Homo-DNA ((4'-->6')-linked oligo-2',3'-dideoxy-beta-D-glucopyranose nucleic acid) constitutes the earliest synthetic model system whose pairing properties have been studied within an etiology of nucleic acid structure. Its conception as part of a program directed at a rationalization of Nature's selection of pentoses over other candidates as the carbohydrate building block in the genetic material was motivated by the question: why pentose and not hexose? Homo-DNA forms an autonomous pairing system and its duplexes are entropically stabilized relative to DNA duplexes. Moreover, the base pairing priorities in homo-DNA duplexes differ from those in DNA. A deeper understanding of the particular properties of homo-DNA requires knowledge of its structure. Although diffraction data for crystals of a homo-DNA octamer duplex were available to medium resolution in the mid-1990s, it took another decade for the structure to be solved. In this tutorial Review we describe the odyssey from the crystallization to the final structure determination with its many failures and disappointments and the development of selenium chemistry to derivatize nucleic acids for crystallographic phasing. More than fifty years after the discovery of the DNA double helix, the story of homo-DNA also provides a demonstration of the limits of theoretical models and offers a fresh view of fundamental issues in regard to the natural nucleic acids, such as the origins of antiparallel pairing and helicality.