--with computer map graphics by Chuck Sutherland


A "comb grave" is a burial that features a grave cover made, normally, of two rectangular slabs of stone leaned together to form a gable roof over the grave (Plate 1). The term "comb" signifies "the crest or ridge of a roof" (OED, 1971)1. Combs have also been likened to pup tents and persons unfamiliar with the term "comb grave" commonly call them "tent graves". However, comb grave is thought to be the traditional name for this vernacular type of grave cover, and its essential form is that of a gable roof set directly on the grave, with no supporting walls. The above-ground space beneath the "roof" is normally empty, not filled.

Plate 1: Typical comb graves, Mt. Gilead Cem., Cassville quad, White Co.

Comb graves have been further described as “triangular and prismatic” in form (Cantrell, 1981, Montell, 1993). Indeed, with each end of the typical comb closed off by a stone slab cut to fit the triangular opening under the side slabs, the basic shape of a comb is that of a simple triangular prism.

Most comb graves feature a headstone that is separate from the comb structure, and some feature both head- and footstones. These head- and footstones typically are of the same local stone as the comb; this is especially true of the older comb graves. However, it is not rare for the headstone to be of marble or other non-local “store-bought” stone. Some combs, especially in the older comb cemeteries, are “headless” (Plate 2). Many of these headless combs have no inscriptions, making it difficult to know who is covered by the comb or when the burial occurred. However, some headless combs have inscriptions on one side slab (Plate 3a) and a few have inscriptions on one of the gable-end stones (Plate 3b).

Plate 2: Old Jericho Cem., Doyle quad, White Co.

Plate 3a: Side-scribed headless comb, Anderson Cem., Doyle quad, White Co.

Plate 3b: Gable-scribed headless comb, Oakley Cem., Okalona quad, Overton Co.

Traditionally, and most commonly, the long side slabs and the end pieces are made of sandstone. However, where suitable sandstone is not readily available, other materials may be used: limestone, sheet metal, cement or concrete, and, in at least one instance, shale. Two marble combs are known, the marble no doubt chosen not so much out of necessity as a desire to make the comb more elegant than normal.

In Tennessee, over 3600 extant combs are found in over 500 cemeteries scattered along a NNE-SSW-trending band paralleling the western front of the Cumberland Plateau (Figure 1 shows the distribution. N.B.: Since this figure was prepared for publication, a number of additional comb cemeteries have been located; for a continuously updated map showing all the known comb cemeteries, see Chuck Sutherland's interactive comb cemetery map found on Comb graves are most common in older graveyards lying on the Eastern Highland Rim, below the Plateau. However, many are also found on the Hartselle Bench halfway up the Plateau escarpment, and a lesser number on top of the Plateau along its western side. The southernmost examples in the main Tennessee comb range are 13 combs in Perkins Cemetery just northeast of Winchester, in Franklin Co. Roper Cem., in Lincoln Co., well to the west of the contiguous Tennessee comb range (not shown in Figure 1), boasts the south-westernmost combs known in the state. The northernmost extant example is a single comb in Rector Cem. in Pickett Co.2 However, one additional comb grave formerly existed in Taylor Grove Cem. 8.6 miles north of the Tennessee-Kentucky state line, and this cemetery should still be considered part of the geographic range of the Tennessee comb graves.

Although at least 72 comb cemeteries are associated with churches, the majority are local graveyards or family plots not associated with church buildings.

The tradition of erecting combs over graves appears to have commenced around 1815 -1820. The custom was strong throughout the remainder of 19th century and during the first half of the 20th century, but by the 1950s-60s erection of combs was uncommon. Even so, a comb was erected in 2012 (Finch, 2013).

Comb graves are known to be present in eight other Southern states (Alabama, Arkansas, Kentucky, West Virginia, Texas, Mississippi, Louisiana and North Carolina). A few comb graves have also been found in eastern Oklahoma and southernmost Missouri, areas that are culturally part of the Upland South (Jordan-Bychkov, 2003). However, Tennessee has more known comb cemeteries and comb graves than all the other states combined. Additionally, the oldest known combs are found in Tennessee graveyards. Probably, the comb grave custom is indigenous to Tennessee.

A variety of ideas have been suggested to account for the “why” of combs, i.e., the purpose served by placing a comb over a grave. Protecting the grave seems to have been a motive, but there is no definitive single reason for combs. It is likely that different reasons motivated different people to erect combs, but that ultimately the comb became a highly popular style, indeed the dominant grave style in numerous small graveyards within the comb range. And style alone was probably sufficient reason for many. Of the other various well-known types of grave covers (ledger stones, box graves, stone table markers, gravehouses, coffin graves, and cairn graves) within the main comb range, none were used in numbers comparable to the comb graves.

Methodology and comb grave data sources:

The present study is an update and extension of earlier work (Finch, 1982, 2004) which began as a simple attempt to learn the geographic range of the comb graves, using USGS and TVA 7.5 minute topographic quadrangle maps (“topos” or “quads”) as guides to cemetery locations. In the somewhat naïve belief that the topo maps showed all or nearly all existing graveyards, each indicated cemetery was visited to determine the presence or absence of comb graves. The happenstance discovery of numerous graveyards not shown on the quads, plus news of additional cemeteries from helpful informants quickly showed that numerous other cemeteries were out there. In time, a variety of resources were found to be useful in locating cemeteries not indicated on the topo maps.

