If it has feathers it is a bird. If it has hair it is a mammal. What defines a reptile. I hear TV shows saying Alligators are more similar to birds then turtles. Right?
Right.
What’s a reptile is a thorny philosophical question actually, as Class Reptilia as construed now is essentially a paraphyletic group which is bad juju under more modern classification schemes ( i.e. not a complete grouping of most recent ancestor-descendant lineages - ideally the Reptilia should include the birds, Class Aves ).
Here’s one definition for ya:
*Reptilia is defined as “the most recent common ancestor of extant turtles and saurians, and all of its descendants” (Gauthier et al., 1988).
Characters supporting Reptilia include:
Tabular small. The tabular (a bone on the posterolateral corner of the skull table) of early synapsids and diadectomorphs is large, but reptiles have only a small tabular, when it is present.
Suborbital foramen present (Fig. 1B-D). The suborbital foramen is a small hole near the lateral edge of the palate, between the pterygoid, palatine, and ectopterygoid (or jugal, when the ectopterygoid is absent). This structure was not found in early synapsids and diadectomorphs (Fig. 1A).
Supraoccipital anterior crista present. The supraoccipital of mesosaurs, synapsids, and diadectomoprhs lacks anterior parasagittal flanges. The supraoccipital of reptiles has a paired anterior parasagittal flange called an anterior crista.
Supraoccipital plate narrow. The supraoccipital is a bone in the back of the braincase. The supraoccipital plate of mesosaurs, synapsids, and diadectomorphs is broad and extends farther laterally than the postparietal. *
Now there’s an easy one to remember :D.
Classically reptiles have generally been defined as ectothermic vertebrates, with a complete terretrial life-cycle ( not always terrestrial, but what this implies is no larval stage like some amphibians ), that have internal fertilization, produce aminotic eggs with leathery shells ( not always, there are ovoviviparous and viviparous reptiles ), and have impermeable skin ( no gas diffusion ). Plus some other things like metanephric post-embryonic kidneys.
- Tamerlane
IIRC the theories are that amphibians left (or evolved from fish that left) the water, but still laid eggs in the water and are tied to it throughout their life. Thier skin must be kept damp - something they cannot do without the aide of a nearby pool. Reptiles, on the other hand, have evolved the ability to leave the water completely behind. They can reproduce on land, and their tough, dry skin allows them to survive the toughest heat without required H2O.
Birds & mammals evidently separately evolved from reptiles (cold blooded). Both are warm blooded, and while birds still lay eggs like reptiles and amphibians, mammals evolved internal fertilization. The first mammals were marsupials (kangaroos, Possums), and later the pouch was lost (in most) and the mammals you see today became the standard.
Reptiles have watertight eggs and dry scaly skin so that they can survive long periods of time out of water (unlike amphibians such as frogs). They also breathe differently from amphibians; reptiles can expand and contract their rib cage while breathing, but amphibians have to pump air into their lungs by squeezing their throat.
Both mammals, birds, and modern-day reptiles evolved from ancient reptiles, although the three lines split up hundreds of millions of years ago. Mammals, of course, developed hair and mammary glands, and birds developed feathers, hollow bones, and other adaptions for flying.
Crocodiles and alligators, however, have evolved from the same common ancestor as birds and dinosaurs, and so share some more “advanced” features of these groups, including a fully four-chambered heart. All the other reptiles do not have two full ventricles, and so the oxygenated blood from their lungs gets mixed in with the unoxygenated blood from their veins. Crocodiles also take care of their young, which other reptiles do not but birds do (and dinosaurs are believed to have).
However, crocodiles are still classified as reptiles because they’re ectothermic (cold-blooded). This is considered an important enough evolutionary change that crocodiles and alligators really can’t be considered “birds” without it.
This is all from a textbook I have in front of me, Raven and Johnson’s Biology: Sixth Edition, 2002.
Doncha think monotremes might have preceded them?
IANAEB
Oh yeah, forgot about those odd things. Australia seems like a world left behind sometimes.
