Large-scale human/mouse DNA comparison studies have discovered numerous conserved noncoding sequences of which only a fraction has been functionally investigated. A question therefore remains as to whether most of these sequences are conserved due to functional constraints or are the result of a lack of divergence time? Based on the suppostition that actively conserved human/mouse noncoding sequences will be present in a third mammal while noncoding regions that are similar because of an insufficient accumulation of random mutations will be absent, we sequenced and compared -200 kb of orthologous human (5q31), mouse and dog DNA. To optimized recognition of conserved noncoding sequences, we developed a program to search for blocks of similarity in globally aligned sequences (eliminating biases against small insertions/deletions which are commonly found in regulatory elements) and coupled it to a new visualization tool. VISTA, to analyze regions of similarity in the human, dog, and mouse DNA sequences simultaneously. Comparative analysis revealed that the content and order of the 6 genes in this region is the same for all three species. Baseline similarity levels of noncoding sequences in humans and dogs were -2 fold higher than in humans and mice and -35 fold higher than in dogs and mice. Varying criteria was thus used to define noncodina elements as conserved in the twoway sequence comparisons of the different species: human/dog (more than 90% identity over more than 120bp), human/mouse (more than 80% identity over more than 120bp), and dog/mouse (more than 75% identity over more than 120bp). Fifteen human/mouse conserved noncoding sequences were identified, all 15 elements overlapped with mouse/dog and 11 overlapped with human/dog conserved noncoding sequences. At these cutoff criteria, all the conserved noncoding element discrepancies fell into two categories, either putative false negatives 9the conserved noncoding elements are present in all 3 species but fall slightly below the cut off value) or correspond to missing draft dog sequence. Our study suggests that the majority of identified human/mouse conserved non-coding regions are likely the result of active conservation due to function. Furthermore, the VISTA tool developed for this analysis is an effective means for displaying multiple two-way sequence alignments to generate hypothesis using comparative genomics data sets.