Many county road maps contain cemetery location information. For some Tennessee counties, maps have been published for the exclusive purpose of showing cemetery locations. Comprehensive cemetery books have been written for a number of counties, e.g., Overton, White and Van Buren counties. The “Fred Clark Book of Cemeteries of White County” proved exceptionally useful inasmuch as it denotes which cemeteries contain comb graves. Online cemetery maps are available for some counties, but the locations given are not always correct. Two online sources that have proved especially useful are Find-A-Grave ( ) and the Tennessee Property Viewer ( ). Find-A-Grave can be mined for leads to cemeteries, but must be used with caution as much information on this website is incorrect, especially cemetery location descriptions. The Property Viewer can be used to search for additional cemeteries because many graveyards are deeded apart from the adjacent or surrounding parcels. Such cemeteries show up as unusually small parcels, often completely surrounded by a larger tract; when a “postage stamp” parcel is spotted on the property viewer map, the “identify” function can be used to determine whether or not it is a cemetery. Somewhat surprisingly (or perhaps not?) this state government website also is troubled with some incorrect property lines and even completely mis-located parcels.

With the numerous and varied sources available, it is clear that the quest --already over four decades old-- to locate all Tennessee comb cemeteries is a task that can never be truly completed before the writer himself is in need of a comb. But the search goes on!

In sum, for this survey, all cemeteries --excepting those that have disappeared3-- on 129 quads have been visited and any combs noted. Out of a total of over 3100 cemeteries, exactly 516 comb graveyards have been identified in the main comb range, and 3662 extant combs counted and photodocumented. The 516 comb cemeteries include 471 graveyards that currently contain combs and 45 ex-comb graveyards known to have formerly had combs. Ex-comb cemeteries were mostly identified in two ways: 1) they had combs when first visited, but the combs were gone on a subsequent visit; 2) they contained clear-cut remnants of comb graves, e.g., a matched pair of triangular gable stones indicating the former presence of a comb. Historic photographs showing combs in graveyards that no longer feature comb graves facilitated the identification of three former comb cemeteries. Graves with only a single triangular stone, or a single rectangular slab laid flat, or sets of head- and footstones of a style commonly associated with comb graves were noted as “probable” or “possible” ex-combs, depending on the evidence, but were not counted as definite ex-combs. Cemeteries with only possible or probable ex-combs were not included in the total number of comb cemeteries, even though at least some of them surely once were comb cemeteries. Appendices A-1 and A-2 are inventories of all the comb cemeteries, ex-comb cemeteries, and combs surveyed.

The locations of the comb cemeteries listed in Appendices A-1 and A-2 are given in latitude and longitude, recorded in the field with a GPS unit (set to the NAD-27 datum, the datum of the published maps) or measured from the published quads (using a TopoTool Coordinate Ruler by Neff Scientific). Where the local name for a cemetery was not known, a name was given for the purposes of this survey based on the most prominent surname or surnames in the cemetery.

To ensure that the full contiguous range of comb graveyards was defined, the survey was extended into quads beyond the core geographic area of the combs until a border of “comb-free” quads was established around the entire contiguous comb range.

Having commenced as a purely geographic endeavor, this study eventually morphed to incorporate more aspects of the comb grave tradition: its temporal range, physical materials used, church associations, origin and reasons for the custom. If a number of comb cemeteries have been missed by this survey, as they most certainly have been, some of the statistics presented in this paper will be changed when these overlooked cemeteries are ultimately discovered and reported. Nonetheless, for the present, we can be relatively confident that the geographic range has been defined and documented, and insights gained regarding other aspects of the comb tradition.

Comb grave materials and construction:

Combs are normally constructed of stone, the preferred material being relatively thin slabs of sandstone taken from the stratigraphic unit known as the Hartselle Formation (note outcrop trace on Figure 1). In places the Hartselle contains thin, planar beds of sandstone that lend themselves to the quarrying of broad, flat slabs, ideal for comb side slabs. Thin-bedded sandstone was undoubtedly prized because of the weight of the stone: typical sized side slabs, even when just 1½ inches thick, require four men to carry safely. Planar-bedded sandstone was preferred as it required less dressing, coming out of the ground with a more-or-less finished surface. The same beds from which the side slabs were quarried could also be cut into smaller pieces and shaped to fill the triangular gaps at each end of the main comb structure (Plate 4), and for that matter, head- and footstones. Two quarries from which Hartselle sandstone gravestones were “pulled” are the Vaughn quarry (Plate 5a) above the community of Allred (Crawford quad), and the Ogletree quarry (Plate 5b) not far from the Holly Springs community (Hilham quad), both in Overton Co. But gravestones are also known to have been quarried at other sites, such as the quarry worked by Dan Elrod in the upper Calfkiller Valley (Cantrell, 1981) and the quarry on Gum Spring Mtn., White Co., worked by Thomas Keathley. There likely were numerous local quarries from which gravestones were extracted. To a certain degree, local sources were necessary, as the weight of the stones and the poor condition of 19th- and early 20th-century roads made long distance transport difficult and generally impractical.

Plate 4: Triangular gable stone supporting one end of a comb in Bethlehem Cem., Bald Knob quad, White Co.

Plate 5a: Old Vaughn quarry in Hartselle sandstone, above Allred, Crawford quad, Overton Co.

Plate 5b: Ogletree quarry in Hartselle Formation, Hilham quad, Overton Co. Note unused headstones leaning against quarry wall.

