What? You've never heard of a Mosasaur? Imagine a crocodile with flippers and you're superficially there. See, way back around 100 million years ago the North American continent was split by a vast epeiric sea (a shallow, salty, inland sea) called the Western Interior Seaway (really creative science people).
The whole physics and biology of the Western Interior Seaway and other epeiric seas is pretty amazing, and we'll get into that later. For now, let's just focus on what is known: The sea was shallow (for a sea), probably extremely productive, and was fed by mini-continents on either side. In limnology we tend to think of shallow waters as being the most productive, because of the ability of rooted plants to grow. I think the same is true of shallow seas, but I don't think it is for the same reasons (any oceanographers care to enlighten me on this?).
Anyway! The top predator of this inland sea and the ocean at large became the mosasaur (pictured below). Although this is a reptile, this isn't a dinosaur, and it isn't a crocodile. Think lizard. In fact, think monitor lizard (Family Varanidae), which includes the still living and awesome Komodo dragons and the recently extinct Megalania (the feared giant lizard of Australia). Many modern monitor lizards are highly adapted for living in and around water (e.g., water monitors), and so it isn't hard to imagine a scenario wherein these amphibious lizards took the next evolutionary step and became aquatic.
And become aquatic they did! Unlike turtles, crocodiles, walruses, and quite like whales and dolphins, the Mosasaurs gave birth at sea, to live offspring. The arms and hands of this species developed into flippers and became somewhat detached from the backbone (meaning they could not support their own weight on land). Pull this lizard out of the water and it would asphixiate.
One of the many interesting things about this invasion of the seas by lizards is that they weren't exactly getting themselves into an unoccupied niche. The first real mosasaurs occurred in the Cretaceous, and by 90 million years ago (MYA), we had three big subfamilies with lots of known species. However, way back in the Jurassic reptiles had already colonized the oceans with the famous plesiosaurs (think the mythical Loch Ness Monster) and of course the everpresent sharks had managed to come up with a particularly devastating breed (the Ginsu sharks) and you had other random nastiness lurking out there (just don't go swimming in the Cretaceous). Nevertheless, the Mosasaurs got really big and are generally thought to have been the top predator.
The reason I started getting interested again in Mosasaurs was an interesting paper by Mike Everhart published in the latest issue of the Kansas Academy of Science. The paper explores the occurence of mosasaur on mosasaur violence. And for something that happened 90 MYA, we can actually figure out quite a bit. Everhart was able to, fairly convincingly, estimate the size and mass of the attacker and the way in which it bit down on the victim.
A plate depicting the first Mosasaur discovery. The name is from Latin Mosa meaning the 'Meuse river' in the Netherlands, and Greek sauros meaning 'lizard'. The first specimen was found in a Meuse limestone quarry.
The way Mosasaurs ate is interesting, and has lead to some debate about the origin of snakes. Mosasaurs, unlike sharks or dolphins or most other big predators we're familar with, didn't have teeth that cut, just teeth that crushed. As a result, Mosasaurs either had to bite their prey in half (although this might have been possible, the family lacks the heavy skull that is common in crocodilians, making this somewhat unlikely) or swallow it whole. In order to maximize the food consumption, the Mosasaurs apparently had a somewhat hinged jaw that has lead many people to think: SNAKES!
Indeed, the theory of a shared snake and mosasaur marine ancestory has recently been in vogue, although it was first proposed in the 1890s (it's called Pythonomorpha if you're curious). The idea of snakes evolving from water lizards sounds inherently unlikely to me (why evolve flippers if you are just as good off just getting rid of limbs altogether?). Luckily, I don't need to push my uninformed logic, because apparently more recent fossils have shoved this theory to the wayside (read the link on the Phythonomorpha to get a better understanding). You can rest comfortably, however, knowing that this issue will continue to be fought bitterly by individuals committed to one side or the other until they die, at which point the next generation of paleontologists will find some other issue to argue about. And you thought politics lasted forever.
Mosasaurs have been found with a variety of food items in their stomaches: Mostly fish, but also a plesiosaur, other mosasaurs, turtles, birds, and sharks. I know sharks are fish too, but at least some species of sharks appear to have gone extinct as the mosasaurs became more prominent, suggesting they might have out-competed them (via direct predation?). Some species appeared to specialize on clams, and at least one late evolving species (Leiodon) managed to evolve teeth that cut. All of this evolutionary achievement probably put the Mosasaurs at the top of the oceanic food chain for around 20 million years before the K-T event happened and the entire family was wiped out. According to Everhart:
"Mosasaurs ruled the oceans of the Late Cretaceous and were beginning to invade fresh water environments such as estuaries, swamps and rivers when the Age of Dinosaurs ended. Did they die suddenly due the catastrophic effects of an asteroid impact in the Yucatan, or was their extinction more gradual following the general collapse of the marine ecosystem? We may never know."
Everhart doesn't mention a third possibility, which is that they are still out there and we just don't know it. Seriously, don't go swimming in the ocean.