All About Serpentine
Serpentine is an intriguing and complex subject, of interest from the casual observer to the research scientist. With the goal of offering accurate and up-to-date information on the subject of Soldiers Delight and its serpentine barrens, look here for content on this fascinating subject.
These articles are meant to inform all readers. Each discusses its subject matter in detail, for the benefit of practicing conservationists, students, teachers, journalists and anyone seeking an in-depth analysis of what make serpentine so intriguing to study. Additional information can be found in publications listed in the references cited in each article.
Please be forewarned that, more often than not, Google searches will result in out-of-date studies and erroneous information. Unfortunately, this leads to perpetuation of misconceptions about serpentine. The articles we present here offer a wealth of accurate information, yet the reader is encouraged to explore even further. Biosis Previews and Georef databases are especially important to finding up-to-date, firsthand results of peer-reviewed research publications. These databases and others are available online from major universities and libraries.
Readers are invited to leave comments and questions on our "contact" page, and to recommend future topics you’d like to see featured on our website.
Summary: Soldiers Delight probably originated from the upper mantle of the Earth, during convergence of an ancient oceanic plate with the continental plate which preceded North America, called Laurentia. It is part of a large fragment of an oceanic plate which was thrust upon the edge of Laurentia during the collision, becoming part of the continent. This fragment of the oceanic plate, termed ophiolite (“snake stone”), consists of crust and underlying upper mantle. Based on current science, Soldiers Delight is from the upper mantle part of the ophiolite. This ophiolite, named the Baltimore Mafic Complex, formed during development of the Appalachian Mountains more than 450 million years ago, about 200 million years before the Age of Dinosaurs.
Discussion: Tectonic plates are large sections of the outermost shell of the Earth comprised of crust and adjoining upper mantle, collectively called lithosphere (Alexander et al. 2007). Oceanic plates are heavier than continental plates and, therefore, almost always slide under (subduct) continental plates during the Wilson Cycle. Infrequently, however, a fragment of an oceanic plate is thrust onto the margin of a colliding continental plate and is called an ophiolite (“snake stone”). During the past 500 million years, less than 0.001 % of the crust generated at mid-oceanic ridges, worldwide, has been incorporated into continental margins. None of the Atlantic Ocean lithosphere has been incorporated into any continental margin during the last 160 million years (Dilek 2003).
Soldiers Delight is part of the largest ophiolite in the central and southern Appalachians, the Baltimore Mafic Complex. This complex stretches from central Maryland into southern Pennsylvania (Hunan and Sinha 1989). Obduction of the Baltimore Mafic Complex occurred around 450 million years ago, about 200 million years before the Age of Dinosaurs (Smith 2008).
Although some ophiolites are preserved in their original condition, the Baltimore Mafic Complex became very fragmented over the span of hundreds of millions of years. Soldiers Delight originated from upper mantle in one of the dismembered portions of this complex (Guice et al. 2021). Upper mantle is comprised of ultramafic rocks; i.e., rocks high in magnesium and iron (Alexander et al. 2007, 2009). Soldiers Delight is serpentinite (serpentine rock) which formed from one of these ultramafic rocks, peridotite (Alexander et al. 2009). That is, the minerals that formed peridotite (olivine and pyroxenes) were changed to serpentinite minerals (antigorite, chrysotile, and lizardtite) by certain elevated temperatures in the presence of water, such as seawater. (Magnetite is also produced from the parent rock.) Serpentinite is still ultramafic because Mg was not lost during metamorphosis. However, Ca was lost during the conversion to serpentine, leaving serpentine rock and derivative soil with very high Mg:Ca ratios, with dramatic consequences to plant life as discussed in subsequent sections. Based on five samples from three locations studied by Guice et al. (2021), the parent ultramafic rock of Soldiers Delight has been transformed to serpentine rock with a Mg:Ca ratio of about 6:1. Most plants grow best at a Mg:Ca ratio of about 1:1, and the average ratio of Mg:Ca on continents is about 0.7 (Alexander et al. 2007).
Ophiolites have generated many geologic studies since they contain portions of the Earth too deep to reach by exploratory drills. They have been invaluable in understanding the complex geologic history of different parts of the Earth. In addition, they have contributed to the development of plate tectonic theory as the presence of oceanic rock on continents could not be explained by earlier theories. Types of ophiolites are discussed and diagrammed in Dilek and Furnes (2011), and evolution of the concept is presented in Dilek (2003). The Baltimore Mafic Complex has recently been interpreted as a suprasubduction zone type of ophiolite, meaning that the rocks formed in association with a continental margin, rather than a mid-ocean ridge (Guice et al. 2021).
The distribution of ultramafic rocks within the Appalachian Mountains is diagrammed in Rajakaruna et al. (2009). It should be noted that not all ultramafic rocks are associated with ophiolites, and the origin of many ophiolites remains unknown or debated, especially in the central and southern Appalachians (Guice, personal communication, January 2022).
Alexander, E.B. 2009. Serpentine geoecology of the eastern and southeastern margins of North America. Northeastern Naturalist 16 (Special Issue 5) : 223-252.
Alexander, E.B., R.G. Coleman, T. Keeler-Wolf, and S. Harrison. 2007. Serpentine Geoecology of Western North America. Oxford University Press, Inc. New York, New York.
Dilek, Y. 2003. Ophiolite concept and its evolution. In: Dilek, Y. and S. Newcomb (eds.). Ophiolite concept and the evolution of geological thought. Geological Society of America Special Paper 373:1-16.
Dilek, Y. and H. Furnes. 2011. Ophiolite genesis and global tectonics: geochemical and tectonic fingerprinting of ancient oceanic lithosphere. Geological Society of America Bulletin 123(3/4):387-411.
Guice, G.L., M.R. Ackerson, R.M. Holder, F.R. George, J.F. Browning-Hanson, J.L. Burgess, D.I. Foustoukos, N.A. Becker, W.R. Nelson, and D.R. Viete. 2021. Suprasubduction zone ophiolite fragments in the central Appalachian orogen: Evidence for mantle and Moho in the Baltimore Mafic Complex [Maryland, USA]. Geosphere 17(2):561-581.
Hanan, B.B. and A.K. Sinha. 1989. Petrology and tectonic affinity of the Baltimore mafic complex, Maryland. In: Mittwede, S.K. and E.F. Stoddard (eds.) Ultramafic rocks of the Appalachian piedmont. Geological Society of America Special Paper 23:1-18.
Rajakaruna, N., T.B. Harris, and E.B. Alexander. 2009. Serpentine geoecology of eastern North America: a review. Rhodora 111(945):21-108.
Smith, II, R.C. and J.H. Barnes. 2008. Geology of the Goat Hill serpentine barrens, Baltimore Mafic Complex, Pennsylvania. Journal of the Pennsylvania Academy of Science 82(1):19-30.
Summary:The term "barrens" was used by English surveyors and settlers to refer to areas which were "bare of timber." "... when the whites commenced settling here, they found no timber, hence they applied the term Barrens, a common appellation at that time, to such portions of the country, however fertile the soil." Tree species, especially oaks, were present but open grown, with trunks too short for timber. The landscape was predominantly oak savanna maintained by frequent autumnal fires ignited by Native Americans mainly for fire-hunting white-tailed deer. Prior to settlement, beginning about 1750, the serpentine barrens covered expansive areas in Baltimore, Harford, and Carroll Counties of Maryland, and in adjacent York and Lancaster Counties of Pennsylvania. Access to the barrens was provided by a system of Amerindian "highways," with a main artery called the Old Indian Road. Native American fires ended about 1730 as American Indians succumbed to European diseases and other factors. Livestock grazing quickly became widespread on the barrens, and Soldiers Delight was a favorite range for stock. Barrens which were not grazed, burned, or cut by settlers soon developed trees with long trunks suitable for timber. By 1800, timber was advertised in Annapolis and Baltimore newspapers as growing in the barrens, and all timber stands in Soldiers Delight were logged before 1914. Also in the 1800s, chromite was mined from five shaft mines and from chromite-rich alluvial sand deposits (placers). After 1930, the non-indigenous Virginia pine expanded rapidly across Soldiers Delight, in the absence of grazing and fire. In just 50 years, Virginia pine dominated more than 50 % of areas which had been in native serpentine vegetation. Ecological restoration of the serpentine ecosystem began in 1989 as a six-year research phase, and expanded in 1995 with the passage of a restoration and management plan. By 2020, more than 500 acres had been manually cleared of pine and more than 100 acres burned at least once. White-tailed deer management began on a limited basis in 2008 and fully expanded beginning in 2014. Invasive species management began in the 1990s, after thousands of tree-of-Heaven (Ailanthis altissima) trees were discovered in Soldiers Delight. A number of other invasive plants continue to be managed annually.
Historical vegetative conditions on the barrens in Maryland and Pennsylvania have been described by Maryland historian William Bose Marye (1886-1979, pronounced Marie) (Marye 1920, 1955a, b, c). Marye determined that the term "barrens" was used by English surveyors and settlers to refer to areas which were "bare of timber" (Marye 1955b). "... when the whites commenced settling here, they found no timber, hence they applied the term Barrens, a common appellation at that time, to such portions of the country, however fertile the soil." The term barren was applied to individual exposures of serpentine as well as the vast serpentine-centric region which included areas of adjacent non-serpentine soils.
