Several procedures are described for the synthesis of arabinofuranosyl-5-ethylpyrimidine nucleosides: (a) condensation of benzylated α-chloroarabinofuranose (5) with 2,4-bis-O-(trimethylsilyl)-5-ethylpyrimidine (6) in the presence of SnCl₄ or molecular sieves yielded exclusively the β-anomer 7 and proved to be the best pathway to 1-(β-D-arabinofuranosyl)-5-ethyluracil (1); (b) condensation of benzoylated α,β-bromoarabinose 9 with 6 led to a mixture of anomeric nucleosides 10 and 11 in the ratio 5:2; (c) condensation of benzoylated α-bromoarabinose 8 with 6 led exclusively to the α-anomer 10 which, on thiation and amination, gave 1-(α-D-arabinofuranosyl)-5-ethylcytosine (4); (d) condensation of benzylated α-halogenoarabinose with 2,4-diethoxy-5-ethylpyrimidine (15) in the presence of SnCl₄ or molecular sieves gave exclusively the β anomer of the 4-ethoxy nucleoside 16 which, on amination and reduction, was the most convenient procedure for the synthesis of 1-(β-D-arabinofuranosyl)-5-ethylcytosine (2). The mechanisms of the various condensation methods are compared and discussed. CD and NMR data are presented for 1 and 2 and their corresponding α anomers. Antiviral activities of 1 and 2 and their α-anomers 3 and 4 were tested in primary rabbit kidney cells and in human skin fibroblasts. Herpes simplex virus was inhibited by 1 at a concentration as low as 2 μg/mL, whereas concentrations exceeding 200 μg/mL were required to inhibit vaccinia virus replication or normal cellular metabolism. Compound 2 was significantly less active as an antiherpes agent than compound 1. The relative lack of activity of 2 could be attributed to its low susceptibility to enzymatic deamination by mammalian cytosine nucleoside deaminase. The α anomers of 1 and 2 were totally inactive as antiviral agents.