The Green River Formation by Tom Caggiano

Historical Background

In 1856 Dr. John Evans collected the first fossil fish known to science from the Green River Formation. The Knightia eocaena was described by Dr. Joseph Leidy in Philadelphia. Although fossils had been reported from the area, this was the first described. And so begins the story of one of the greatest fossil sites known to science.

Soon after, geologists, railroad workers, and others started reporting fossil sites in surrounding areas. Most of the fish collected were sent to Edward Drinker Cope, who did the majority of the work on the fish fauna of what became known as the Green River Formation. Cope travelled west and did some collecting of his own, and in 1871, published a major work on the fishes.

The Green River Formation outcrops in the states of Wyoming, Utah and Colorado and showcases the remains of a large lake system. Scientists have determined that the system was made up of three lakes that varied in size and depth through time. The system existed for just over twenty million years, forming about sixty million years ago and disappearing around the end of the Eocene.

The formation preserves the remains of a subtropical environment. The extremely fine-grained sediments produced beautiful fossils that grace the collections of museums around the world. They also grace the collections of many amateurs, the fish being for sale in almost every rock shop in the country, many museum shops and every souvenir shop in the Midwest. The formation has produced over a million specimens with over one half million in the last thirty years alone.

The Lakes

Fossil Lake was the deepest of the three lakes in the system. It had a small surface area and was short-lived. Fossil Lake produces the largest share of fish specimens of the three lakes.

Lake Gosiute produced a smaller variety of fossil fishes and a different state of preservation as well. It outcrops near the Fontenelle Reservoir and near the Farson Dam. The shales and fossils are not the typical Green River specimens most of us are familiar with.

Lake Uinta was the largest in surface area but was the shallowest. In existence for 17 million years, it was the longest-lived of the three lakes. Its sediments form one of the thickest lacustrine sediments known, almost 7000 feet in some places.

Field Work

The fish are collected from two different units. One, called the "split fish layer," is so called because the layers split, exposing the fossils. This sometimes reduces the quality, as part of the fossil ends up on each side of the matrix. The other unit is called the "eighteen inch layer." It produces the finest specimens because it splits above the specimens, leaving them under a thin layer of rock. Careful preparation exposes a complete fossil.

Eighteen-inch-layer specimens are collected by removing the caprock, then cleaning off the fossil-bearing layers. Once clean, one looks for the outline of a vertebral column covered in matrix. A saw is used to cut a square around the fossil to a depth suitable to reach below the bottom of the specimen. The surrounding rock is then removed, leaving the fossil sticking up off the strata. A blade is then used to pry up the fossil and its surrounding matrix.

Split-fish fossils are collected by selecting a block of suitable size and splitting off thin layers using thin blades. Blades are driven into the block at various points around the block until the layer breaks free. One then carefully examines both sides of the rock for fossils including those that may still be covered with a thin layer of matrix, in which case the line of vertebrae show as a linear bump on the rock surface. A rock saw is used to square up the collected pieces at which time they can be packed for shipping.


Since specimens can end up on very thin fragile slabs of rock, it is sometimes wise to mount the entire piece onto a plywood board cut to match the shape of the specimen before beginning preparation work. Epoxy works well as an adhesive.

The method of preparation varies, depending on whether you are working with a split-fish-layer or eighteen-inch specimen. The split-fish specimens generally require little or no work. One might use an optical magnifier and clean off any stray pieces of matrix still covering parts of the fish. Remove dust with a mild burst of air and finish with a light coat of a clear acrylic spray.

Eighteen-inch specimens require much more work. While the matrix is very soft and easy to remove, the fossils tend to be extremely fragile. The bones tend to be fairly stable but the scales are very delicate. The utmost care must be taken in order to produce a museum-quality specimen.