Prior to settlement of the barrens (beginning about 1750), they covered expansive areas in Baltimore, Harford, and Carroll Counties in Maryland, and in adjacent York and Lancaster Counties of Pennsylvania. In 1722, the barrens of central Maryland were described in correspondence as "... a Vast Body of Barrens ... what is called so, because there is no wood [timber] on it; besides Vast Quantities of Rockey Barrens ...". "... the Lands ... all along the west side of Baltimore Co[unty], are cut off & separated ... by large Barrens, many miles over..." (Marye 1955a). Soldiers Delight was described as "... an immense barrens" circa 1730 (Marye 1920). The following description occurred in a 1753 letter: "... about thirty miles from Navigable Water is a Range of barren dry Land without Timber about nine miles wide which keeps a Course about North East and South West parallel with the mountains thro this province Virginia & Pennsilvania..." (Marye 1955a). In 1771, 620 acres in Baltimore County were described as "Verry poor bare Barrens"; "Barrense, hilly and stony"; "Poore hilly Barrance & much broke with stone & Verey scarce of Timber"; and "exceedingly poor & much broke with stone and Little or no Timber of any sort." (Marye 1955a). Similar descriptions of other barrens in 1770 are provided in Marye (1955a).
Barrens in Pennsylvania were noted by a Quaker in 1683 as extensive treeless spaces in the wilderness (Marye 1955a). In 1737, they comprised about 130,000 acres, and their width at the Maryland border was reported to be 20 miles (Marye 1955b). This historical estimate correlates well with the 23-mile strip of ultramafic bedrock mapped by Pearre and Heyl (1960).
Between 1580 and 1652, serpentine in Maryland was within the hunting domain of the Susquehannock Indians (Marye 1955c). Their villages were concentrated mostly on both sides of the lower Susquehannock River in York and Lancaster Counties, Pennsylvania, near the Maryland border (Cadzow 1936, Witthoft 1959). (The Shenk's Ferry culture preceded the Susquehannocks in this region (Witthoft 1959)). "Fire hunting" was the primary technique for harvesting wildlife on serpentine (Marye 1955b), as it was in a variety of habitats throughout the Southeast (Maxwell 1910, Swanton 1946). The following quotations are from Marye's transcriptions of historical documents (Marye 1955b). "It was the custome of the Indians in the autumn to set fire to and burn the barrens of York [Pennsylvania] and Baltimore [Maryland] Counties...". "When the Indians no longer set their fires, trees began to creep back...". Another source, in reference to the York Barrens of Pennsylvania, stated "...that the Indians for many years, and until 1730 or 1731, to improve this portion of their Great Park for the purpose of hunting, fired the copse or bushes as often as their convenience seemed to call for it...".
Between 1652 and 1730, hunting in the barrens was shared with other tribes (Marye 1955c). By 1675, the Susquehannocks had succumbed to European diseases and to warfare with both settlers and Iroquois Indians (Hunter 1959, Witthoft 1959). They disappeared as a distinct group by 1680, and the last identifiable descendent was killed in 1763.
Access to the barrens was provided by a system of Indian "highways." The Old Indian Road (Marye 1920), the main travel artery, passed through the barrens region and was within 3.5 miles of Soldiers Delight to the north, east, and west. A resident of a Baltimore County plantation gave sworn testimony in 1697 that it was located "in the 'walks' which Indians usually take when they move to their hunting Quarters ..." (Marye 1955c). The plantation "was situated on or very near a highway followed by Indians in going to or returning from certain hunting grounds or when travelling 'on the warpath'," and the "highway" was commonly referred to as the Indian Road or Old Indian Road in a "considerable number" of eighteenth-century records (Marye 1920). Fire was probably used for other reasons such as communication and warfare (Moore 1972).
White-tailed deer were probably the primary object of fire-hunting, based on archaeological evidence and the widespread distribution of deer. Donald Cadzow (1936) led archaeological excavations of Susquehannock sites along the Susquehanna River in 1930 and 1931. He reported the contents of 31 food storage pits at the Schultz site, located in one of the largest Indian villages. White-tailed deer bones were discovered in 61 % of the storage pits. Elk bone were found in 32 % of the pits, and other bone included fish, black bear, beaver, raccoon, and turkey. Buffalo bone was not found in storage or fire pits at the Schultz site and not reported from the other four sites studied. At the Strickler site, Futer (1959) excavated a large midden-filled pit and found many bones of deer and bear. The regional abundance of elk before settlement is uncertain (Mansueti 1950 and Paradiso 1969), and the presence of elk herds at Soldiers Delight was not reported in the historical literature.
Fire-hunting improved the harvest of low-density wildlife populations compared to other hunting techniques. According to a review by Knox (1997), the white-tailed deer population prior to English arrival in 1607 was only about 8.0 – 10.9 deer/mi2 in neighboring and physiographically similar Virginia. For comparison, in March 2008, a helicopter-mounted forward-looking infrared camera (FLIR) survey estimated a density of 93 deer/mi2 in Soldiers Delight (Helicopter Applications, Inc. 2008).
European settlement of the barrens region was delayed until the 1740s (Porter 1975, 1979). Among factors causing this delay were the psychological effect of a vast barren landscape, scarcity of timber, and lack of an effective treaty with the Five Nations of Iroquois before 1744. After Indian extirpation, barrens were used by settlers for livestock grazing (Marye 1955c). In addition, "... lands which lay within easy distance of the Barrens, were considered to be more valuable on that account." (Marye 1955c). For example, one parcel was advertised in the Maryland Gazette (1746) as "...convenient for stock, there being an outlet to the Barrens of Patapsco...". Marye concluded that this outlet must have been by way of Soldiers Delight. The name of "Graziers Delight", 892 acres, surveyed on Soldiers Delight in 1774 implied that the barrens of Soldiers Delight were a favorite range for stock (Marye 1955c). The longevity of grazing as an ecological factor controlling woody plants has not been documented. Although the use of fire by settlers to maintain and create grazing land may have been widespread initially, the practice was "...largely abandoned..." by 1780 (Marye 1955b). Grazing may still have been a factor in the 1800s and into the early 1900s, based indirectly on historical vegetation descriptions and photographs (Tyndall and Hull 1999). However, information on grazing intensity and distribution have not been published.
Barrens which were not grazed, burned, or cut by settlers soon developed trees with long trunks suitable for timber. By 1800, timber was advertised in Annapolis and Baltimore newspapers as growing in the barrens (Marye 1955a). Before settlement, the landscape was described as "... acres upon acres overgrown with nothing but saplings; other considerable areas, with bushes only; still others, denuded and bare." (Marye 1955a). After settlement, "... the sapling lands produced timber trees, hardwood seedlings sprang up on the bushey grounds, and the Barrens vanished..." (Marye 1955a). Rapid timber development within a barren was probably from tree species already in place on deeper silt loam soils ("sapling lands"); i.e., not from afforestation by oaks from surrounding non-serpentine areas. "Bushey" may have referred to stunted oaks on shallower silt loam or deeper gravelly sandy loam, and "denuded and bare" to gravelly sandy loam too shallow for woody plant root systems. However, the definition of these terms was not provided in Marye (1920, 1955a, b, c). Referring to the "York Barrens of Pennsylvania," one source stated "Portion of the country that were sixty or seventy years ago [1775-1785] without timber are now  thickly covered with sturdy oaks and large hickories." (Marye 1955b). Since hickories do not grow on serpentine, this quote apparently includes non-serpentine areas adjacent to serpentine barrens.
In 1910, Shreve et al. reported that the "Barrens ... in the Soldiers Delight area ... have an open park-like stand of trees ... The Black Jack Oak and Post Oak are often the sole trees on the thinnest soil, or they may be accompanied by Red Cedar." In a corresponding photograph, Virginia pine is not evident nor is red cedar (first reported by Knox ). Oak trees are scattered or absent on ridges of various aspect. In 1914, the State Forester (Besley 1914) did not identify seedlings or stands of Virginia pine in Soldiers Delight. Only stands of "culled hardwoods" were mapped indicating that all hardwood forests had been logged. In 1929, "The desolation of the serpentine barrens around Soldiers Delight, with its rocky soil and stunted vegetation of cedar and meagre grass..." was reported by the Maryland Geological Survey. And also, "... characterized by a scanty vegetation of pine and cedar growing in a thin grey soil mantled in spring with the purple ground pink ...". In 1937, during an automobile drive along Deer Park Road, Bowen states, "All about us are small ridges and steep barren slopes ... The road winds along the barren hillsides, through occasional clumps of bushes ... The scrub oaks grow with gnarled trunks and have very glossy dark green leaves. Sassafras bushes dot the hills in little communities in spite of the very thin covering of soil...".