Tamerlane gives the standard hard-parts definition that is important in paleontology (where you can’t check for scales, feathers, or fur). But one very good definition would be (and I’m adlibbing this, paraphrasing a traditional one) reproduction by amniotic eggs (which may be laid, or be retained internally in some species), presence of only one ear ossicle, the stapes, and absence of feathers and of the distinctive forelimb bone structure of birds.
Amniota, having embryos encased in an amnion, is a clade comprising birds, mammals, and reptiles; the reptiles are the paraphyletic group left over when the first two are removed; hence the secondary clauses above.
Within Amniota, four main groups are distinguished:
[list=A][li]Synopsida, characterized by a fenestra (open area) in the skull in the temporal region, and including the extinct sailbacks and mammal-like reptiles and all mammals.[/li][li]Anapsida, characterized by a skull without fenestrae, characterized by very archaic late Paleozoic forms and the turtles[/li][li]Archosauria, characterized by two fenestrae, one at the temple and one higher on the skull, and including a wide assortment of extinct reptiles, notably the dinosaurs and pterosaurs, and also crocodilians and birds[/li][li]Lepidosauria, also originally having the two fenestrae but with a strong tendency to bone reduction and jointing for a very flexible skull, including several small extinct forms and modern snakes, lizards, and tuatara.[/list][/li]
A fifth group, Euryapsida, with only the high-on-the-side-of-the-skull fenestra, seems not to have been a true group, but rather several lepidosaur groups that independently eliminated the temporal fenestra. Most of the Mesozoic sea monsters were “Euryapsids.”
Some fish are warm blooded aren’t they??
Yes, sort of, though not quite in the same way as birds and mammals.
What certain fish like Tuna have is a clever counter-current exchange mechanism that allows them to maintain a sort of endothermy. This is often referred to as ‘partial endothermy’, though some don’t make any such distinction.
- Tamerlane
I think Tamerlane’s answer was outstanding. But, if you want a simple definition, along the lines of the ones quoted, you could go with “if it’s hairless and completely covered with scales”. Amphibians don’t have scales, so there’s your separation point for them. There are some scaly animals that aren’t reptiles, such as Armadillos and Pangolins (scaly anteaters), but they have body parts that aren’t covered in scales. Plus, they have hair.
With the discovery of feathered dinosaurs, not even this is really true anymore. It is becoming pretty clear that feathers evolved well before true birds did, which means the diagnosis of “bird” will likely have to be changed. Downy feathers, for example, appear to be a feature common to most (if not all) coelurosaurs, while vained feathers are now considered a Eumaniraptoran trait (a group which includes birds, dromaeosaurs like Deinonychus and Velociraptor, and some early “almost birds” like Caudipteryx).
It may also be worth noting that while Gauthier’s definition (as posted by Tamerlane) tends to remain intact, exactly who is included in Reptilia remains somewhat controversial. The center of this controversy lies in the placement of Testudines (turtles) - some authors place Testudines among the Diapsida (which would mean Gauthier’s Reptilia is composed only of Diapsida!), others place them in the Anapsida, along with various “parareptiles”: Pareiasauria, Procolophonoidea, Milleretidae, and a number of others. This dramatically broadens the scope of Reptilia.
Darwin’s Finch, IIRC Tyrannosauridae are now considered a family of Maniraptores; the possibility of a feathered T. Rex is something to definitely intrigue one! 
Actually, Tyrannosauridae are still separate from Maniraptora*. However, Tyrannosauridae is a member of Coelurosauria - which means the downy feathers may well have been (may have been - there is no evidence, as yet, to support such) present in at least the young. And I certainly agree that the idea of feathered tyrannosaurs (even if only the juveniles) is intriguing!
*Thomas Holtz, a vertebrate paleontologist with the University of Maryland, had erroneously redefined Maniraptora to include Arctometatarsalia (Holtz 1994), a group which itself included Tyrannosauridae, Troodontidae and Ornithomimosauria. In 1996, he revised his proposed phylogeny to create Maniraptoriformes (=[Ornithomimisauria + Tyrannosauroidiae] + Maniraptora). You can see various current cladograms here.