Postglacial fragments, featuring Hartman's Cave

Although eastern North America contains many things, it is not noted for its terrestrial Cenozoic sedimentary deposits. Conditions just weren't favorable for the long-term preservation of extensive terrestrial sedimentary formations. Therefore, our knowledge of terrestrial life in this region is largely confined to some transitional coastal settings and what we might call "point sources" in comparison to the great formations of the West: sedimentary deposits of caves, fissures, bogs, ponds, and so on. These in turn are strongly biased to just the Pleistocene and Holocene. We've visited a few of these sites before, around Minnesota (Hidden Falls, I-35, Kirchner Marsh, Loring Park, and in general), and in the District of Columbia, Kentucky (Mammoth Cave), and Pennsylvania (Marshalls Creek Mastodon, Port Kennedy Bone Cave). Here is another site in northeastern Pennsylvania, plus some brief commentary on sites nearby.

If you "scratch" a geographic area long enough, eventually you start coming up with reports of Quaternary fossils: a mammoth tooth here, a bison skull there. It's usually big obvious stuff found at or near the surface, or some bones in a cave if the area has caves, and fairly typically there's some initial publicity (maybe even a paper or two) and then the find is largely forgotten, dwindling into the background noise of the local community to be occasionally resurrected in a newspaper article, or a school project, or by some strange person like me.

By far the most famous of the terrestrial finds in the vicinity of Delaware Water Gap is our old acquaintance, the Marshalls Creek Mastodon, but it is not alone. Less complete fossils of another mastodon were found at Shotwell Pond in New Jersey in the 1930s (Jepsen 1960; Sussex County in New Jersey is well-supplied with mastodons in general). The "elk moose" Cervalces scotti is represented by a couple of good specimens found within a few miles of the water gap, near Hainesburg and Mount Hermon, New Jersey (Parris et al. 1987). There are also several bog/pond microfossil sites, such as Francis Lake (Cotter et al. 1985), Glovers Pond (Erickson 1988), Saddle Bog (Sirkin and Minard 1972), Silver Lake (Zelanko et al. 2012), Tannersville Bog (Cai and Yu 2011), and White Lake (Yu 2007), all of which are in New Jersey except for Tannersville Bog. The most prolific vertebrate site in the vicinity of the water gap, though, was a site known in the literature as Hartman's Cave (Leidy 1880, 1889; Darton 1885; Mercer 1894; Hay 1923). The cave itself is still extant, but it is now protected bat habitat and the fossiliferous sediments have been heavily excavated and otherwise disrupted.

The fossils of Hartman's Cave first show up in the literature in a brief report delivered to the Academy of Natural Sciences of Philadelphia on September 21, 1880, by eminent vertebrate paleontologist Joseph Leidy. Leidy by this time had given up dinosaurs due to the "Bone Wars", and we find him here in an activity with lower stakes and less pressure: describing cave material that had a mix of paleontological and archeological specimens. A Mr. T. Dunkin Paret had begun investigating and excavating the cave, and invited Leidy and a Dr. T. C. Porter to examine the results. Leidy observed that the cave originally contained a bed of water-lain unfossiliferous clay about 10 feet thick, succeeded by rocky fragments, and then about a foot of windblown earth containing organic remains. For a significant part of the cave the sediment fill was too thick to allow large animals, but there was certainly enough space for smaller animals, and many of the bones had been gnawed by rodents or small carnivores, which probably also transported the pieces of large bones. Leidy reported that "about half a bushel" of bones and fragments had been obtained.

The fauna Leidy observed was primarily extant animals living in the area, with a few exceptions. Representing extinct animals were some teeth of the giant beaver Castoroides and partial jaws of a peccary. Of animals extirpated from that part of Pennsylvania, there were some caribou bones, a fragment of bison jaw, and a great deal of woodrat/packrat pieces, representing Neotoma magister or floridana (apropos of nothing: the Allegheny woodrat, Neotoma magister, was first described scientifically from Pennsylvania cave fossils in 1857, and not known to be a living species until decades later [Rhoads 1894; Poole 1940]). In hindsight, the abundance of Neotoma suggests packrat middens, albeit in an area too humid for rat-induced midden cementation.