Mining for chromite could have contributed to treeless conditions in small localized areas, but widespread surface mining was not employed due to the nature of chromite deposits. Merchantable quantities of chromite were found in perennial stream beds, at five primary upland locations, and in scattered small shallow pits. Mining of chromite-rich alluvial sand deposits ("placers") in Soldiers Delight began between 1810 and 1820 (Pearre and Heyl 1960), and some level of production continued into the early 1920s (Johnsson 2017). Placer mining consisted of hand-digging chromite-rich sand deposits in streambeds and banks, and manually removing much of the sand with a screen mesh (Singewald 1928). The sifted material was then carried to a troughlike wooden structure, called a buddle, for final separation of chromite grains. Buddles were about 12 feet long and a foot wide positioned along the stream with a small dam to provide for water flow down the buddle. The lighter chromite grains would concentrate at the end of the buddle, and repeated washing would increase the concentration.
Woody plant growth could also have been impacted in the vicinity of the five shaft mines for lode chromite. Records support the beginning of Choate Mine ore production soon after 1825 (Johnsson 2017). Operation of each mine ended by about 1880, with the exception of an aborted restart of the Choate Mine in 1917-1918 (Johnsson 2017, Pearre and Heyl 1960). Peak periods of activity at the Choate Mine were the 1830s and 1865 - 1875 (Pearre and Heyl 1960). The number of mine shafts at each mine ranged from one to four, and three mines had concentrating mills and other structures (Johnsson 2017, Singewald 1928).
After 1930, Virginia pine invasion and expansion rapidly ensued (Tyndall 1992a, b). In 1937-1938 aerial photographs of the U.S. Department of Agriculture, pines are widely scattered in upland areas of Soldiers Delight. But in just 50 years, Virginia pine dominated more than 50 % of areas which had been in grassland or oak savanna vegetation. Pines became established first in the deeper and wetter soils of steep ravines and floodplains and then spread into shallow ravines and upper and lower ends of ridges. Virginia pine usually invaded the midsection of south-facing slopes last as they were hotter and drier than other habitats, leaving a landscape impression of scattered "openings" surrounded by pine.
The origin of invading Virginia pine is not known. Preliminary study of a single sediment core from a small depression of unknown origin in serpentine near the perimeter of Soldiers Delight suggested that Virginia pine was present in small numbers as early as 1810 (Hilgartner et al. 2009). Regardless of the source(s), the absence of fire, grazing, mining, and other disturbance activities, in conjunction with abundant bare mineral soil coverage which is favorable to pine seedling establishment, allowed Virginia pine to expand rapidly as a non-indigenous invasive species.
Recent Conditions at Soldiers Delight
Ecological restoration of the serpentine ecosystem began in 1989 as a six-year research phase. Three research areas were selected to verify predicted indigenous vegetation response to the clearing and fall burning of Virginia pine invaded areas. Spring burning was not tested since historical fires were autumnal (Marye 1955). Results for 1989–1992 were presented in Tyndall (1994), while permanent-plot sampling continued in 1993 and 1994 and reported in Tyndall (2020). Research results were consistently positive and warranted implementation of a restoration and management plan in 1995 for the entire Natural Environment Area. Based on research results, the strategy was the manual removal of pines during winter followed by prescribed burning in fall. Spring burning was also implemented in greenbrier-dense areas. By 2020, more than 500 acres had been cleared of pine and more than 100 acres burned at least once. Vegetation response to clearing and burning was robust as predicted by research results and the scientific literature. The rate of prescribed burning lagged behind clearing mainly because of smoke management limitations in a rapidly developing landscape.
White-tailed deer began to be conspicuous in 1994 during daylight hours, and excessive deer herbivory became a serious problem. By 2007, nearly 100% of oak seedlings were being browsed, and herbaceous flowers were noticeably uncommon. Most plants of gray goldenrod (Solidago nemoralis) and serpentine aster (Symphyotrichum depauperatum), a serpentine endemic and State Endangered species (Gustafson and Latham 2005, Maryland Department of Natural Resources 2019), were consumed to a height of only a few inches aboveground. In March 2008, a helicopter-mounted forward-looking infrared camera (FLIR) survey estimated a density of 93 deer/mi2 (36 deer/km2) (Helicopter Applications, Inc. 2008). In an attempt to reduce deer density, a two-day managed firearm hunt was conducted in January 2008 and continued annually through 2016. Since signs of herbivory continued to be extreme, a four-night sharpshooter harvest was conducted during March 2014 (USDA 2014), and State regulated public hunting was expanded beginning with the September 2014–January 2015 hunting season. This expanded public hunting program has continued annually (Maryland Department of Natural Resources 2020).
In 2019 serpentine aster had recovered to within about 50% of 1994 levels, thanks to the sharpshooter harvest in conjunction with regulated public hunting (Tyndall 2020). In addition, a resurgence of oak seedling and sapling growth was evident. Although showing abundance in some areas in 2020, gray goldenrod may be slow to recover in areas where seed banks were exhausted by deer.
Invasive species management began with tree-of-Heaven (Ailanthis altissima) in the 1990s, after thousands of trees were discovered in Soldiers Delight. The largest trees had been planted on-site decades earlier and provided abundant seed for rapid spread. Other seed sources were large trees on adjacent properties. Over 10,000 individuals were killed, and this invasive is now in the early detection-rapid response level of management. Mile-a-minute-vine (Persicaria perfoliata), once a major problem like tree-of-Heaven, has been reduced to a minor problem thanks to assertive management in conjunction with the spread of two biocontrol weevils from off-site introductions by the Maryland Department of Agriculture. During 2018 - 2020, Japanese barberry (Berberis thunbergii), sericea lespedeza (Lespedeza cuneata), and autumn-olive (Eleagnus umbellata) required the most management effort, followed by miscanthus (Miscanthus sinensis), wavyleaf basketgrass (Oplismenus hirtellus ), thistles (Carduus nutans, Cirsium arvense, Cirsium vulgare), callery pear (Pyrus calleryana), and beefsteak-plant (Perilla frutescens).
Besley, F.W. 1914. Map of Baltimore County and Baltimore City showing the forest areas by commercial types. Maryland Board of Forestry, Baltimore, Maryland.
Bowen, E.B. 1937. Soldier's Delight Hundred. Federation P.T.A. News.
Cadzow, D.A. 1936. Archaeological studies of the Susquehannock Indians of Pennsylvania. Safe Harbor Report No. 2, Archaeological Section. Vol. III. Pennsylvania Historical Commission, Harrisburg, Pennsylvania.
Futer, A.A. 1959. The Strickler Site. p. 136-147. In: Witthoft, J. and W.F. Kinsey, III (eds.). Susquehannock Miscellany. Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania.
Gustafson, D.J. and R.E. Latham. 2005. Is the serpentine aster, Symphyotrichum depauperatum (Fern.) Nesom, a valid species and actually endemic to eastern serpentine barrens? Biodiversity & Conservation 14:1445–1452.
Helicopter Applications, Inc. 2008. FLIR white-tailed deer survey, Soldiers Delight Natural Environment Area, Maryland Department of Natural Resources, Spring 2008. Gettysburg, Pennsylvania.
Hilgartner, W.B., M. Nejako, and R. Casey. 2009. A 200-year paleoecological record of Pinus virginiana, trace metals, sedimentation, and mining disturbance in a Maryland serpentine barren. Journal of the Torrey Botanical Society 136(2):257-271.
Hunter, H.A. 1959. The historic role of the Susquehannocks. p. 8-18. In: Witthoft, J. and W.F. Kinsey, III (eds.). Susquehannock Miscellany. Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania.
Johnsson, J. 2017. Maryland's Choate chromite mine, 1830 - 1920. Mining History Journal, p. 53-73.
Knox, R.G. 1984. Age structure of forests on Soldiers Delight, a Maryland serpentine area. Bulletin of the Torrey Botanical Club 111:498-501.
Knox, W.M. 1997. Historical changes in the abundance and distribution of deer in Virginia. In: McShea, W.J., H.B. Underwood, and J.H. Rappole (eds.). The Science of Overabundance: Deer Ecology and Population Management. Smithsonian Books, Smothsonian Institution, Washington and London.
Mansueti, R. 1950. Extinct and vanishing mammals of Maryland and District of Columbia. Maryland Naturalist 20:7-12.
Marye, W.B. 1920. The Old Indian Road, Part I: Various Indian roads. Maryland Historical Magazine 15(2):107-124; Part II: 208-229; Part III:345-395.
Marye, W.B. 1955a. The Great Maryland Barrens. Maryland Historical Magazine 50(1):11-23.
Marye, W.B. 1955b. The Great Maryland Barrens: II. Maryland Historical Magazine 50(2):120-142.
Marye, W.B. 1955c. The Great Maryland Barrens: III. Maryland Historical Magazine 50(3):234-253.
Maryland Department of Natural Resources. 2019. List of rare, threatened, and endangered plants of Maryland, March 2019. Natural Heritage Program, Wildlife and Heritage Service. Annapolis, Maryland. (https://www.dnr.maryland.gov/wildlife/Documents/rte_Plant_List.pdf)
Maryland Department of Natural Resources. 2020. Central Region Public Hunting Program 2020−2021. Wildlife and Heritage Service. Owings Mills, Maryland. (http://www.dnr.maryland.gov/wildlife/documents/CR-PublicHuntingProgram.pdf, 25 August 2020).
Maryland Geological Survey. 1929. Baltimore County. John Hopkins University Press, Baltimore, Maryland.
