Question 041117a: Can evolution be directly observed?

  karthaus@photon.chitose.ac.jp

日本語


Answer 041117a: No it can't. Evolution requires the random change of genetic material by mutation. These changes occur very seldomly. Living cells have a very effective repair mechanism and mutations, if they occur, are removed. Evolutionary changes of species require millions of years according to evolution theory.

Mutation occurs at a rate of 1 in 10^10 nucleotides (10 billion nucleotides). The human genome has 3x10^8 nucleotides. That means that one mutation occurs every 33 generations. The chimpanzee and humans have 1% difference in the genes. That means 3x10^6 nucleotides are different. To have this change 33x3x10^6 generations = 10^8 generations must have accumulated all mutations, no mutation is lethal, and no backmutation should have occured. Even with one generation every 10 years (no human can reproduce within 10 years after birth, though), 5x10^8 years must have past between the separation of chimpanzee and human. This is 500 million years!
According to evolution, this split bewteen apes and humans occured 5 million years ago.
Now, lets make the reverse calculation, of how high the mutation rate has to be to allow for this change in a comparable short time span of 5 million years:

5 million years = 500,000 generations (one generation every 10 years. This is a generous estimate).
3 million different nucleotides = 6 permanent mutations per generation are needed.
Asume 10% of mutations are positive. (Most mutations are lethal and the offspring dies before passes this mutation along to his/her children).
So we need 60 mutations per generation (that means per person).
60 mutations within 3 million nucleotides is 1mutation in 50000 nucleotides.
Ask any molecular biologist or genetic biologist and he will tell you that this is impossible.

In this estimation I have not included that a mutation can occur several times at the same nucleotide position, or that the nucleotide can mutate back again.
Also, these mutations have to occur one after another in the same family tree. If one of the many mutations prove to be fatal, this lineage dies out.
It already has been shown by Haldane (see question 041117c), that it is very improbable that mutations spread through the whole population.

back to Main page
-----------------------------277532456332276 Content-Disposition: form-data; name="userfile"; filename="" Content-Type: application/octet-stream