The results of the search for low-energy conformations of poly(dA).poly(dT) and of the poly(dA).poly(dT) "complex" with the spine of hydration similar to that found by Dickerson and co-workers (Kopka, M.L., Fratini, A.V., Drew, H.R. and Dickerson, R.E. (1983) J. Mol. Biol. 163, 129-146) in the minor groove of the CGCGAATTCGCG crystals are described. It is shown that the existence of such a spine in the minor groove of poly(dA).poly(dT) is energetically favourable. Moreover, the spine of hydration makes the polynucleotide conformation similar to the poly(dA).poly(dT) structure in fibers and to the conformation of the central part of CGCGAATTCGCG in crystals; it also acquires features characteristic of the structure of poly(dA).poly(dT) and DNA oligo(dA)-tracts in solution. It is shown that the existence of the TpA step in conformations characteristic of the poly(dA).poly(dT) complex with the spine of hydration is energetically unfavourable (in contrast to the ApT step) and therefore this step should result in destabilization of the spine of hydration in the DNA minor groove. Thus, it appears that the spine of hydration as described by Dickerson and co-workers is unlikely to exist in the poly d(A-T).poly d(A-T) structure. The data obtained permit us to interpret a large body of experimental facts concerning the unusual structure and properties of poly(dA).poly(dT) and oligo(dA)-tracts in DNA both in fibers and in solution. The results provide evidence of the existence of the minor groove spine of hydration both in fibers and in solution on A/T tracts of DNA which do not contain the TpA step. The spine plays an active role in the formation of the anomalous conformation of these tracts.