Maxwell, H. 1910. The use and abuse of forests by the Virginia Indians. William and Mary College Quarterly Historical Magazine 19:73-103.
Moore, C.T. 1972. Man and fire in the central North American grassland 1535-1890: a documentary historical geography. Ph.D. thesis, University of California, Los Angeles, California.
Paradiso, J.L. 1969. Mammals of Maryland. North American Fauna No. 66. Washington, D.C.
Pearre, N.C. and A.V. Heyl, Jr. 1960. Chromite and other mineral deposits in serpentine rocks of the Piedmont Upland, Maryland, Pennsylvania, and Delaware. U.S. Geological Survey Bulletin 1082-K:707-833.
Porter, F.W., III. 1975. From backcountry to county: the delayed settlement of western Maryland. Maryland Historical Magazine 70:329-349.
Porter, F.W., III. 1979. The Maryland frontier 1722-1732: Prelude to settlement in western Maryland. p. 90-107. In: Mitchell, R.D. and E.K. Muller (eds.) Geographical Perspectives on Maryland's Past. Occasional Paper No. 4, Geography Department, University of Maryland, College Park, Maryland.
Shreve, F., M.A. Chrysler, F.H. Blodgett, and F.W. Besley. 1910. The Plant Life of Maryland. The Johns Hopkins University Press, Baltimore, Maryland.
Singewald, J.T., Jr. 1928. Part II. Notes on feldspar, quartz, chrome, and manganese in Maryland. In: Maryland Geologic Survey Reports, Vol. 12, p. 91-194. Johns Hopkins University Press, Baltimore, Maryland.
Swanton, J.R. 1946. The Indians of the Southeastern United States. Smithsonian Institution Press, Washington, D.C.
Tyndall, R.W. 1992a. Historical considerations of conifer expansion in Maryland serpentine "barrens." Castanea 57:123-131.
Tyndall, R.W. 1992b. Herbaceous layer vegetation on Maryland serpentine. Castanea 57:264-272.
Tyndall, R.W. 1994. Conifer clearing and prescribed burning effects to herbaceous layer vegetation on a Maryland serpentine “barren.” Castanea 59:255–273.
Tyndall, R.W. and J.C. Hull. 1999. Vegetation, flora, and plant physiological ecology of serpentine barrens of eastern North America. p. 67-82. In: Anderson, R.C., J.S. Fralish, and J.M. Baskin (eds.) Savannas, Barrens, and Rock Outcrop Communities of North America. Cambridge University Press, Cambridge, United Kingdom.
Tyndall, R.W. 2020. Changes in herbaceous species variables after enhanced hunting effort for white-tailed deer in Soldiers Delight serpentine ecosystem in Maryland. Castanea 85(2):327-342.
USDA. 2014. Activities summary report, 2014 white-tailed deer management, Soldiers Delight Natural Environment Area. Animal and Plant Health Inspection Service (APHIS), Wildlife Services. Annapolis, Maryland.
Witthoft, J. 1959. Ancestry of the Susquehannocks. p. 19-60. In: Witthoft, J. and W.F. Kinsey, III (eds.). Susquehannock Miscellany. Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania.
Witthoft, J. and W.F. Kinsey, III. 1959. Susquehannock Miscellany. Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania.
Summary:The most important role of fire in savannas and grasslands is the regulation of microclimate; i.e., the climate near the ground in which plants and animals grow, reproduce, and evolve. Dead plant stems and leaves which accumulate between growing seasons have a number of adverse effects on growing conditions. By removing this dead material, fire improves microclimatic conditions for plant growth immediately. Light levels increase, the length of the growing season increases, excessive heat and drought conditions improve, along with other more favorable growing conditions. Even small fires can improve growing conditions substantially; dead biomass can be periodically burned on a small scale with great success. Microclimate is also important in the evolution of plants and animals. Over time, species adapt to conditions in which they are growing. In savannas, species have been/are evolving in microclimate maintained by fires occurring every 4 to 10 years. Changing this evolutionary fire regime changes growing conditions, and species which cannot adapt to a different microclimate decline and eventually disappear. Fire effects also differ considerably by season, so knowing the evolutionary fire history of an ecosystem is imperative before designing a prescribed burn program. Fires in the serpentine ecosystem of Maryland and adjacent Pennsylvania occurred in the “fall” (interpreted to be late fall into early winter) based on historical records and vegetation characteristics. Spring fire can be used as a temporary tool to manage invasive greenbrier. Smoke management will continue to limit prescribed burning at Soldiers Delight and other serpentine sites near urban and suburban areas. Although large volumes of smoke can be managed effectively, the high visibility of smoke columns from several miles away can lead to an overloading of calls at 911 centers. As a result, only small-scale burns are feasible in non-rural landscapes, limiting the total number of acres burned annually.
Fire and Microclimate. Perhaps the most important role of fire in savannas and grasslands is the regulation of microclimate; i.e., the climate near the ground in which plants and animals grow, reproduce, and evolve. In the 1960s, biologists started measuring weather parameters in burned and unburned prairies; their results are applicable to savannas with prairie-like vegetation such as serpentine barrens. In brief, growing conditions were found to be more favorable, and occurred earlier in the growing season, in burned prairie than in unburned (Hulbert 1969, Hulbert 1988, Knapp 1984, Knapp and Seastedt 1986, Old 1969, Peet et al. 1975). Unburned prairie has a large quantity of dead biomass, both standing and on the ground (litter). Standing dead stems and leaves intercept sunlight and, therefore, reduce photosynthetic light levels resulting in smaller plants and fewer flowers than in burned prairie. Litter on the ground inhibits the growth of shoots in the spring, by absorbing sunlight and minimizing soil temperature, thus shortening the growing season. Lower soil temperature also inhibits N and P production by microbes, nutrients which are released to the soil and used by plants. Burning off the litter layer substantially increases light levels for shoot growth, lengthens the growing season as shoot tips are exposed to light sooner, and increases soil temperature which results in faster plant growth and more N and P released from bacteria and fungi in the soil. In an Oklahoma tallgrass prairie, optimal soil temperature for growth of dominant prairie plants occurred more than a month sooner in burned prairie than in unburned prairie (Rice and Parenti 1978). In a Kansas tallgrass prairie (Knapp 1984), early spring soil temperature was as much as 30 F degrees warmer in burned versus unburned prairie, and 16 F degrees warmer six inches aboveground. Growth of unburned tallgrass was 55 % lower, and photosynthetic light 58 % lower, compared to burned. In an Illinois tallgrass prairie (Old 1969), a dormant-season spring burn caused a 2- to 3-fold increase in growth of dominant grasses and a 10-fold increase in flowering.
Standing dead plants also restrict air flow which allows excessive heat to build up around live plants, reducing their photosynthesis and growth (Knapp 1984). Leaf temperatures in burned prairie in Kansas were optimal for photosynthesis, but much greater than optimal in unburned due to less air flow. In addition, higher air temperatures cause greater water loss from leaves which limits photosynthesis. When prolonged, heat and water stress also increase mortality particularly during droughts.
Microclimate is also important in the evolution of plants and animals. Over time, species adapt to conditions in which they are growing. In savannas, species have been/are evolving in a microclimate maintained by fires occurring 2 to 3 times per decade. Changing this evolutionary fire regime changes growing conditions, and species which cannot adapt to a new microclimate decline and eventually disappear. Others may be able to persist, but do not exhibit optimum growth or reproduction. Fire-free serpentine barrens typically have a lackluster appearance of stunted grasses and scattered diminutive wildflowers. Removing fire from a fire-dependent ecosystem results in its collapse and eventual replacement, and the end to the evolution of species restricted to that ecosystem.
Fire Season. Fire effects differ considerably by season, so knowing the evolutionary fire history of an ecosystem is imperative before designing a prescribed burn program. Fortunately, Marye (1955) was able to determine that fires in the serpentine ecosystem of Maryland and adjacent Pennsylvania occurred in the “fall” (interpreted to be late fall into early winter). And that finding is corroborated by the characteristics of serpentine vegetation. The dominant plants are very robust, competitive, perennial grasses, especially little bluestem and Indian grass. Less competitive plants, such as annual wildflowers, survive between the clumps and colonies of these grasses. An important effect of periodic fall fires is keeping the spread of little bluestem, Indian grass, and other warm season grasses in check, leaving habitat for many other species. Periodic spring fires can have the opposite effect; i.e., growth and spread of warm season perennial grasses is favored, reducing available habitat for annuals and less competitive grasses. Marye learned that English colonists mimicked fall burning by Native Americans until learning that spring burning would better enhance grass forage for livestock.
Spring fire can be used as a temporary tool to restore serpentine vegetation if implemented correctly. The best example might be with spring burning of the non-indigenous greenbrier. Fire exclusion has led to its invasion, spread, and development of dense thickets. Being fire intolerant, it can be eradicated with spring burning if the soil is dry and, therefore, rhizomes are exposed to lethal fire temperatures. Rhizomes in wet soil are buffered from excessive heat of passing flames. In addition, burning greenbrier in wet soil in the spring can enhance growth through the release of nutrients. Unfortunately, spring rainfall usually precludes adequate drying of surface soil for spring burning-control of greenbrier. Winter soils are almost always saturated on mid-Atlantic serpentine.
