The family Tenrecidae of Madagascar are by far my favorite examples of speciation, well represented by fossil, subfossil and modern forms linked to an ancestral species in Africa AND a distantly related living species (
Protomogale). Otherwise, species flocks of cichlids in East Africa or the ciscoes (
Coregonus spp) of the US Great Lakes can do in a pinch.
If you want more glaring examples of "modern" transitionals - populations at the cusp of incipient speciation - I'd go with ring species like the
Ensatina eschscholtzii group of salamanders in California, the green warblers (
Phylloscopus trochiloides) of China, Tibet and Siberia, and of course the herring gulls (
Larus spp) which has a nice "ancestor" species to go along with it (
Larus smithsonianus), which species has the added advantage of being still extant. For a good review of the ring species discussion, see Irwin DE, Irwin JH, Price TD, 2001, "Ring species as bridges between microevolution and speciation", Genetica 112-113:223-243.
quote:
A demonstration of how small changes can lead to species-level differences is provided by ring species, in which two reproductively isolated forms are connected by a chain of intermediate populations. We review proposed cases of ring species and the insights they provide into speciation. Ring species have been viewed both as illustrations of the history of divergence of two species from their common ancestor and as demonstrations that speciation can occur in spite of gene flow between the diverging forms. Theoretical models predict that speciation with gene flow can occur when there is divergent ecological selection, and geographical differentiation increases the likelihood of speciation. Thus ring species are ideal systems for research into the role of both ecological and geographical differentiation in speciation, but few examples have been studied in detail. The Greenish warbler is a ring species in which two northward expansions around the Tibetan plateau have been accompanied by parallel evolution in morphology, ecology, and song length and complexity. However, songs have diverged in structure, resulting in a lack of recognition where the reproductively isolated forms come into contact in Siberia. Our analysis suggests that these differences could have arisen even with gene flow, and that parallel rather than divergent ecological changes have led to divergence in sexually selected traits and subsequent speciation.
That'll probably be enough to go on. If you'd like more discussion of why I think these examples are compelling, I'd be happy to expand on this answer (
ad nauseum 
)