杏吧原创

RESULTS from one of the largest scientific experiments ever conducted in
the US have proved one thing beyond doubt: that things don鈥檛 always turn out
the way they were planned.

More than 17 000 people across the US volunteered to join the Seed
Preference Test, a survey organised by researchers at Cornell University in
New York. The amateur scientists had to set out three types of seed on pieces
of cardboard and record how many birds of various species ate each type of
seed. The volunteers even paid for the privilege of taking part in the study,
shelling out $7 each for instructions and materials (This Week, 23
October 1993). But at the end of the study, fewer than 5000 鈥渞esearchers鈥�
submitted reports on the five-minute experiments to the organisers.

In some cases, the 鈥渒itchen window scientists鈥� lost interest and either did
not perform the experiment or failed to complete the paperwork. Others
diligently set out the seed but to no avail. 鈥淭housands of the participants
had trouble attracting birds,鈥� says Margaret Barker, one of the leaders of the
project. 鈥淪ome people would stare at the pieces of cardboard for hours, and
the birds didn鈥檛 come.鈥�

Nevertheless, the scientists were pleased with the results. More than 500
000 bird visits were reported, and the data turned conventional wisdom on its
head. Birds are supposed to like sunflower seeds best, but the experiment鈥檚
data show that in the western part of the country birds prefer red milo, a
variety of sorghum that has been included in birdseed mixtures only as filler.
Ground-feeding birds preferred millet, and birds that generally feed in trees
liked black-oil sunflower seeds best.

The Cornell researchers are planning to try again this year, with an
improved scientific protocol. The cardboard has been eliminated in case it
frightens some birds. And volunteers will be told to set out the seed several
days before they begin to collect data. That will give the birds time to
sample everything on the menu before they are asked to review the
restaurant.