As mentioned above, the triangular opening left at each end of a comb was normally plugged with a sandstone slab cut to fill the space. These gable stones improve the visual effect of the comb and, by sealing the comb ends, give the grave more complete protection. But more importantly, the gable stones provide necessary structural support for the heavy side slabs. Without these stones (or some alternative form of support) the side slabs would be very subject to gradual spreading and eventual collapse. These gable stones were sometimes actually triangular in shape, but such stones might in time lean in or out or collapse and cease to function as support stones (Plate 6). Some gable stones only appear to be triangular, having in actuality a buried extension that improves their stability. These features are rarely visible, but cautious pulling and pushing on some exposed gable stones indicates that they have subsurface extensions. Gable stones with extensions have also been seen in collapsed combs.

Plate 6: Fallen gable stones no longer serving to support the side slabs, which have begun to collapse, Mt. Pisgah Cem., Doyle quad, White Co.

Observation of numerous combs suggests that in most cases one side slab (commonly the right hand side as viewed from the head of the grave toward the foot) was laid down first, supported by the two already set gable stones. Next the second side slab (typically the left hand side) was laid down, resting on the gable stones and slightly overlapping the first laid stone at the crest of the comb. While there are combs in which the two side slabs meet at the crest with no overlap, there does seem to have been a tendency to lap the stones, one over the other. If this lap served any purpose, it could have been to minimize the entry of rain into the comb and onto the grave itself.

A very few instances are known where sandstone side slabs have been beveled to meet neatly at the crest of the comb (Plate 7). This nice bit of stone dressing probably is practical mainly where thicker than normal sandstone slabs are employed4.

Plate 7: Thicker than normal sandstone side slabs used with beveled edge combs at Philadelphia Cem., Irving College quad, Grundy Co.

While gable stones are the most common form of support for the comb side slabs, a noteworthy alternative support system involved the use of a thick iron rod running the full length of the grave. The rod, actually a long bolt, has a head at one end and is threaded at the other end. A hole was bored through the headstone and another carefully positioned hole through the smaller, but stylistically matching, footstone (Plate 8). Typically, the bolt was run through the headstone first and then through the footstone. The heavy nut was then turned onto the threads behind the footstone. Probably, the sideslabs were laid before the nut was fully tightened, allowing the head- and footstone positions to be adjusted slightly if needed to mate with the side slabs. In any case, one side slab (typically the right) was laid on the bolt rod itself, and the second side slab (typically the left) was laid against the upper end of the first laid slab (again resulting in lapped comb slabs). When all four stones were in place, the nut could be tightened to draw the head- and footstones tight against the ends of the side slabs. In this system, no supporting gable stones are necessary under the side slabs, but the system requires both head- and footstones, and the holes for the rod had to be accurately placed.

Plate 8: Bolted combs in Falling Springs Cem., Crawford quad, Overton Co. Note that when a side slab slips off the bolt, it has no support at either end and the structure begins to collapse.

The bolt support system was very popular in Overton county and is especially common (almost ubiquitous) in the Allred area, leading to the supposition that it was invented by the Vaughn family of stonecutters. The Vaughns are known to have cut many of the distinctive “truncated triangle” head- and footstones (Plate 8) that are the most common type of comb graves to use the bolt support system. Graveyards such as Falling Springs Cem. (Plate 9) at Allred are impressive both for the number of combs and for the stark geometry of the “truncated triangle” combs5.

Plate 9: Falling Springs Cem., Crawford quad, Overton Co, not far from the Vaughn quarry.

Where the planar-bedded Hartselle was not available, sandstone still seems to have been the preferred rock type. The vast majority of combs are made of sandstone from one source or another.

However, where appropriate sandstone was unavailable, other materials came into play, notably limestone. Limestone is not commonly planar-bedded, and, for that matter, is more often thicker bedded than thin. Nonetheless, limestone (or, in one instance, dolostone, a magnesium-rich variant of limestone) was used for at least 63 combs in 24 cemeteries, primarily where sandstone evidently was not locally available. Some of the limestone combs are made of very finely dressed stone (Plate 10). The thicker limestone slabs sometimes feature beveled edges (Plate 11) or other innovative shapes (Plate 12) In other cases the limestone combs are relatively crude (Plate 13). Phillips Cem. (Hillsboro quad, Coffee Co.) features three limestone combs along with ten combs made of rather rough, thick-bedded sandstone slabs.

Plate 10: Finely dressed limestone combs in Stephenson Cem., Hillsboro quad, Coffee Co.

Plate 11: Beveled edged limestone combs in Perkins Cem., Winchester quad, Franklin Co. These are the southernmost known combs in the main Tennessee comb range.

Plate 12: Limestone comb featuring notched side slabs supported at the head end by a flared base on the headstone, Cash Cem., Alto quad, Coffee Co.

Plate 13: Relatively crude limestone comb in Warren Cem., Burrow Cove quad, Grundy Co.

Figure 2 shows the distribution of 19 of these 24 limestone comb cemeteries; Appendix B is the complete inventory. With the exception of the three combs at Shellsford Cem. (Cardwell Mountain quad, Warren Co.), four combs at Albertson Cem. (Riverton quad, Fentress Co.), and a very crude comb in Gist-Anderson Cem. (Doyle quad, White Co.), all the limestone combs are found along the margins of the comb range. The limestone combs on the western edge of the comb range may be attributed to distance from the Hartselle outcrop band. The limestone combs in the southernmost part of the comb range probably reflect the fact that the Hartselle Formation in this region contains less sandstone and more shale. Two limestone combs near the northern end of the comb range may also reflect a lack of sandstone in the Hartselle, though this is not certain.

