104 Evolution and the Fossil Record
laboratory mice could pass on their immunity directly to their offspring (Steele et al. 1998).
It is hard to see how this is explained by anything other than Lamarckian inheritance.
More recently, molecular biologists have found that acquired inheritance is the norm,
rather than the exception, in most microorganisms. Viruses work entirely this way, inserting
their DNA into the cell of a host and making more copies of themselves. Many bacteria and
some other organisms (including plants such as corn) seem to have “jumping genes” that
exchange gene fragments between strains of organisms without sex or even recombination.
One group of viruses (the retroviruses that cause HIV, among other infections) copies their
own genetic information from host to host and may be capable of carrying the DNA of one
organism into another.
All of these new mechanisms of inheritance suggest the genome is not as simple and
“one-way” as we thought only 40 years ago. John Campbell (1982) summarized a full range
of genetic interactions, starting with the simple “structurally dynamic” genes that respond
to a certain environmental stimulus by producing a particular response. At a more sophis-
ticated level are genes that apparently sense their environment and change their response.
Automodulating genes change their future responsiveness to stimuli when stimulated.
The most Lamarckian of all are “experiential genes,” which transmit specific modifications
induced during their lifetimes into the genome of their descendants. The example from
immunology may fit this, as does bacterial and viral DNA swapping.
Clearly, the simplistic “central dogma” no longer applies to microorganisms, which are
remarkably promiscuous in swapping DNA around. It may also not apply to many multicel-
lular organisms either, if the immunologic experiments are correctly interpreted.
Neutralism, Junk DNA, and Molecular Clocks
One of the first challenges to neo-Darwinism came when molecular biology began to under-
stand the details of the genome in the 1960s. Prior to this, geneticists had assumed that each
gene in the chromosome coded for only one protein (and the structures built from them), so
that inheritance would be simple (the “one gene, one protein” dogma). They also asserted
that every gene was under the constant scrutiny of natural selection (panselectionism) and
no gene was selectively neutral (even if we can’t detect how selection operates). But in the
1960s, a series of discoveries shattered this simplistic idea of the genome. Using a newly
developed technique called electrophoresis, Lewontin and Jack Hubby (1966) found that
organisms had far more genes than they actually use or that can be expressed in the pheno-
type. Soon, geneticists were discovering that as much as 85–97 percent of the DNA in some
organisms (including about 90 percent of human DNA) is not critical for expression of a
phenotypic feature and is either “silent” DNA or “junk” DNA left over from the distant past
when it had some function. If it is not expressed, it cannot be detected by natural selection
and is neutral with respect to selective advantages or disadvantages. This new idea of neu-
tralism completely shattered the old belief in panselectionism. In recent years, a few geneti-
cists have tried to salvage the idea that there is less “junk DNA” than once thought, and
Project ENCODE made the claim that most of the DNA was minimally functional. However,
these claims have been debunked by many different lines of evidence. Most of our DNA is
indeed “junk” that is never read or used in any real functional sense.
At the most basic level, the fundamental structure of the genetic code guarantees that a
high percentage of mutations will be invisible to natural selection. The genetic code (fig. 4.2)
consists of a three-letter “triplet” sequence of nucleotides (adenine, cytosine, guanine, and