areas to afford guidance over what might at first seem to be track- 
less areas. 
Other possibilities that should not be overlooked as possible ce- 
lestial landmarks, are the sun and moon, for we know that insects--the 
honeybee, for example (Frisch, 1950)--set flight courses rather accu- 
rately by means of the sun’s position. Both the suri and moon appear to 
observers in northern latitudes to lie well south of the zenith and thus 
offer a limited degree of guidance. 
We do not know if-any of these factors are actually recognized by 
birds. We do know that their senses are adequate to detect them all. 
Obviously more studies are needed before we can arrive at any definite 
conclusions. Fortunately modern technological advances may help us 
with one problem--the need to learn more about the actual routes flown . 
by migrating birds. We should find out, for example, how direct and 
accurate a course they are able to steer, and determine the relative 
importance of contact flying, guidance by geographical clues (coastlines, 
river systems and mountains), and atmospheric conditions (wind and 
visibility). Probably following the birds in planes isnot the ideal way 
a des 