A single crude comb, possibly unique6, made of slabs of Chattanooga Shale (Plate 14) exists, along with a normal sandstone comb, in Russell Cem. (Buffalo Valley quad, Putnam Co.). This shale comb, just 43-44 inches long, covers the grave of an infant. The presence of a sandstone comb in the same small graveyard raises the question as to why the shale was used when sandstone could apparently be had.

The Chattanooga Shale is a very distinctive rock unit, easily split into thin, planar, black slabs that are attractive when fresh. This feature, plus the fact that it was locally available, probably free for the taking, may explain its use on this grave. Unfortunately, the Chattanooga is so friable that extracting a slab big enough to run even the length of an infant’s grave would have been difficult and this crude comb appears to have been constructed of several relatively small pieces. Additionally, the black shale contains fine grains of pyrite (iron sulfide) which weather easily and promote the relatively rapid disintegration of the shale. Therefore the Chattanooga Shale is not a good choice for grave markers, in spite of its fissility.

Plate 14: Crude, perhaps “homemade” comb made of pieces of Chattanooga Shale, covering the grave of an infant. Russell Cem., Buffalo Valley quad, Putnam Co.

Two marble combs are known, one at Eureka Church Cem. (Welchland quad, Warren Co.) (Plate 15), dated 1899. The other is a pink marble child’s comb at Perkins Cem. (Winchester quad, Franklin Co.), dated 1890. It is reasonable to assume that this use of expensive imported stone was done for elegance rather than necessity. (Figure 3 shows the distribution of the marble and shale combs; Appendix C is the inventory.)

Plate 15: Fancy striped marble comb at Eureka Cem., Welchland quad, Warren Co.

Regardless of the rock type chosen, stone combs were invariably erected without recourse to cement for structural integrity. A very few combs have been found (in 11 different cemeteries) to have cement along the cracks between stones (Plate 16), but this appears to have been a means of sealing the comb against the entry of rainwater, rather than for any structural support.

Plate 16: Comb with crest line cemented (rather than lapped), presumably to keep out rainwater, Little Cem., Livingston quad, Overton Co.

On the other hand, 10 combs made entirely of poured cement or concrete (Plate 17) are known from five cemeteries. Of these cemeteries, two (Lancaster Cem. and Rock Springs Cem., Buffalo Valley quad, Smith Co.) are located on the western margin of the comb range and one (Curlee Cem., Readyville quad, Cannon Co.) lies further west, outside the main comb range. The other two cemeteries containing cement/concrete combs (Okalona and Liberty cemeteries, Okalona quad, Overton Co.) are also populated with numerous sandstone combs. In these latter cases the use of cement/concrete must reflect a preference. Indeed, judging from the style of these combs, the same man constructed all three of the cement/concrete combs in these two cemeteries. (Figure 4 shows the distribution of the cement/concrete combs; Appendix D is the inventory.)

One of the striking characteristics of all these cement/concrete combs is that they are molded to mimic the basic form of two slabs resting on triangular gable supports, just like ordinary sandstone combs.

Plate 17: Plate 17: Matching husband and wife combs of concrete, Liberty Cem., Okalona quad, Overton Co.

Finally, a number of combs have been constructed of sheet metal over wooden frames. The combs are made of corrugated “tin” (Plate18), flat sheet metal (Plate 19), and 5-Vee roofing (Plate 20). During this survey, 26 metal combs were found in 13 graveyards, widely scattered throughout the comb range, some on the fringes, some in the heart (Figure 5 and Appendix E). The distribution of these metal combs is not easily tied to geology. It may be that some examples along the western margin of the comb range relate to the lack of sandstone. But this cannot be said of metal combs in the central part of the comb range. It may be that cost was a factor in some cases. It may be that more men were capable of the carpentry necessary to create a sheet metal comb than were qualified to quarry stone. Two of the three most recent combs known (1983 and 2001) are metal combs (Plate 20) erected by the same man for members of his family (Finch, 2013)7.

Plate 18: Combs of corrugated roofing metal over wooden frames; Pierce Cem., Obey City quad, Overton Co.

Plate 19: Combs of flat sheet metal, riveted and sealed, Cunningham Cem., Dry Valley quad, White Co.

Plate 20: Combs of 5-Vee roofing over wooden frames with plywood end caps, erected 1983 and 2001, Bear Creek Cem., Cookeville East quad, Putnam Co.

Ball (1977, Fig. 4) described a grave cover in Lambert Cem. (Hillsboro quad, Coffee Co.) that was a metal comb grave, albeit somewhat unusual in having a low (two inches above ground level) poured concrete base. Although this structure was a complete ruin by the time of the present study, Lambert Cem. is counted as the thirteenth metal comb location.

Being supported by wooden frameworks, metal combs are obviously not as durable as stone combs. Of the 26 metal combs known, around a half dozen seen during the early stages of this survey have since collapsed or otherwise been destroyed.

Geologic influence on the geographic range of the comb graves:

The general distribution of the Tennessee comb graves has been described in the introduction as forming a band paralleling the western escarpment of the Cumberland Plateau. The preferential use of thin slabs of Hartselle sandstone for comb construction has been noted. From these two facts it may be argued that there was a geologic control on the distribution of the comb graves.