Fire Size. Since a fundamental role of fire is on microclimate, even small fires can improve growing conditions substantially. Dead biomass can be periodically burned on a small scale with great success. However, small fires tend to be homogeneous and lack the diversity of conditions created by landscape fires. Large-scale fires result in a landscape of areas ranging from completely burned to unburned thereby leaving a diversity of habitat conditions, and these conditions can fluctuate with subsequent fires. In addition, landscape fires leave refugia for populations of plants and animals to quickly recolonize high-intensity burned areas. This is especially important for insect populations which have been persisting in ecosystems subject to fire exclusion.
Fire Frequency. Large landscapes under indigenous conditions do not burn completely every year, except during extreme circumstances such as megadroughts. For example, in a large serpentine oak savanna landscape, south- and west-facing slopes burn more frequently or completely than north- and east facing slopes, and slopes burn more frequently or thoroughly than ravine bottoms. Areas which burn frequently, every 1-3 years, usually support grasslands, or sparse savanna, since tree seedlings and saplings, including oaks, have insufficient time to become established. Areas which burn infrequently, every 8 years for example (Peterson and Reich 2001), support woodland or open forest.
Species diversity increases with fire frequency in oak savanna as well as closed forest communities. For example, a 40-year fire frequency experiment in Minnesota, involving study areas of open savanna to closed forest, found highest species diversity in the most frequently burned plant communities (Cavender-Bares and Reich 2012). Fire frequency in the study ranged from nearly annual to none. High frequency fires favor the development of grasses and forbs, while low frequency fires support woody plant dominance which suppresses grass and forb production (Peterson et al. 2007).
In a 32-year study of Minnesota oak savanna and woodland stands by Peterson and Reich (2001), low frequency burning, fewer than two fires per decade, produced woodland stands with dense sapling thickets. Frequent burning, three fires per decade, maintained oak savanna. The authors concluded that more frequent burning, more than four fires per decade, would result in grassland as too many oak seedlings and saplings would be killed to sustain populations.
Post oak savannas in the Missouri Ozarks were maintained by low intensity ground fires which occurred every 4.3 years on average, during the century before Native Americans emigrated from the region (1710 – 1810) (Guyette and Cutter 1991). During periods of severe drought, however, fires burned once every 11 years and were more severe.
Fire and Species Diversity. A universal outcome of fire research in many ecosystems, especially savannas and grasslands, has been an increase in species diversity with increasing fire frequency. At a long-term ecosystem science reserve in the Midwest, a 40-year prescribed burn experiment was conducted as part of an oak savanna restoration program (Cavender-Bares and Reich 2012). The total number of plant species, and their distribution across the savanna, were lowest in unburned areas and highest in areas which were burned almost annually. The same result has been documented repeatedly in prairie studies; annually burned prairies are richest in species diversity. Although this research has shown that plant species diversity peaks with very frequent fire, staying within the evolutionary fire regime of an ecosystem is important as other ecological components would suffer. For example, burning savannas annually would kill too many oak seedlings and saplings which are needed to maintain self-sustaining populations. In addition, excessive burning can have significant adverse effects on certain insect species.
Future of Fire. Smoke management will continue to limit prescribed burning at Soldiers Delight and other serpentine sites near urban and suburban development. Although large volumes of smoke can be elevated for dissipation at high altitudes with modern smoke management techniques, the high visibility of smoke columns from several miles away can lead to an overloading of calls at 911 centers. As a result, only small-scale burns are feasible in non-rural landscapes, limiting the total number of acres burned annually.
Cavender-Bares, J. and P.B. Reich. 2012. Shocks to the system: community assembly of the oak savanna in a 40-year fire frequency experiment. Ecology 93(8) Supplement, pp. S52-S69.
Guyette, R.P. and B.E. Cutter. 1991. Tree-ring analysis of fire history of a post oak savanna in the Missouri Ozarks. Natural Areas Journal 11(2):93-99.
Hulbert, L.C. 1969. Fire and litter effects in undisturbed bluestem prairie in Kansas. Ecology 50:874-877.
Hulbert, L.C. 1988. Causes of fire effects in tallgrass prairie. Ecology 69:46-58.
Knapp, A.K. 1984. Post-burn differences in solar radiation, leaf temperature, and water stress influencing production in a lowland prairie. American Journal of Botany 71:220-227.
Knapp, A.K. and T.R. Seastedt. 1986. Detritus accumulation limits productivity of tallgrass prairie. Bioscience 36:662-668.
Old, S.M. 1969. Microclimate, fire and plant production in an Illinois prairie. Ecological Monographs 39:355-384.
Peet, M., R. Anderson, and M.S. Adams. 1975. Effect of fire on big bluestem production. American Midland Naturalist 94:15-26.
Peterson, D.W. and P.B. Reich. 2001. Prescribed fire in oak savanna: fire frequency effects on stand structure and dynamics. Ecological Applications 11(3): 914-927.
Peterson, D.W., P.B. Reich, and K.J. Wrage. 2007. Plant functional group responses to fire frequency and tree canopy cover gradients in oak savannas and woodlands. Journal of Vegetation Science 18:3-12.
Rice, E.L. and R.L. Parenti. 1978. Causes of decreases in productivity in undisturbed tallgrass prairie. American Journal of Botany 65(10):1091-1097
Summary:Calcium is the most important nutritional element for the development of self-sustaining plant populations. It is critical to cell division and cell elongation; i.e., plant growth. And large healthy plants are needed to produce large numbers of seed (or other reproductive structures) to establish and maintain populations. Magnesium is only needed by plants in very small quantities, but it can contribute to very poor growth if in large amounts when calcium levels are low. Good plant growth occurs in soils with a Mg:Ca ratio of 1:1; gardeners lime their soils to achieve this ratio. At Soldiers Delight, the Mg:Ca ratio in topsoil (A horizon) is over 3:1, and it can be over 27:1 in subsoil (B horizon). This excessive Mg:Ca ratio is the main reason why serpentine areas around the globe develop distinctive vegetation, very different from surrounding vegetation. Heavy metal toxicity was once hypothesized to contribute to the distinctive composition of serpentine vegetation, because nickel (Ni), chromium (Cr), and cobalt (Co) occur naturally in serpentinite, and Ni and Cr are at high levels. However, none of these elements has been found to be toxic on serpentine in eastern North America and elsewhere because of unsuitable pH levels. That is, these elements do not occur in a toxic chemical state in the circumneutral pH range typical of serpentine. Low to very low levels of these elements have been found in plants at Soldiers Delight. Other essential elements which are naturally low in serpentinite, and therefore low in serpentine soils, are phosphorus and potassium, but their effects to plant growth are masked by excessive Mg:Ca. Nitrogen is not particularly low in serpentine soils.
Discussion: Calcium is the most important nutritional element for the development of self-sustaining plant populations as it is essential to cell division and cell elongation; i.e., plant growth. And large healthy plants are needed to produce large numbers of seed (or other reproductive structures) to establish and maintain populations. When parent ultramafic rock is converted to serpentine rock, calcium is lost because it cannot fit in the molecular structure of the new minerals (Alexander et al. 2007). As a result, this essential element for the growth of plants is in short supply in serpentine soils (Rabenhorst et al. 1982). In contrast, the high level of magnesium in ultramafic rock remains unchanged during the formation of serpentinite. And this combination of high magnesium and low calcium is deadly for most species of plants. Good growth occurs in soils with a Mg:Ca ratio of 1:1. Gardeners lime their soils to achieve this ratio. At Soldiers Delight, the Mg:Ca ratio in topsoil (A horizon) is over 3:1 in both silt loam and gravelly sandy loam soils (Tyndall 2012), and it can be over 27:1 in subsoil (B horizon) (Rabenhorst et al. 1982). This Mg:Ca factor is the main reason why serpentine areas around the globe develop distinctive vegetation. That is, the vegetation on serpentine is very different from surrounding vegetation, worldwide. In order for plants to grow and develop self-sustaining populations on serpentine, they must be able to compensate for this excessive ratio of high Mg to very low Ca. The physiological mechanisms for Mg:Ca tolerance can be found in Baker (1981).
Heavy metal toxicity was once hypothesized to contribute to the distinctive composition of serpentine vegetation, because nickel (Ni), chromium (Cr), and cobalt (Co) occur naturally in serpentinite, and Ni and Cr are at high levels. But Cr and Co fell out of scientific discussions when it was understood that they are not in biologically-available chemical forms in serpentine soil (Alexander et al. 2007). That is, they are not present in chemical forms which are useable by plants and, therefore, cannot become toxic to plants.