Remains of living animals show about the kind of fauna a person might expect to see in eastern Pennsylvania: terrestrial snails, freshwater mussels, turtles, snakes, turkeys, moles, various rodents, rabbits, bats, bobcats, foxes, wolves, skunks, weasels, raccoons, deer, and elk. Molars of a fetal or newborn horse were interpreted by Leidy as potentially representing a modern domestic animal. Mixed in were some charcoal fragments and seeds of dogwood, pignut hickory, and walnut. A few human artifacts were found as well, including a few bone awls and other small bone and antler implements (some of which had been damaged by gnawing), and most interestingly a bored shell of the snail Conus tornatus (now Conasprella tornata), found today on the Pacific coast from Mexico to Peru. This would indicate long-distance trading networks. Leidy compared the site to Durham Cave, another small bone-producing cave, albeit apparently destroyed by 1880 (this site was in Riegelsville, Pennsylvania, about 20 miles south of the water gap [Hay 1923]).

This was the state of the site in 1880. Darton (1885) published a review that was in large part a recapitulation of Leidy's article. Leidy revisited the topic in an 1889 publication. (Just to make things confusing, the cave was also known as the Crystal Hill cave by this time, and Leidy occasionally mixed his usage.) With more space to work with, he expanded upon the artifacts and added information about the number of specimens recovered for each species, as well as adding a few illustrations. If you're curious, judging by lower jaws the most abundant animals were woodrats, followed by raccoons, woodchucks, and skunks, each represented by at least 44 to as many as 92 mandibular rami (briefly, the part of the lower jaw that sticks up and includes the articulation with the upper jaw), with nothing else represented by more than 11. None of these four seem particularly surprising to find holed up in a cave. By this time he now suggested that the horse teeth might represent an extinct indigenous species rather than the modern domestic horse. He also decided that the peccary represented a new species, which he named Dicotyles pennsylvanicus. This species is still occasionally mentioned today, under the genus Mylohus, but is usually sunk into another species.

Plate II from Leidy 1889; it was spread over two pages, which I put together, leaving a light strip to show the join (and also because all the versions I could find were missing the innermost edges, so it wouldn't be accurate to pretend there is no seam). Most of the elements are from Hartman's Cave: Figures 3–6 are of Dicotyles pennsylvanicus, Figures 7–9 and 13–20 are of the Hartman's Cave Castoroides, Figure 21 is of a beaver incisor, and Figure 22 is of caribou lower premolars and molars. Included for comparison, Figures 1 and 2 are of the peccary Platygonus vetus from Mifflin County, Pennsylvania, and Figures 10–12 are of a Castoroides incisor from Illinois.

Since Leidy's two publications, the only substantial report on the cave fossils has been an 1894 article by Henry Chapman Mercer (ignoring shorter extracts), who was also working on the Port Kennedy Bone Cave at about the same time. Actually, to describe the paper as if it is about the cave fossils is misleading, because the main areas of focus are the stratigraphy of the deposits and the implications for the archeology. The excavations that produced Leidy's fossils were not done the way we'd do them now, especially if there was the chance of finding human artifacts with extinct Pleistocene megafauna, and Mercer recognized that there was essentially no way of determining associations from the excavated specimens and the spotty documentation (to say nothing of what enterprising packrats can do by transporting small tools from the vicinity of the cave and redistributing them). Mercer found that practically all of the upper, bone-and-artifact-bearing unit had been excavated by 1893, except for fringes near the opening, but conducted some additional excavations in the talus beyond the opening and in the clay within the cave.

A transect across the cave 40 feet beyond the opening found that the clay deposit was indeed unfossiliferous fine clay, deposited in thin layers, with the occasional sand pocket. Mercer did come across a layer in the talus containing human artifacts and bones of small animals, but found it difficult to correlate any features within the talus pile outside of the cave with the much thinner and by then lost debris layer inside the cave. He could say nothing definitive about whether or not humans had killed and eaten the extinct animals represented in the cave fauna except to note it was possible, and also observed that he had no way of knowing how long the extinct animals had been extinct. (Curiously, Mercer described Castoroides not as a "giant beaver", but as a "giant chinchilla". In terms of dignity, Castoroides has had a hard time catching a break.) Mercer also made a point of noting that there was nothing in the artifacts to suggest they hadn't been made by the known native people of the Delaware Valley. Most of the bones were probably there thanks to small animals.