The Hartselle Formation is a stratigraphic unit deposited during the Mississippian Period (350-299 million years before present). Its type section is near Hartselle, Alabama. In Tennessee the Hartselle crops out along the eroded western slopes of the Cumberland Plateau, where it forms a continuous north-south outcrop band across the state, typically at elevations around 1100 –1400 ft. (The Hartselle outcrop trace is shown in the figures.) It is also found in many erosional outliers of the formerly more extensive Plateau, such as Gum Spring Mountain, just west of Sparta.

The Hartselle Formation consists mainly of sandstone and shale, with the relative proportions of the two rock types varying significantly over the length of its outcrop belt. Where well-indurated, quartz-cemented sandstone is prominent, the Hartselle forms a resistant unit which erodes more slowly than the overlying Bangor Limestone. As the Bangor is eroded back, a nearly flat topographic surface, known as the Hartselle Bench, is formed on top of the Hartselle. Where well developed, the Hartselle Bench forms a distinct stair-step in the western escarpment of the Cumberland Plateau. The Hartselle Bench also forms the flat upper surface of various of the erosional outliers lying to the west of the main Plateau. It is in these areas, where the Hartselle Formation contains significant amounts of sandstone, that the comb graves are found in abundance.

South of Warren county, the Hartselle Formation becomes more shaly, with a lower sandstone content8. It is in this region that most limestone combs are found. Not much further to the south, the Tennessee combs die out.

The Hartselle Formation is not found on the east side of the Cumberland Plateau; nor are any combs found here.

The range of the Tennessee combs also dies out a little north of the Tennessee-Kentucky line, for reasons that are at present obscure. Nonetheless, it is fair to say that in Tennessee, the comb grave tradition was strongly influenced by the presence or absence of appropriate sandstone in the Hartselle Formation, and that the primary control on the geographic range of the combs was the lithology of the Hartselle.

In sum, combs are found in graveyards on the Hartselle Bench, in cemeteries along the western portion of the Plateau relatively close to the Hartselle outcrop band, and on the Eastern Highland Rim below the Plateau. Combs become rarer (or made of substitute materials such as limestone, concrete and sheet metal) in areas distant east or west of the Hartselle outcrop band, the distance likely related to the difficulty of moving the heavy stones via 19th- and early 20th-century transportation facilities.

Local concentrations within the comb range:

Cantrell (1981) visited 168 cemeteries, finding 1104 comb graves and noting two distinct areas of comb concentration, which he designated the Caney Fork Valley group (White and northern Van Buren counties) and the Overton County group. These same areas of comb concentration were documented during the present survey.

Doyle quad (White and Van Buren counties) contains 38 comb cemeteries (the 2nd greatest number of comb cemeteries in a single quad), with a total of 546 extant combs, by far the highest comb count for a single quad. The adjacent Cassville (34 comb cemeteries) and Bald Knob (29 comb cemeteries) quads have comb counts of 423 and 350 respectively. Thus, these three quads alone account for 36% of the 3662 extant combs in the main Tennessee comb range. Not surprisingly, these three quads also contain the individual cemeteries with the highest comb counts: Mt. Gilead Cem. (Cassville quad), 138 combs; Mt. Pisgah Cem. (Doyle quad), 126 combs; Old Union Cem. (Bald Knob quad), 102 combs.

Combs are numerous in Overton Co., with Crawford quad presenting the highest comb cemetery density of any quad: 23 out of 24 cemeteries, i.e., 96% have combs, for a total extant comb count of 435, the second highest count of any quad. The Falling Springs Cem., with a comb count of 102, is located in the community of Allred, as is the old Vaughn stone quarry, source of so many comb rocks. Okalona quad contains 45 comb cemeteries, the highest number of any quad, and has an extant comb count of 387. Thus the Crawford and Okalona quads alone account for 22% of all extant combs in the main Tennessee comb range.

Beyond the pale:

The geographic survey done for this study has defined the main range (Figure 1) of the Tennessee comb graves by mapping out the comb cemeteries in each 7.5 min quad and establishing a border of “comb-free” quads around the range. However, as should be expected for a once-popular and deeply entrenched burial custom, the use of combs spread beyond the boundaries outlined above. There are outliers, combs “beyond the pale”. In the Readyville quad, well west of the core comb range, a cement comb exists at Curlee Cem. (Cannon Co.) and a sheet metal comb was photographed in Science Hill Cem. (Rutherford Co.). To the south and west, two children’s combs, made of limestone, are known from Shofner Cem. (Normandy quad, Bedford Co.), and two additional limestone combs are found in Roper Cem. (Dellrose quad, Lincoln Co.). The Roper combs are the south-westernmost combs known in Tennessee.

Including these outliers, the total number of Tennessee comb graveyards rises to 520 and the total number of extant combs to 3668. A statewide search would undoubtedly reveal more comb graves, but these scattered instances of combs--likely the result of the practice being introduced by families migrating westward from the main comb region—do not represent a widespread coherent local tradition in these areas.

Temporal range of the comb graves:

Before considering what is the earliest known comb, it needs to be noted that there is no easy way today to scientifically ascertain the exact date when a comb was erected. Crissman (1994) emphasizes that gravehouses were erected “almost as soon as the burial was completed”, and it seems likely that most combs were also erected not long after the burial. Thus, the death year (if any) on the headstone is, in most cases, the date of the comb itself. However, it is known in some cases that combs were erected over graves years after the actual burial.