The thin layer of rock that covers the fossil can be thinned down using one of several methods. An airscribe, mini sandblaster or simply bulk gouging with a large needle all work equally well. One must be careful to leave a thin layer of matrix. The next step should be done under strong magnification. The "scratch and blow" method has proven to be the best way to produce a fine specimen -- a light scratch at the matrix using a fine steel needle and then a burst of air to remove the dust. Patience is the name of the game, as prying up sections of matrix will cause the fish to "split," resulting in a poor outcome. Care must be taken not to put much pressure on the fossil itself, as this will cause pieces of it to flake off. Eventually, the complete fossil will be uncovered and should then be sprayed with a protective acrylic spray. The spray is important, as prolonged exposure to humidity can have an adverse effect on the quality of the fossil.

The Fish

While the formation preserves over twenty species of fish, five constitute over 98% of collected specimens. We will confine our discussion to common types.


This is probably one of the most common fossils in the world. [Editor's note: I read recently that it is indeed the most common vertebrate fossil thus known.] The sheer numbers boggle the imagination. In 1978 alone, over 20,000 were collected. You can find them for sale at any place that offers natural history items.

Two species are known with K. eocaena being the most common. A member of the herring family, they average five inches in length with a maximum size of ten inches. They have a deeply forked tail and a single dorsal fin in front of the mid-body line. Knightia fed on algae, ostracods and insects and they were the major food source for many of the larger fish from the lakes. They were schooling fish and are frequently found in mass mortality or "death" layers. Fossils have been found packed as densely as several hundred fish per square meter of slab rock.


Cope described two species of this perch-like fish, however, present-day researchers doubt the existence of M. sauvagenus. With the only specimen found having been lost, no further study is possible. M . labracoides is a fairly common predator easily distinguished by the two dorsal fins, large fan-shaped tail and an anal fin equal and opposite its second dorsal fin. A voracious predator, Mioplosus is the fish most commonly found eating other fish, something for which the Green River fish fossils are famous.


The upturned mouth of this herring type fish indicates that it was probably a surface feeder. They have a wide anal fin, single dorsal fin, and a deeply forked tail. The maximum size is listed at twenty-six inches, but they are more commonly three to six inches. D. dentatus is also known from the Cretaceous of South America. When Cope originally erected the genus, he listed four species, however, more recent research has pared it down to D. dentatus only. Another Knightia predator, "Diplos" have been found containing fossils of their last meals.


Priscacara has grinding-type teeth, indicating that it probably fed on snails and crustaceans. It is easily identified by its stout dorsal and anal fin spines, which, in my opinion, make it the most attractive of the Green River fossils. Two species, P. serrata and P. liops, are the most common. Several others are under review by current researchers. Priscacara are known only from the Eocene. They have a large oval body and range in size from one inch to fifteen inches with four to six most common. They were schooling fishes and are more commonly found in the eighteen-inch layer.


The largest of the common fishes is Phareodus. Two species, P. encaustus and P. testis, are known. They average fifteen inches, with a maximum of thirty inches. With the dorsal and anal fins located at the very back of the body, these fish cannot be confused with any other Green River species. They have large very sharp teeth and have been found with the remains of Mioplosus and Priscacara inside.

Collecting Opportunities

Fortunately, gaining access to collecting sites in the Green River Formation is easy. One has a choice of several quarries open to fee collecting. I've collected in three. Years back I spent a few hours at Warfield's Quarry and I collected at Ulrich's about three years ago. Most recently I collected at Antares Fish Quarry. All three are commercial quarries that allow individuals to collect for a fee. These quarries are usually leased from local landowners or governments for fossil collecting rights. Overburden is bulldozed off and you are given an area to work. Ulrich's provides more assistance but charges a higher fee. You drive your own vehicle into Antares and Warfield, while Ulrich drives you to the site in their four-wheel drives. There are others in the Kemmerer, Wyoming area; I suggest you check them all out and pick the one that best suits your situation.

The Green River formation is a world-class site that offers amateur collectors the opportunity to collect museum-quality specimens with limited expertise, expense, and effort.


Grande, Lance, 1984. Paleontology of the Green River Formation, with a Review of the Fish Fauna. Geological Survey of Wyoming, Bulletin 63.