There is an additional complication in the genes of eukaryotes. The mRNA that is to be translated contains stretches of noncoding sequence that are removed before translation begins. These noncoding stretches are called introns (for INtervening sequences) and the sequences that are translated are called exons (for EXpressed sequence). Before translation, enzymes called spliceosomes get to work on the mRNA, cutting out the introns and splicing the exons back together. The locations of exon/intron boundaries are called splice sites, more specifically an exon-intron boundary is called a donor site and an intron-exon boundary is called an acceptor site. There are certain characteristics typical to these sites that we will discuss later. Note that donor sites may occur within codons as well as between them, in which case the next acceptor site must occur at the correct position to ensure that the 3-bases-per-codon pattern is conserved.
As a general rule, higher eukaryotes (such as mammals) have a higher proportion of genes with introns, in lower eukaryotes (yeast, eubacteria) the proportion is smaller.
There is also a phenomenon know as polyadenylation, in which a sequence of As is added to the end of the mRNA (located just before the transcription termination) though this is not coded in the DNA. In mammals this extra sequence of adenylate residues is typically on the order of a couple of hundred base pairs in length. This is the handiwork of an enzyme called poly A polymerase and it occurs during the translation process converting DNA to mRNA. As an mRNA ages the poly A section typically becomes shorter. About 10-30 bp upstream of the poly A tail the sequence AAUAAA (or some slight variant) always appears. This is know as the polyadenylation signal.