Mercer (1894)'s depiction of the interior of the cave. He didn't explicitly identify what each of the letters stood for, but the dotted line from "a" to "d" is an approximation of the original surface of the excavated debris layer, "f" represent his excavations in the talus "k", "g" is the transect through the clay layer "h", and "i" is the cave bedrock.

The last issue Mercer considered was how the clay got there in the first place. He found it unlikely that the clay was simply the result of decomposition of the cave rock, and could see no evidence for recent fluvial transport from the outside, or for that matter any evidence the cave was exposed to the outside while the clay was forming. He hit upon the idea that the cave, which was within a glaciated area, had been blocked by ice and filled by muddy water draining from the glaciers, making the cave fill essentially glacial flour, but wasn't entirely confident in the idea. This hypothesis would also mean that the organic remains accumulated entirely after the glacial retreat, in the latest Pleistocene and Holocene.

References

Cai S., and Yu Z. 2011. Response of a warm temperate peatland to Holocene climate change in northeastern Pennsylvania. Quaternary Research 75(3):531–540.

Cotter, J. F. P., J. C. Ridge, E. B. Evenson, W. D. Sevon, L. Sirkin, and R. Stuckenrath. 1985. The Wisconsinan history of the Great Valley, Pennsylvania and New Jersey, and the age of the “terminal moraine”. Pages 1–58 in E. B. Evenson, organizer. Woodfordian deglaciation of the Great Valley, New Jersey. Guidebook for the Friends of the Pleistocene Field Conference 48.

Darton, H. 1885. Ancient bone cave in Pennsylvania. Scientific American Supplement to 1885, 472:7541–7542.

Erickson, J. M. 1988. Fossil oribatid mites as tools for Quaternary paleoecologists; preservation quality, quantities, and taphonomy. Bulletin of the Buffalo Society of Natural Sciences 33:207–226.

Hay, O. P. 1923. The Pleistocene of North America and its vertebrated animals from the states east of the Mississippi River and from the Canadian provinces east of longitude 95°. Carnegie Institution of Washington Publication 322.

Jepsen, G. L. 1960 [1962 and 1964 editions as well]. A New Jersey mastodon. New Jersey State Museum, Trenton, New Jersey. Bulletin 6.

Leidy, J. 1880. Bone caves of Pennsylvania. Proceedings of the Academy of Natural Sciences of Philadelphia 32:346–349.

Leidy, J. 1889. Notice and description of fossils in caves and crevices of the limestone rocks of Pennsylvania. Annual Report of the Geological Survey of Pennsylvania for 1887:1–20.

Mercer, H. C. 1894. Re-exploration of Hartman’s Cave, near Stroudsburg, Pennsylvania, in 1893. Proceedings of the Academy of Natural Sciences of Philadelphia 1894:96–104.

Parris, D. C., S. S. Albright, and W. B. Gallagher. 1987 [or 1990]. Road log, paleontological sites on the Ordovician, Silurian and Devonian systems of the Delaware Water Gap region. Pages 1–11 in W. B. Gallagher, editor. Paleontology and stratigraphy of the lower Paleozoic deposits of the Delaware Water Gap area. Geological Association of New Jersey, Newark, New Jersey. Annual Field Conference 4.

Poole, E. L. 1940. The technical name of the Allegheny woodrat. Journal of Mammalogy 21(3):316–318.

Rhoads, S. N. 1894. A contribution to the life history of the Allegheny cave rat, Neotoma magister Baird. Proceedings of the Academy of Natural Sciences of Philadelphia 46:213–221.

Russell, E. W. B., R. B. Davis, R. S. Anderson, T. E. Rhodes, and D. S. Anderson. 1993. Recent centuries of vegetational change in the glaciated north-eastern United States. Journal of Ecology 81(4):647–664.

Sirkin, L. A., and J. P. Minard. 1972. Late Pleistocene glaciation and pollen stratigraphy in northwestern New Jersey. U.S. Geological Survey, Washington, D.C. Professional Paper 800-D:D51–D56.

Yu Z. 2007. Rapid response of forested vegetation to multiple climatic oscillations during the last deglaciation in the northeastern United States. Quaternary Research 67(2):297–303.

Zelanko, P., Yu Z., G. E. Bebout, and A. J. Kaufman. 2012. Multiple early Holocene climate oscillations at Silver Lake, New Jersey and their possible linkage with outburst floods. Palaeogeography, Palaeoclimatology, Palaeoecology 350–352:171–179.