Several sheet metal combs in Pierce Cem. (Obey City quad, Overton Co.) are likely younger than the graves they cover, as suggested by the apparent newness of the wooden frames supporting the combs. Four cement combs in Lancaster Cem. (Buffalo Valley quad, Smith Co.) are all structurally connected and appear to have been constructed concurrently, some time after the most recent date of death, 1921, which is some three years after the earliest date of death, 1918. In Stockton Cem. (Stockton quad, Fentress Co.) there are five headstones with dates ranging from 1847 to 1862, that all appear to be the work of a prolific stonecutter who operated in the White Co. area. A Stockton family tradition holds that these stones were brought on an oxen-pulled sled from White Co., an exceptionally long and tedious trip for hauling gravestones. It is thought by a Stockton family member9 that all five stones were probably commissioned and brought to the family burying ground at the same time. This assumption is far more reasonable than assuming that five separate trips were made over such a distance (66 miles by today’s highway system). If true, then at least one of the Stockton combs was erected 15 or more years after the date of death.

In spite of the foregoing examples, unless there is evidence to the contrary, the year of death inscribed on a comb grave is considered to be the date of the comb itself.

The earliest dated comb grave thus far found is in the Herd-Hurd-Bryant Cem. (Cassville quad) in White Co., the comb of William W. Herd, who was “Born Oct. 10, 1816, Died the Same Day” (Plate 21). Remarkably, this same small cemetery also contains combs dated 1817 and 1822 --the 2nd and 3rd oldest comb dates—in the same row of graves with the 1816 comb. The 1816 and 1822 combs have simple round-topped headstones, whereas the 1817 comb (Plate 22) features a necked discoid headstone (McNerney, 2017).

In an earlier version of this paper, it was stated that the comb grave of Rosey Hutson, found in Mt. Pisgah Cem. (Doyle quad, White Co.) bore the oldest known date on a comb, 1817, inscribed on a headstone of the necked discoid type. However, based on stylistic considerations, it was thought that this particular comb was likely erected some years after the death date10. The discovery of the Herd-Hurd-Bryant Cem. with a second necked discoid dated 1817 (Plate 22) renders it more likely that the 1817 date in the Mt. Pisgah Cem. is the actual date the comb was erected. As noted earlier, it is always possible that a comb was erected in a later year than the date of death, and in the case of the Stockton combs, a group of five combs were almost certainly erected years after the death dates. But there is not, at present, any reason to suspect that the combs at the Herd-Hurd-Bryant Cem are significantly younger than the dates they bear.

Plate 21: Marker for the oldest dated comb. Herd-Hurd-Bryant Cem., Cassville quad, White Co.

Plate 22: Morgan Bryant comb, date of death 1817, with a necked discoid headstone. Herd-Hurd-Bryant Cem, Cassville quad, White Co.

In addition to the early dated combs, there are numerous uninscribed combs, commonly (but not always) of more rustic construction, that may have been erected before some of the dated combs. In the Cold Hollow Cem. (Bald Knob quad, Van Buren Co.) an uninscribed comb covers the grave of George T. Sparkman, who died in 1816 (Plate 23). In Austin (Anderson) Cem. (DeRossett quad, White Co.) an uninscribed comb covers the grave of Rachel Austin, who died in 1818.

Plate 23: Uninscribed comb for Geo. T. Sparkman, deceased 1816, Cold Hollow Cem. (formerly listed as Sparkman-Yates Cem.), Bald Knob quad, Van Buren Co.

At the present time, a total of five combs (three dated, two uninscribed) are known covering pre-1820 burials. Although the age of the earliest comb will probably never be definitively established, the available data suggest that the comb grave tradition was aborning during the years 1815-1820 in White Co., where most of the oldest combs are found, and where the Hartselle sandstone was widely available.

In addition to the 1822 comb in the Herd-Hurd-Bryant Cem., several combs bearing mid-1820’s dates are found in Wilson Cem. (Sparta quad, White Co.) and in Austin (Anderson) Cem. (DeRossett quad, White Co.). An 1822 comb is known to the north in Roaring River Cem. (Okalona quad, Overton Co.), and an 1828 comb is found to the south in Blue Springs Cem. (Viola quad, Warren Co.). Clearly by the mid-1820s the Tennessee comb grave tradition was well established in the White Co. area and strong enough to begin to spread to north and south.

Appendix F is a partial listing of combs with early dates and their locations; their distribution is shown in Figure 6.

The comb grave style appears to have “caught on” and spread rapidly throughout its full range. A comb dated 1835 is found in Perkins Cem. (Winchester quad, Franklin Co.) at the southernmost end of the main comb range, and a comb dated 1839 may be seen in Fitzgerald Cem. (Dale Hollow Dam quad, Clay Co.) near the northern end of the comb range. By the mid-19th century comb graves were very popular throughout the comb range, and in some cemeteries became the dominant form of burial.

Cantrell (1981) meticulously logged the death dates inscribed on over 700 combs in his Caney Fork Valley and Overton County groups. His plots of these data show that the use of comb graves peaked in the 1876-86 decade in White and Van Buren counties, and around 1906 in Overton Co. The plots indicate a rapid decline in the erection of combs in both areas, with some combs still being erected in the Overton Co. area into the 1950s, some three decades longer than in the Caney Fork Valley area.

No collection of dates comparable to Cantrell’s work has been compiled for the present study; however a number of more recent combs were noted. Combs dated 1956 and 1958 were noted in Flat Creek Cem. (Hilham quad, Overton Co.) during the early stages of this project, but have since been removed. Comb graves dated 1967 and 1969 occur in Stockton Cem. (Stockton quad, Fentress Co.), and a pair of sheet metal combs cover the graves of a husband (d. 1983) and wife (d. 2001) in Bear Creek Cem., (Plate 20, Cookeville East quad, Putnam Co.)7. To date, the most recent comb is that of Kathleen Rodgers, who was buried at her home in 2012 (Plate 24, Windle quad, Overton Co.). The two 21st century combs (2001 and 2012) and the separate circumstances leading to their erection have been described in detail by Finch (2013).

