The genus was named in 1895 by American paleontologist George Reber Wieland based on a skeleton from South Dakota, who placed it into the extinct family Protostegidae.

It had an especially hooked beak and its jaws were adept at crushing, so it probably ate hard-shelled crustaceans, mollusks, and possibly even sponges, while slowly moving over the seafloor.

With its large and strong foreflippers, Archelon was likely able to produce powerful strokes necessary for open-ocean travel and, if need be, escape from fellow marine predators.

It inhabited the northern Western Interior Seaway, a mild to cool temperate area, dominated by plesiosaurs, hesperornithiform seabirds, and mosasaurs.

It may have gone extinct due to the shrinking of the seaway, increased infant mortality rates (in the sea), higher instances of egg and hatchling predation (on land), and a rapidly cooling climate.

The holotype specimen, YPM 3000, was collected from the Late Campanian-age Pierre Shale of South Dakota along the Cheyenne River in Custer County by American paleontologist George Reber Wieland in 1895, and described by him the following year based on a mostly complete skeleton excluding the skull.

In the same study, Protostega copei from Kansas, which was first described by Wieland in 1909 and named in honor of Edward Drinker Cope who first erected the family Protostegidae,[10] was moved to the genus Archelon as A.

[12][13] In 1992, a fourth and the largest specimen to date, nicknamed "Brigitta", was discovered in Oglala Lakota County, South Dakota and resides in the Natural History Museum Vienna.

The nostrils are elongated and rest on the top of the skull, slightly posited forward, and are unusually horizontal compared to sea turtles.

[18] The humeri in the upper arms are proportionally massive, and the radii and ulnae of the forearms are short and compact, indicating the animal had strong flippers in life.

[18] In the absence of firmly jointed neck and pleural plates, the skin over the carapace was probably thick, strong, and leathery in order to compensate and properly support the shoulder girdle.

[20] A turtle plastron, its underside, comprises (from head-most to tail-most) the epiplastron, the entoplastron, which is small and wedged in between the former and the hyoplastron; then is the hypoplastron and finally, the xiphiplastron.

[24] In 1953, Swiss paleontologist Rainer Zangerl split Protostegidae into two subfamilies: Chelospharginae and Protosteginae; to the former was assigned Chelosphargis and Calcarichelys, and the latter Archelon and Protostega.

[11] The sister group of Protostegidae had, in the past, been considered to be Dermochelyidae, and thus their closest living relative would have been the dermochelyid leatherback sea turtle (Dermochelys coriacea).

[12] However, phylogenetic studies conclude that protostegids represent a completely separate, basal (ancient) lineage that originated in the Late Jurassic, removing the family from the superfamily Chelonioidea (which includes all sea turtles).

The thick plastron indicates the animal probably spent a lot of time on the soft, muddy seafloor, likely a slow-moving bottom feeder.

According to American paleontologist Samuel Wendell Williston, the jaws were adapted for crushing, implying the turtle ate large mollusks and crustaceans.

[29] It might have been able to target larger fish and reptiles,[9] as well as, similar to the leatherback sea turtle, soft-bodied creatures such as squid and jellyfish.

[29] Archelon probably had weaker arms, and thus less swimming power, than the leatherback sea turtle, and so did not frequent the open ocean as much, preferring shallower, calmer waters.

This is indicated by the similarity of the humerus/arm and hand/arm ratios of it and cheloniids, which are known to have poor development of the limbs into flippers and a preference for shallow water.

[34] The Late Cretaceous Dakotas were submerged in the Northern Inland Subprovince, an area characterized by moderate to cool temperatures, with an abundance of plesiosaurs, hesperornithiform seabirds, and mosasaurs, particularly Platecarpus.

[34] However, some Maastrichtian-age Kansas Pierre Shale fossils may have been eroded millions of years ago, and it is possible Archelon survived well into the Maastrichtian.