From the record
Real cases, written from the published record. Predict first; the record and its sources come after.
1977 · Galápagos
Daphne Major, 1977
On the small island of Daphne Major, Peter and Rosemary Grant measured the beaks of the medium ground finches year after year.
In 1977 almost no rain fell. The small, soft seeds ran out first; what was left were large, hard seeds that take a strong bite to crack.
When the drought ended, how did the surviving finches' beaks compare with the beaks before it?
What happened. Most of the finches died. The survivors had beaks a few percent deeper on average, and their chicks inherited much of that difference. Then the very wet years of 1982–83 brought plenty of small seeds, and over the next dry spell smaller beaks did better: the push went the other way.
Selection has a direction, but the direction can change with the weather.
1890s–today · Pacific coast
The seals came back
Northern elephant seals were hunted for their oil until perhaps only a few dozen were left in the 1890s. Protected, they recovered: today there are well over a hundred thousand.
A bottleneck is when a population shrinks to a few individuals and then grows again.
Did their genetic variety come back with their numbers?
What happened. In 1974 Bonnell and Selander checked two dozen genes in northern elephant seals and found no variation in any of them. The few survivors carried only part of the old population's alleles; drift in a tiny population lost the rest, and new variety arrives only through rare mutations.
Numbers can come back in a century. Lost variety mostly does not.