Plate 24: Comb grave of Kathleen Rodgers, erected in 2012, Windle quad, Overton Co. Made of Hartselle sandstone slabs from the Walker quarry in Overton Co.

In sum, the comb tradition spans nearly two centuries of Tennessee cultural history, from around 1816 to at least 2012. Perhaps the custom will yet continue.

The purpose of the combs:

“What was the purpose of combs?” is a commonly asked question, one for which there is no definitive or “one size fits all” answer.

Hoping to answer this question, Cantrell (1981) interviewed funeral directors, a monument dealer, other cemetery researchers, a longtime resident of Allred (in the heart of the Overton County comb group), and even a member of the Vaughn family of stonecutters who had quarried gravestones in his younger days. The most common reason cited was to protect the grave from rain. The second most common reason given was to protect the grave from animals, whether domestic or wild. At least one informant said that one purpose for the combs was to make the grave more permanent. Oddly enough, the retired stonecutter said he knew of no reason for combs other than people wanted them.

During the course of the present survey, informal conversations with a few old timers also elicited the same reasons: protection from rain or from animals. With regard to rain, this notion may be supported by the observation of combs sealed with cement in 11 graveyards, by a caulk-like substance in one cemetery, and by metal combs sealed with a tar-like substance in another graveyard. As noted earlier, the normal lapping of one side slab over the other might be for the same reason. However, protecting the grave from rain does not bear up well to logic, inasmuch as there is no real purpose in doing so. Cold logic, of course, cannot be relied upon to explain the reasons for customs surrounding emotional events such as burials.

With regard to protecting the grave from animals, one person remarked that the soil was so rocky that the graves were not very deep, and therefore had to be protected from being dug up by wild animals. This cannot be the general reason, inasmuch as combs are found in many different soil conditions.

An Overton Co. man explaining why he had a box grave erected over his mother in 1954 stated that she requested it, fearing that the graveyard –located in a field—might become neglected and she did not want the cows to walk on her. A second informant reported that as a boy he visited numerous cemeteries with his father who told him that combs were erected to prevent unfenced cattle from walking on graves. “I remember him demonstrating with his foot how a cow's foot would slip off of the slanted side [of a comb].”11 It is certainly true that cattle can severely damage a graveyard by rubbing against headstones to scratch themselves. Worse yet, pigs, rubbing against and rooting underneath gravestones, can and do topple markers and bury them in the mud.

On the other hand, cattle may have been deliberately allowed into some graveyards. According to the White county cemetery book (Pollard, 2003), “In the days before power mowers, the easiest way to keep a cemetery mowed was to allow livestock to graze it. The cover [comb] protected the grave from…trampling.”

While some combs may well have been erected to protect the grave from animals, personal observations show that many a groundhog has found a happy home beneath a comb, and the occasional blacksnake has taken refuge there.

Some combs were erected because they were specifically requested by the decedent or by a family member. Myrtle Webb Judd asked her brother Clio Webb to build a comb for her husband Joe, who died in 1983. Whether or not Joe himself wanted a comb is not known, but he may have, as his parents were buried beneath combs in the same cemetery where he was interred (Bear Creek Cem., Cookeville East quad, Putnam Co.). According to Clio, Myrtle wanted a comb to keep the grave dry. Myrtle also requested a comb for herself, matching that of her husband, and Clio built hers, too, when she died in 2001 (Plate 20) (Finch, 2013)7.

Ball (1999), noting the “desire to protect the burial from the elements or animal disturbance”, concluded that comb graves served the same purpose as gravehouses and other types of grave covers: “Functionally, [combs] are but one of a class of grave marker designed to cover the entire grave which fulfilled an emotional need to provide a level of protection to the remains of the deceased from rain and other factors.”

While agreeing with Ball, I would add that combs make graves more visible and more permanent, less susceptible to being lost (when the cemetery is abandoned) or obliterated (by rain, animals, or vandals). This is an important effect of the comb structures, and was probably the intent or reason behind the erection of some combs. Twenty lone combs were noted in this survey, two in the side yards of homes, one behind a barn, six in pastures/fields and 11 “lost” in thickly wooded areas. Headstones alone might or might not have preserved these grave locations; any or all of these graves could easily have completely disappeared had they not been covered with combs.

None of the above effectively explains why combs became so popular in the comb range. Other forms of grave covers were known and used to some extent. Box graves were more expensive than combs12. Gravehouses, commonly constructed with a wooden frame, did not last as long. Cairn graves and some coffin graves required more stone cutting and dressing labor. All of these forms of grave covers served the same purpose as combs; however, none of these alternate forms of grave covers achieved the enormous popularity of the comb graves.

The very rapid spread of the comb tradition throughout the length of the comb range, documented earlier in this article, suggests that combs may have first become a sort of graveyard “fad,” and later an enduring fashion. They were relatively simple to construct, cheaper than some alternative forms of grave cover, and served a variety of purposes, both emotional and functional, including marking the grave in a more permanent manner than a simple headstone. Combs became fashionable, and like all fashions, this one ran its course. As modern, factory-produced stones of marble and granite became widely available, they became more popular. Concomitantly, the use of combs made by local artisans of local stone declined and nearly died out.