Ni continues to be part of the metal toxicity discussion as a number of plants on serpentine, including eastern North America, can have higher levels of Ni in their tissues than plants on adjacent non-serpentine (Briscoe et al. 2009, Milton and Purdy 1988, Pollard 2016). In addition, laboratory grown seedlings of a common plant on serpentine in Maryland and Pennsylvania (lyreleaf rockcress, Arabidopsis lyrata) have shown genetic signs of local adaptation to Ni (Veatch-Blohm et al. 2017). However, Ni phytotoxicity has not been verified for any plant species in eastern North America (Pollard 2016, Rajarakuna et al. 2009). Hochman (2001) measured low levels of Ni in the tissues of plants growing in serpentine soil at Soldiers Delight, with some plants having higher levels in nearby non-serpentine. Wood (1984) found no differences in available Ni between Soldiers Delight serpentine and non-serpentine soils. These results indicate that only a small proportion of total Ni in the soil is in a chemical form biologically available to plants. Ni experts have shown that biologically available Ni decreases exponentially with increasing pH (Kukier and Chaney 2004). Areas at Soldiers Delight with indigenous oak savanna or grassland vegetation do not have phytotoxic levels of Ni due to pH being above 5.8 (Kukier and Chaney 2004, Siebielec et al. 2007, Tyndall 2012). Areas which have been lost to pine-greenbrier woodland were hypothesized to have higher potential for Ni phytotoxicity as their soils become acidic, but a study by Cumming and Kelly (2007) found very low levels of Ni in leaves of herbaceous plants and lower total Ni levels in the pine woodland soil.
In the tropics, especially in Cuba and New Caledonia (Reeves 2003), over 300 species hyperaccumulate Ni to levels which would be toxic to most plants, including many plants which grow on serpentine. But no hyperaccumulator of Ni has been verified in eastern North America, and only two plant species in western North America (Rajakaruna et al. 2009).
Other essential elements which are naturally low in serpentinite, and therefore low in serpentine soils, are phosphorus and potassium (Alexander et al. 2007, Alexander 2009). But their effects to plant growth are masked by excessive Mg:Ca. Nitrogen is not particularly low in serpentine soils.
Alexander, E.B. 2009. Serpentine geoecology of the eastern and southeastern margins of North America. Northeastern Naturalist 16 (Special Issue 5) : 223-252.
Alexander, E.B., R.G. Coleman, T. Keeler-Wolf, and S.P. Harrison. 2007. Serpentine Geoecology of Western North America. Oxford University Press, Inc. New York, New York.
Baker, A.J.M. 1981. Accumulators and excluders: strategies in the response of plants to heavy metals. Journal of Plant Nutrition 3:643-654.
Briscoe L.R.E., T.B. Harris, W. Broussard, E. Dannenberg, F.C. Olday, and N. Rajakaruna. 2009. Bryophytes of adjacent serpentine and granite outcrops on the Deer Isles, Maine, U.S.A. Rhodora 111:1-20.
Cumming J.R. and C.N. Kelly. 2007. Pinus virginiana invasion influences soils and arbuscular mycorrhizae of a serpentine grassland. Journal of the Torrey Botanical Society 124:63-73.
Hochman, D.J. 2001. Pinus virginiana invasion and soil-plant relationships of Soldiers Delight Natural Environment Area, a serpentine site in Maryland. M.Sc. Thesis. University of Maryland, College Park, Maryland.
Kukier, U. and R.L. Chaney. 2004. In situ remediation of nickel phytotoxicity for different plant species. Journal of Plant Nutrition 27:465-496.
Milton, N.M. and T.L. Purdy. 1988. Response of selected plant species to nickel in western North Carolina. Castanea 53:207-214.
Pollard, A.J. 2016. Heavy metal tolerance and accumulation in plants of the southeastern United States. Castanea 81(4):257-269.
Rajakaruna, N., T.B. Harris, and E.B. Alexander. 2009. Serpentine geoecology of eastern North America: a review. Rhodora 111(945):21-108.
Reeves, R.D. 2003. Tropical hyperaccumulators of metals and their potential for phytoextraction. Plant and Soil 249:57-65.
Siebielec, G., R.L. Chaney, and U. Kukier. 2007. Liming to remediate Ni contaminated soils with diverse properties and a wide range of Ni concentration. Plant and Soil 299:117-130.
Tyndall, R.W. 2012. Soil differences between extant serpentine oak savanna and grassland in Soldiers Delight Natural Environment Area, Maryland. Castanea 77(3):224-230.
Veatch-Blohm, M.E., B.M. Roche, and E.E. Dahl. 2017. Serpentine populations of Arabidopsis lyrata ssp. lyrata show evidence for local adaptation in response to nickel exposure at germination and during juvenile growth. Environmental and Experimental Botany 138:1-9.
Wood, S.G. 1984. Mineral element composition of forest communities and soils at Soldiers Delight, Maryland. M.Sc. Thesis. Towson State University, Towson, Maryland
Summary: The serpentine ecosystem was classified as "barren" by English surveyors during the 1700s as barren was their term for landscapes without ("bare of") timber. Tree species were present, mainly oaks, but they were open-grown and not the tall, long-trunk trees found in the fireless forests of today. The individual serpentine areas and surrounding landscapes were primarily in savanna condition due to frequent and widespread low-intensity ground fires ignited by Native Americans mainly for fire-hunting white-tailed deer. (Savannas are dominated by grasses but, in contrast with grasslands such as prairies, savannas have an open canopy of trees with a canopy coverage of at least 5 - 10 %.) Fire in this fire-frequent oak savanna ecosystem ended about 1730, with the demise of Native American tribes, and the ecosystem changed quickly without fire. Savanna developed into forest unless subject to grazing by livestock and other disturbances. By 1800, timber-size trees were available for harvest, and before 1914, all timber stands at Soldiers Delight were logged. Beginning in the 1930s, in the absence of fire and grazing, the non-indigenous Virginia pine and greenbrier rapidly destroyed most native vegetation. Due to fire exclusion, grazing, cutting, pine-greenbrier invasion, and excessive deer browsing, the extent of the current oak savanna at Soldiers Delight is smaller than before settlement (ca. 1750). Today, oaks continue to expand and grow in restoration areas, and ground vegetation is dominated by little bluestem (Schizachyrium scoparium) and other grasses, with a variety of wildflowers, sedges, and rushes.
Serpentine vegetation has been, and continues to be, a product of climate change. Maryland climate toward the end of the Pleistocene, 20,000 to 17,000 years before the present (BP), was very cold and dry as the Laurentide Ice Sheet was nearby in Pennsylvania. The main forest type in the piedmont was an open conifer forest dominated by spruce, fir, and white pine, named the cool temperate conifer forest. Between 16,000 and 8,000 BP, plant distributions and abundances changed rapidly across North America (Williams et al. 2004). Cold-tolerant trees migrated northward and were replaced by deciduous ones, especially oaks. During the warm and dry periods of the mid-Holocene Hypsithermal Interval, 8,000 BP to 4,000 BP, the temperate deciduous forest, dominated by oaks, was fully developed in the mid-Atlantic region. Like the temperate conifer forest, this deciduous forest had an open canopy and, therefore, more herbaceous cover than deciduous forest of today. The serpentine oak savanna ecosystem could be thousands of years old, based on knowledge about the development and expansion of prairie and savanna in the Midwest. Warm and dry periods of the Hypsithermal Interval resulted in major expansion and migration of prairie and savanna, including an eastward migration from the Midwest augmenting the regional flora already in place. Those plants which tolerated magnesium-rich calcium-poor soils would have been able to colonize serpentine, notably blackjack oak, post oak, little bluestem, big bluestem, and Indian grass. As in the Midwest, Amerindian landscape fires would have aided the migration and expansion of savanna plant populations especially during periodic wet cycles. Their maintenance would also have been assisted by periodic megadroughts after the Hypsithermal, the most recent ending just before English settlement in the 1700s.
Vegetation 1700 - Present
In the 1700s, the indigenous serpentine ecosystem was classified as "barren" by English surveyors (details in Landscape History section). Barren was the term used by the English for landscapes without ("bare of") timber. Tree species were present, mainly oaks, but they were not the tall, long-trunk trees found in the fireless forests of today. The individual serpentine areas and surrounding landscapes were primarily in savanna condition due to frequent and widespread low-intensity ground fires ignited by Native Americans. Savannas are dominated by grasses but, in contrast with grasslands such as prairies, savannas have an open canopy of trees with a canopy coverage of at least 5 - 10 % (Anderson et al. 1999). Grassland and open forest were present, but minor in comparison to the coverage of savanna. The main savanna tree species at Soldiers Delight were blackjack oak (Quercus marilandica), post oak (Q. stellata), and sassafras (Sassafras albidum) (Weakley 2015 nomenclature).
Vegetation type at Soldiers Delight and other serpentine areas is partially controlled by soil type. The shallowest soils are only a few inches deep, well drained, and drought-prone (details in Soils and Vegetation section). Where this soil is only 2 - 3 inches deep, only grassland vegetation was able to persist as woody plants need deeper soils and more moisture for their rooting systems. On deeper soils, the English reported oak and sassafras "bushes", and "saplings" on even deeper soils. And the deepest soils provided both adequate rooting volume and moisture for the development of large open-grown trees such as "gnarled stunted oaks".
Fire in this fire-frequent oak savanna ecosystem ended about 1730, with the demise of Native American tribes, and the ecosystem changed quickly without fire. While the shallowest soils remained grassland, savanna developed into forest (Marye 1955). By 1800, timber-size trees were available for harvest, and before 1914, all forests in Soldiers Delight were logged (Besley 1914).
