This uncertainty about the appearance and ecology of Enantiornis is due in part, ironically, to the sheer number of enantiornithe species known from the rocks that yielded the E.leali fossils, the Lecho Formation of Salta, Argentina. The Lecho Formation dates to the early Maastrichtian age, about 70 million years ago, so its fauna was among the last to harbor opposite birds. The problem with reconstructing Lecho birds is that none are known from complete specimens preserving both fore and hind limbs, and all are characterized by specific features of wither the wing or the tarsometatarsus.
Frustratingly, these features point to a sheer diversity of ecological forms that's almost unknown in other enantiornithes. In general, enantiornithes tend to be relatively conservative in terms of overall body plan. Most are small, and many have short-snouted, toothy, Archaeopteryx-like skulls, though a few, like the longipterygids and gobipterygids, heavily modified the snout to suit different ecological niches. Almost all enantiornithes have feet adapted specifically to an arboreal lifestyle, with reversed halluces and long, curved claws. Their metatarsals tend to be mid-length relative to their tibiotarsi, and are often at least partly fused to each other in a tarsometatarsus.
The Lecho ennatiornithes, however, differ significantly from the norm. Three named species are known from metatarsals: Soroavisaurus australis, Lectavis bretincola, and Yungavolucris brevipedalis. Of these, Soroavisaurus is closest to the standard enantiornithe anatomy, with fairly broad, mid-length, partially fused metatarsals. It has been assigned to the family Avisauridae, and is overall very similar to the North American Avisaurus archibaldi (also known right now from the metatarsals alone, but more of the skeleton is awaiting description as hinted at by Scott Hartman). Though one of the largest enantiornithes, if one were to reconstruct a Soroavisaurus, it would be most conservative to use scaled-up "normal" enantiornithes like Cathayornis or the possible basal avisaurids Cuspirostrisornis. Soroavisaurus had large, strong talons and an inward-oriented second toe, possible perching adaptations like those seen in smaller arboreal species.
Lectavis and Yungavolucris, though, were weird. Very weird. Lectavis had very long and thin legs with fully fused metatarsals which were very long and relatively slender compared to related species. The long, thin tibia ended in a strange forward-protruding knob at the ankle, which would probably have given these birds a "knob-kneed" appearance similar to modern waders like flamingos. The toes aren't known, but the strange leg anatomy strongly suggests that these were not arboreal, but may have been specialized waders (as suggested by Padian & Chiappe 1998). Yungavolucris brevipedalis has almost the opposite foot anatomy. As its name suggests, the metatarsals are extremely short, squat and very unusual in shape. The foot was asymmetrical, with the outer toe small and thin, and the outer two metatarsals flared out to the side like a pair of bell-bottoms. The metatarsal for digit 2 was large and robust. The entire tarsometatarsus was extremely flattened. This odd arrangement of the foot bones is somewhat reminiscent of hesperornithes, and it's likely that Yungavolucris were strong swimmers if not fully aquatic foot-propelled divers.
The rest of the Lecho birds are known only from wing bones, which, at least compared to the feet, were all pretty conservative as enantiornithe wings go and don't tell us nearly as much about ecology. The "wing species" include Martinavis saltariensis, Martinavis vincei, Martinavis saltarienensis, Martinavis minor, Elbretornis bonapartei, and Enantiornis leali itself. For those keeping count, that means there are at most seven named species of enantiornithes in the Lecho. But were there really?
Given the fact that there is no overlapping material between the wing species and the foot species, it's highly likely that we have a Chirostenotes-type situation here, with various unassociated parts of single species being given different names (in the case of Chirostenotes, the feet were named Macrophalangia, the hands Chirostenotes, and the jaws Caenagnathus). But is is possible to sort out the Lecho specimens into groups and figure out which wing goes with which foot?
Several papers have offered conflicting speculation on this point. For example, Walker et al. (2007) suggested that Soroavisaurus is probably the same animal as either Enantiornis leali or one of the Martinavis species (if they are in fact distinct and not growth stages). If Soroavisaurus = Enantiornis, the name Avisauridae would have to go: Enantiornithidae has priority. However, Walker & Dyke (2010) suggested that Enantiornis may instead be a synonym of Lectavis, or, on the other hand, their newly named Elbretornis may represent Lectavis wings. So, as you can see, while almost all workers in this area agree that the Lecho birds are probably over-split, there is no consensus which wings go with which feet, with some researchers changing their opinions from year to year, and yet continuing to name foot-free wings as new species while acknowledging that they will likely end up as synonyms.
One key to the whole issue could be size. Mickey Mortimer (online) suggested that Elbretornis may, based on size, be the wings of Yungavolucris, rather than Lectavis as suggested by Walker & Dyke (2010). Lectavis was approximately twice the size of Yungavolucris (based on my reconstructions of it an unusually long-legged bird, Lectavis was around 60cm long). The humerus of Elbretornis was only about 9cm long. If Elbretornis = Lectavis, that species must have been relatively short-winged. But, keep in mind that Lectavis may not be as long-legged as its slender bones suggest. If, by chance, it is a larger, relatively shorter-legged bird with unusually slim lower legs, it could have been large enough to belong to the large Enantiornis wings. And that's the kicker--Enantiornis is one of the largest known Mesozoic birds, with a wingspan of at least a meter, probably more. Most Lecho fossils come from very small individuals, except for Soroavisaurus australis and Lectavis bretincola, which probably had a nearly 1m wingspans themselves. Based on large size alone, Soroavisaurus and Lectavis look like the best contenders to represent the feet of Enantiornis. However, the humerus of Martinavis vincei is not much smaller than that of Enantiornis (11cm vs. 15cm), which means it needs to be considered as a possible synonym of Soroavisaurus or Lectavis as well, using size alone (also noted by Mortimer, online).
So, what is Enantiornis? We know of three major types of bird in the Lecho: Possibly hawk-like perching birds, long-legged possible waders, and stout-footed diving birds. Any one of these three (or none, as there appear to be more species known from wings than from feet) could be Enantiornis. For now, there's no way to know with any reasonable degree of certainty. We can only hope that a few more complete skeletons are found.
As mentioned above, there may be hope to at least narrow down the field. If the rumored new Avisaurus archibaldi specimens preserve wing bones, it should be possible to compare these with the Lecho "wing species" and figure out which one is most likely to belong to an avisaurid. Comparison with the likely avisaurid species Neuquenornis volans, known from both wings and feet, would also shed some much-needed light on the situation. Doing so would present one or more clear candidates to match a set of wings to Soroavisaurus, and we would then be able to either rule it out as a synonym of Enantiornis, or consider sinking it.
* Padian, K. and Chiappe, L.M. (1998). "The origin and early evolution of birds." Biological Reviews, 73: 1-42.
* Walker, C.A., Buffetaut, E., and Dyke, G.J. (2007). “Large euenantiornithine birds from the Cretaceous of southern France, North America and Argentina.” Geological Magazine, 144(6): 977-986.
* Walker, C.A. and Dyke, G.J. (2010). "Euenantiornithine birds from the Late Cretaceous of El Brete (Argentina)". Irish Journal of Earth Sciences, 27: 15–62.