Changes

::But first, because the question insists, I will address the boring stuff.

::But first, because the question insists, I will address the boring stuff.

::Genes duplicate, entire sections of chromosomes, and sometimes even whole chromosomes, or sets of chromosomes, can be duplicated or even fused.  Sometimes, just a portion of a gene is duplicated, or maybe even swapped with a part from a different gene.  On a smaller scale, errors in DNA replication, among other factors, can cause single base pairs in DNA to be changed (this is called a "single nucleotide polymorphism" or a SNP--pronounced "Snip"), deleted, or inserted.  Sometimes, it's a transposable element (chunks of DNA that like to move around in the genome--a lot) that inserts in the middle of an exon.  Or maybe even an endogenous retrovirus that's still intact enough to move around finding itself in between two enhancer elements for an essential gene.  Chromosomes are not static structures either.  High school biology was only partially right when they taught you that recombination occurs between homologous chromosomes during meiosis; as it turns out, non-homologous chromosomes trade parts too sometimes (quite often, actually, on an evolutionary time scale). In short, on the scale of hundreds of millions of years, the genome has more in common with Mardis Gras in New Orleans than it does with a leather-bound book.  To call mutations "accidents", overlooks how commonly they occur on evolutionary time scales.

::Genes duplicate, entire sections of chromosomes, and sometimes even whole chromosomes, or sets of chromosomes, can be duplicated or even fused.  Sometimes, just a portion of a gene is duplicated, or maybe even swapped with a part from a different gene.  On a smaller scale, errors in DNA replication, among other factors, can cause single base pairs in DNA to be changed (this is called a "single nucleotide polymorphism" or a SNP--pronounced "Snip"), deleted, or inserted.  Sometimes, it's a transposable element (chunks of DNA that like to move around in the genome--a lot) that inserts in the middle of an exon.  Or maybe even an endogenous retrovirus that's still intact enough to move around finding itself in between two enhancer elements for an essential gene.  Chromosomes are not static structures either.  High school biology was only partially right when they taught you that recombination occurs between homologous chromosomes during meiosis; as it turns out, non-homologous chromosomes trade parts too sometimes (quite often, actually, on an evolutionary time scale). In short, on the scale of hundreds of millions of years, the genome has more in common with Mardis Gras in New Orleans than it does with a leather-bound book.

::On the genomic level, when and where mutations occur is largely random.  The vast majority of mutations have no effect whatsoever on phenotype.  However, when a mutation does affect phenotype, whether or not it survives to be present in future populations is not random.  Any genomic change that increases an organism's ability to reproduce efficiently, will gradually become increasingly prevalent in subsequent generations because the more-efficiently reproducing individuals generate more offspring.  Alternatively, if a variation impairs the ability of an individual to reproduce efficiently, then that variation will become less prevalent in subsequent generations because the individuals carrying it will produce proportionately fewer offspring.  This seemingly obvious observation '''is''' natural selection.  To say that a trait is "selected against" simply means that it has a deleterious effect on the relative reproductive potential of the individuals carrying it.

::On the genomic level, when and where mutations occur is largely random.  The vast majority of mutations have no effect whatsoever on phenotype.  However, when a mutation does affect phenotype, whether or not it survives to be present in future populations is not random.  Any genomic change that increases an organism's ability to reproduce efficiently, will gradually become increasingly prevalent in subsequent generations because the more-efficiently reproducing individuals generate more offspring.  Alternatively, if a variation impairs the ability of an individual to reproduce efficiently, then that variation will become less prevalent in subsequent generations because the individuals carrying it will produce proportionately fewer offspring.  This seemingly obvious observation '''is''' natural selection.  To say that a trait is "selected against" simply means that it has a deleterious effect on the relative reproductive potential of the individuals carrying it.