Retrotransposons have proliferated extensively in eukaryotic lineages; the genomes of many animals and plants comprise 50% or more retrotransposon sequences by weight. There are several persuasive arguments that the enzymatic lynchpin of retrotransposon replication, reverse transcriptase (RT), is an ancient enzyme. Moreover, the direct progenitors of retrotransposons are thought to be mobile self-splicing introns that actively propagate themselves via reverse transcription, the group II introns, also known as retrointrons. Retrointrons are represented in modern genomes in very modest numbers, and thus far, only in certain eubacterial and organellar genomes. Archaeal genomes are nearly devoid of RT in any form. In this study, I propose a model to explain this unusual distribution, and rationalize it with the proposed ancient origin of the RT gene. A cap and tail hypothesis is proposed. By this hypothesis, the specialized terminal structures of eukaryotic mRNA provide the ideal molecular environment for the lengthening, evolution, and subsequent massive expansion of highly mobile retrotransposons, leading directly to the retrotransposon-cluttered structure that typifies modern metazoan genomes and the eventual emergence of retroviruses.