The current grassland community at Soldiers Delight is much larger than before settlement, due to grazing, cutting, and other activities after settlement (ca. 1750), and excessive deer browsing in recent decades (see Landscape History section). With effective stewardship, oak and sassafras recolonization will convert much of the current grassland into savanna. This is an important process since savannas are even more diverse than grasslands primarily because of the oaks and the numerous insects which feed upon them. In addition, oak trees create shaded habitat for forbs which cannot tolerate or compete well in full sunlight conditions (Pavlovic et al. 2006).
Grassland vegetation today is primarily dominated by little bluestem (Schizachyrium scoparium), with Indian grass (Sorghastrum nutans) more dominant in some areas. These grasses are the same plants which are dominant in prairies and savannas of the Midwest. In spring, seasonally characteristic plants in the grasslands at Soldiers Delight are parasol sedge (Carex umbellata), round-fruited witchgrass (Dichanthelium sphaerocarpon), serpentine chickweed (probably a variety or subspecies of Cerastium velutinum), and lyreleaf rockcress (Arabidopsis lyrata). In summer, conspicuous plants are papillose nutrush (Scleria pauciflora), Appalachian ragwort (Packera anonyma), starved witchgrass (Dichanthelium depauperatum), and grey goldenrod (Solidago nemoralis). In fall, serpentine aster (Symphyotrichum depauperatum), Indian grass (Sorghastrum nutans), arrowfeather (Aristida purpurascens), fork-tip three-awn grass (Aristida dichotoma), and glade knotweed (Polygonum tenue) are characteristic. The relative abundance of these, and less common plants, can be found in Tyndall (1992), and detailed botanical descriptions can be found in Weakley (2015).
The flora of serpentine oak savanna unfortunately was not documented before Soldiers Delight was invaded by non-indigenous and highly invasive Virginia pine and greenbrier, beginning in the 1930s (see Landscape History section). However, a remnant of oak savanna was sampled after it was uncovered in 1991 by the removal of a dense pine stand (Tyndall 2005). Before pine clearing, plant cover was sparse under the pines, but it had tripled in just 12 years after clearing, with little bluestem becoming the dominant plant. In addition, the oak savanna flora was somewhat different from adjacent grassland, adding additional diversity to the ecosystem, and it was still changing at the end of the study.
Vegetation Before 1700
Serpentine vegetation has been, and continues to be, a product of climate change. Rrecords do not exist specifically for Soldiers Delight, but excellent information is available about pre-historical climates and vegetation types for the mid-Atlantic region, Chesapeake Bay area, and the Midwest. That information provides insight into the development of serpentine oak savanna in the piedmont of Maryland, with the understanding that much more research is needed to fill in the blanks of current knowledge.
Maryland climate toward the end of the Pleistocene, 20,000 to 17,000 years before the present (BP), was very cold and dry as the Laurentide Ice Sheet was nearby in Pennsylvania. The main forest type in the piedmont was an open conifer forest dominated by spruce, fir, and white pine, named the cool temperate conifer forest (Delcourt and Delcourt 1984, Williams et al. 2000). This conifer forest was not a closed canopy forest because of dry conditions and low CO2 levels. Prairie forbs were present, but low in number (Williams et al. 2004). Oaks were essentially limited to the southeastern United States during this glacial maximum.
Between 16,000 and 8,000 BP, plant distributions and abundances changed rapidly across North America (Williams et al. 2004). Plant ranges expanded in both south-north and west-east directions, and centers of abundance also changed during this period. Cold-tolerant trees migrated northward and were replaced by deciduous ones, especially oaks. Oaks migrated rapidly between 20,000 and 14,000 BP, and may have become most abundant in Maryland between 10,000 and 9,000 BP (Williams et al. 2000). Although common today, oaks were more abundant in the past.
During warm and dry periods of the mid-Holocene Hypsithermal Interval, 8,000 BP to 4,000 BP, the temperate deciduous forest, dominated by oaks, was fully developed in the mid-Atlantic (Williams et al. 2000). Like the temperate conifer forest, this deciduous forest had an open canopy and, therefore, more herbaceous cover than deciduous forest of today (Williams et al. 2004). In the southern portion of the Chesapeake Bay area, oaks became mixed with southern pines during 5500 BP to 4800 BP, as pines migrated northward in response to warmer and wetter winters (Willard et al. 2005), adding a mixed deciduous-conifer forest to the region. Both of these forest types remained prominent until widespread deforestation by the English beginning in the 1700s AD in Maryland.
Between 1450 AD and 2000 AD, the Chesapeake Bay region experienced 14 wet-dry climatic cycles, and some of the dry cycles were extreme (Cronin et al. 2000). Megadroughts, lasting for more than a decade, occurred during dry cycles between 1450 AD and 1650 AD, and some were more severe than the Dust Bowl of the 1930s. These megadroughts also occurred during the coolest part (1450 - 1600 AD) of the Little Ice Age (1300 - 1900 AD) (Cronin et al. 2003). The end of the 1450 - 1650 AD megadroughts nears the beginning of historical descriptions of the serpentine barrens by English surveyors and colonists (Marye 1955). Precipitation increased in the late 1600s, and wet conditions continued until about 1930.
A similar but more complex pattern of climate and vegetation change occurred in the Midwest, where spruce and jack pine were dominant at the peak of glaciation about 18,000 BP (Anderson 2006). As climate warmed and dried during the early Holocene, 10,000 - 8,000 BP, conifers were replaced by grassland, oak savanna, and open oak-hickory forest. Between 8,000 and 5,000-3500 BP, depending on the location, the Hypsithermal Interval had cycles of even warmer and drier conditions, and the landscape was subject to frequent fires ignited by Native Americans (Anderson 2006, Moore 1972, Nelson et al. 2006). As a result, oak-hickory forest was converted to oak savanna and prairie, except in the fire shadows of steep slopes, lakes, and rivers (Anderson 2006, Thomas-Van Gundy 2020). During wet cycles of the Hypsithermal and the wetter climate following it, frequent landscape fires prevented re-expansion of the oak-hickory forest and maintained prairie and savanna. Prairie expanded farther eastward, after a wet pause, ending about 6200 BP, reaching into central Ohio (Mack and Boerner 2004). Dominant grasses included two which are dominant at Soldiers Delight, Indian grass and little bluestem. Although wet cycles occurred after prairie expansion, frequent landscape fires maintained prairies as well as oak savannas, woodlands, and open forests until European settlement in the 1800s (Nelson et al. 2006, Thomas-Van Gundy 2020).
The serpentine oak savanna ecosystem could be thousands of years old, based on knowledge about the development and expansion of prairie and savanna in the Midwest. The warm and dry periods of the mid-Holocene Hypsithermal Interval, 8,000 - 4,000 BP, allowed prairie and savanna plants to migrate eastward, augmenting the regional flora already in place. Those that tolerated magnesium-rich calcium-poor soils were able to colonize serpentine, notably blackjack oak, post oak, little bluestem, big bluestem, and Indian grass. As in the Midwest, Amerindian landscape fires would have aided the migration and maintenance of savanna plant populations during periodic wet cycles. Their maintenance would also have been assured by periodic megadroughts after the Hypsithermal, the most recent ending just before English settlement in the 1700s.
Anderson, R.C. 2006. Evolution and origin of the Central Grassland of North America: climate, fire, and mammalian grazers. Journal of the Torrey Botanical Society 133(4):626-647.
Besley, F.W. 1914. Map of Baltimore County and Baltimore City showing the forest areas by commercial types. Maryland Board of Forestry, Baltimore, Maryland.
Cronin, T., D. Willard, A. Karlsen, S. Ishman, S. Verardo, J. McGeehin, R. Kerhin, C. Holmes, S. Colman, and A. Zimmerman. 2000. Climatic variability in the eastern United States over the past millenium from Chesapeake Bay sediments. Geology 28(1):3-6.
Cronin, T.M., G.S. Dwyer, T. Kamiya, S. Schwede, and D.A. Willard. 2003. Medievel Warm Period, Little Ice Age, and 20th century temperature variability from Chesapeake Bay. Global and Planetary Change 36:17-29.
Delcourt, P.A. and H.R. Delcourt. 1984. Late Quarternary paleoclimates and biotic responses in eastern North America and the western North Atlantic Ocean. Paleogeography, paleoclimatology, paleoecology 48:263-284.
Mack, J.J. and R.E.J. Boerner. 2004. At the tip of the Prairie Peninsula: vegetation of Daughmer Savanna, Crawford County, Ohio. Castanea 69(4):309-323.
Marye, W.B. 1955. The Great Maryland Barrens. Maryland Historical Magazine 50(1):11-23; 50(2):120-142; 50(3):234-253.
Moore, C.T. 1972. Man and fire in the central North American grassland 1535-1890: a documentary historical geography. Ph.D. thesis, University of California, Los Angeles, California.
Nelson, D.M., F.S. Hu, E.C. Grimm, B.B. Curry, and J.E. Slate. 2006. The influence of aridity and fire on Holocene prairie communities in the eastern Prairie Peninsula. Ecology 87(10):2523-2536.
Pavlovic, N., R. Grundel, and W. Sluis. 2006. Groundlayer vegetation gradients across oak woodland canopy gaps. Journal of the Torrey Botanical Society 133(2):225-239.
Thomas-Van Gundy, M.A., G.J. Nowacki, R.C. Anderson, M.L. Bowles, L. Marlin, R.B. Brugam, N.B. Pavlovic, S.J. Halsey, and J. McBride. 2020. Visualizing the ecological importance of pre-Euro-American settlement fire across three Midwestern landscapes. The American Midland Naturalist 183(1):1-23.
Tyndall, R.W. 1992. Herbaceous layer vegetation on Maryland serpentine. Castanea 57:264-272.
Tyndall, R.W. 2005. Twelve years of herbaceous vegetation change in oak savanna habitat on a Maryland serpentine barren after Virginia pine removal. Castanea 70(4):287-297.
Weakley, A.S. 2015. Flora of the southern and mid-Atlantic states, working draft of 21 May, 2015. University of North Carolina Herbarium, Chapel Hill, North Carolina.
Willard, D.A., C.E. Bernhardt, D.A. Korejwo, and S.R. Meyers. 2005. Impact of millenial-scale Holocene climate variability on eastern North American terrestrial ecosystems: pollen-based climate reconstruction. Global and Planetary Change 47:17-35.
Williams, J.W., T. Webb III, P.H. Richard, and P. Newby. 2000. Late-Quaternary biomes of Canada and the eastern United States. Journal of Biogeography 27:585-607.
Williams, J.W., B.N. Shuman, T. Webb III, P.J. Bartlein, and P.L. Leduc. 2004. Late-Quaternary vegetation dynamics in North America: scaling from taxa to biomes. Ecological Monographs 74(2):309-334
Summary: Vegetation types at Soldiers Delight occur on different types of serpentine soil. The transition from one soil type to another is very narrow resulting in distinct sharp boundaries between vegetation types. Grassland vegetation is on gravelly sandy loam averaging only 3 inches deep. Oak savanna occurs on the deeper silt loam, averaging about 3 feet deep in one study area. Oak forest is on the deepest silt loam, measuring over 5.5 feet deep in one study. These soil types are easily distinguished by the gravelly surface and black color of sandy loam, and the reddish-brown color of the gravel-free silt loam. Sandy loam is about two-thirds sand, while the silt loam is about two-thirds silt. Gravelly sandy loam soil is too shallow and drought-prone to support trees. In contrast, the deeper silt loam soils have adequate rooting space and moisture for trees. Although very different in physical characteristics, soil chemistry is similar for these soil types. In particular, Mg:Ca ratios are similar and pH is high for both (5.8 - 6.7). After recently discovering that the gravelly sandy loam is a new soil series, scientists of the U. S. Department of Agriculture, National Resources Conservation Service excavated two pedons (surface to bedrock soil pits) for detailed analysis and description, along with the silt loam soil type. In addition, they produced a block diagram of serpentine geology catena; i.e., the sequence of distinct soils which vary with slope from the same bedrock.
Discussion: Plant communities at Soldiers Delight occur on different types of serpentine soil. The transition from one soil type to another is very narrow resulting in sharp boundaries between vegetation types. Grassland vegetation is on a gravelly sandy loam averaging only 3 inches deep (Tyndall 2012). Oak savanna occurs on the deeper silt loam, averaging about 3 feet deep in one study area. Oak forest silt loam soils are the deepest, measuring over 5.5 feet deep in one study (Rabenhorst et al. 1982). These soil types are easily distinguished by the gravelly surface and black color of sandy loam, and the reddish-brown color of silt loam. Sandy loam is about two-thirds sand, while the silt loam is about two-thirds silt.
These differences in soil texture and depth result in very different growing conditions and, therefore, different vegetation. Tree and shrub establishment is severely limited on the gravelly sandy loam due to very shallow soil depth and high rock fragment content. Shallow stony soils have limited rooting volume, nutrient capacity, and available water, plus extremes in maximum temperature and drought (Lutz and Chandler 1946, Poesen and Lavee 1994, Wesemael et al. 2000). In other words, the gravelly sandy loam soil produces a hostile environment for woody plants. In contrast, the deeper silt loam soils, have adequate rooting space and moisture for tree development.
Although very different in physical characteristics, soil chemistry is similar for these soil types (Tyndall 2012). In particular, Mg:Ca ratios are similar and pH is high for both (5.8 - 6.7). (Soil pH is high because the Mg-rich bedrock results in the dominance of Mg2+ on the exchange complex (Rabenhorst et al. 1982).)
Although woody plant establishment and growth are severely inhibited on gravelly sandy loam, pre-settlement survey descriptions (Marye 1955) suggest that oak shrubs were present before settlement and, therefore, are expected to slowly reestablish where soils are not too shallow. Exemplary descriptions of Soldiers Delight include, “… one half barrens the rest small saplings and bushes soil thin…”, “… 200 acres of sapling Land 300 acres of Bare Barrens the Rest small bushes…”; and “Soil of Both Bushy and barren Land is very thin and both Hilley, and Stoney, the soil of the Sapling Land is Middling…”. The descriptor “barren land” may have referred to gravelly sandy loam too shallow for woody plant development. “Bushy … land” may have been deeper gravelly sandy loam as well as patches of shallow silt loam within large units of gravelly sandy loam. “Sapling Land” and “Middling” soil may have referred to small trees on deeper silt loam. Definitions of these descriptors were not provided in historical publications.
After the gravelly sandy loam soil was first analyzed (Tyndall 2012), soil scientists of the U. S. Department of Agriculture, National Resources Conservation Service, David Verdone and Aaron Friend, determined that it was a new soil series which had never been classified. That finding led to official USDA descriptions of the two main soil types, with the excavation of two pedons (surface to bedrock soil pits). The following is a summary of their findings (Soil Survey Staff, National Resources Conservation Service, United States Department of Agriculture 2013), beginning with a block diagram of serpentine geology catena; i.e., a sequence of distinct soils which vary with slope from the same bedrock. The gravelly sandy loam described in Tyndall (2012) is in a still nameless soil series, referred to as "Fine-silty, magnesic, mesic Lithic Hapludoll" in the Taxonomic Class, Loamy, Mesic Lithic Hapludolls. The silt loam in Tyndall (2012) is in the Chrome soil series, in the Taxonomic Class, Fine, Mixed, Active, Mesic Typic Hapludalfs. The Travilah soil series (associated with intermittent streams) was not sampled in their study.
Serpentine Geology Catena
Fine-silty, magnesic, mesic Lithic Hapludoll ["gravelly sandy loam" in Tyndall (2012)]
This unique well drained soil has yet to be established, but can be found in the serpentine geology on crests and nose slopes in the northern piedmont region. The profile is less than 25 cm to an aquitard of serpentine bedrock, high in OM throughout, extremely low Ca:Mg ratios (e.g. 0.25:1) and high in Ni and Cr. This thin profile has a very low AWC which drastically limits vegetative growth in the dry summer months. These edaphic properties create a unique grassland forb ecosystem that has drawn international attention.
Chrome ["silt loam" in Tyndall (2012)]
This well drained soil is found on broad convex interfluves and linear side slopes. It has many of the same chemical properties as the Lithic Hapludoll, but is 50 to 100 cm to serpentine bedrock. The silty clay loam textures found at 25 cm creates an aquitard, which results in a low Ksat and moderate AWC. This soil becomes very dry during summer months. The vegetative community is still very unique with slightly more tree species.
This somewhat poorly drained soil can be found on base slopes, head slopes and concave areas within interfluves of serpentine geology. These are discharge areas for seeps, and accumulate overland flow during precipitation events. This soil has silty clay loam textures at 25 cm that act as an aquitard with a representative soil depth between 50 and 100 cm to bedrock. The profile has a moderate AWC that is sustained into the summer months by ground water discharge from higher elevations. This prolonged ground water table allows for more water demanding vegetation such as trees to persist.
Oe--0 to 2 centimeters; loose; clear smooth boundary.
A1--2 to 7 centimeters; black (10YR 2/1) very gravelly loam, very dark gray (10YR 3/1), dry; 22 percent clay; moderate fine granular structure; very friable, nonsticky, nonplastic; many fine roots throughout and common medium roots throughout and many very fine roots throughout; 45 percent nonflat subangular strongly cemented 2- to 75-millimeter serpentinite fragments; neutral, pH 6.6, pH indicator solutions; clear smooth boundary.
A2--7 to 15 centimeters; black (10YR 2/1) gravelly loam, dark gray (10YR 4/1), dry; 22 percent clay; moderate fine subangular blocky structure parting to weak fine granular structure; very friable, nonsticky, nonplastic; many fine roots throughout and common medium roots throughout and many very fine roots throughout; 29 percent nonflat subangular strongly cemented 2- to 75-millimeter serpentinite fragments; neutral, pH 6.8, pH indicator solutions; clear smooth boundary.
BC--15 to 21 centimeters; dark yellowish brown (10YR 3/6) gravelly clay loam; 28 percent clay; weak
fine subangular blocky structure; very friable, nonsticky, slightly plastic; many fine roots throughout and common medium roots throughout and many very fine roots throughout; 15 percent nonflat subangular strongly cemented 2- to 75-millimeter serpentinite fragments; neutral, pH 7.0, pH indicator solutions; abrupt smooth boundary.
R--21 to 46 centimeters; few very fine roots in cracks