Origin of the comb grave custom:

Cantrell (1981) concluded that “the early popularity of these structures [combs] indicates that the settlers were already familiar with the tradition and brought it with them to Tennessee.” This would suggest that the use of combs came to Tennessee from North Carolina. But Cantrell adds that “preliminary inquiries about the existence of comb graves in North Carolina have yielded negative results”.

More recently, a single comb grave has been located in the old City Cemetery of Raleigh, N.C. It is made of weakly foliated granitic rock (Plate 25), and it bears the date 1834, making it more than a decade younger than the oldest known combs in Tennessee. To date, there is no evidence for a North Carolina origin for the comb style.

Plate 25: Comb grave of Mary Beasle, died 1834; in old City Cemetery, Raleigh, NC. The comb is made of weakly foliated granitic rock. (Image courtesy of Susan K. Wiley.)

In Tennessee, no combs have been reported east of the Cumberland Plateau, where they would be expected had the style been brought from North Carolina.

That said, one must admit that emigrants from North Carolina westward into Tennessee undoubtedly carried with them familiarity with the concept of covered graves, such as ledger graves and box graves. Ball (2005, 2008) makes a strong logical case for his supposition that the idea, if not the style, behind comb graves (as well as gravehouses and other forms of grave covers) derives from box and stone table graves popularized in England and Scotland during the 18th century. Several types of grave covers can be seen in the Old Burying Grounds at Beaufort, N.C., including a number of graves covered by comb-shaped structures made of carefully fitted bricks (Plate 26). By construction technique the graves are cairn graves, but the resulting form is distinctly comb-like. None are known to pre-date the Tennessee combs (the four grave covers pictured date from the mid-nineteenth century) and it seems unlikely that these comb-like grave covers inspired the Tennessee combs or vice-versa. The similarity of form is most likely a coincidental repetition of a basic shape.

Plate 26: Comb-like mid-19th century grave covers made of carefully fitted bricks, in Old Burying Grounds, Beaufort, N.C. (Image courtesy Beaufort Historical Assoc.)

While considering a possible source for the Tennessee comb grave custom, one must ask if there are comb graves in England. And the answer is yes, for in 2015 a single comb grave was noted in the cemetery at Otley All Saints Church, in West Yorkshire, England13. This comb is essentially identical in form and construction to the Tennessee combs in that it covers a single grave, and is made of two long sandstone slabs supported at the ends by triangular gable stones. Unfortunately, it is uninscribed except for the single letter “P”, so at the time of this writing, its age is only hinted at by the nearest surrounding graves which range from 1794 to 1856. The Otley comb could be older or younger than the oldest Tennessee combs. If it is older, was it the progenitor of the Tennessee comb grave tradition? The answer to this question is “probably not.” As far as is currently known, the Otley comb is an isolated example and not part of any established tradition. It is unlikely that a single English comb grave would inspire a widespread custom in central Tennessee without any intermediate occurrences in North Carolina or eastern Tennessee.

Several stone grave covers or “tomb shrines” of triangular form are also known in Ireland14. Three are found in County Kerry (two on the Iveragh Peninsula, one on Illaunloughan Island) and two are located at St. Cronan’s Temple in County Clare. (See ) One of the tomb shrines at St. Cronan’s Temple is thought to cover the grave of St. Cronan himself and date to the 7th century. It has been described as “made of only four pieces of stone, two for the sides, two for the ends, rather like a tent. One end piece is only a half piece—allowing the disciple to put his hand inside and touch the earth where his dear one is buried.” (See )

The Irish tomb shrines have a distinctive look, being much more steeply pitched than the Tennessee combs. Nonetheless, we clearly have a type of Irish comb grave tradition that predates the Tennessee tradition. Was it the inspiration for the later tradition? The difference in overall appearance, the great difference in age, and the apparent paucity of examples in Ireland, along with the lack of any “trail” through North Carolina and eastern Tennessee mitigate against --but do not disprove—an Irish origin for the Tennessee comb grave custom.

The similarities between the Tennessee combs and these scattered examples in the UK are most easily attributed to two facts: 1) many cultures evince a desire to protect graves by some means, and 2) a comb-type grave cover is a simple geometric construction that could easily be invented by different peoples at different times.

Based on the evidence currently available, Ball (1999) is correct in stating that the comb grave “likely originated in the early 19th century in the northeastern Highland Rim area of middle Tennessee as a regional adaption of a yet earlier prototype grave covering.” What that earlier prototype was may never be known with certainty.

Regardless of conceptual inspiration, it seems that the comb grave custom as known in the Upland South originated in Tennessee, in the White Co. area, in the 1815-1820 period. The use of combs spread rapidly throughout the main Tennessee range, achieved a popularity that greatly exceeded that of other types of grave covers, and, as discussed below, was later carried into other areas.

Religious connections:

Ball’s (1997) assertion that comb graves “seem to have no particular association with any churches in the region”, is reconfirmed by the present survey. Seventy-two comb cemeteries were found associated with churches, 69 in the main Tennessee comb range and three outside the main range. The church denominations, where known, were recorded. A few church buildings are known to have changed from one denomination to another; in these cases, the earlier denomination was recorded. A few church buildings are now in disuse with no indication of their denomination and in one case the church building has long since been razed.

The denominationally identified churches include 30 Baptist churches, 18 Methodist, 11 Churches of Christ, 5 Presbyterian, 1 Christian Church, 1 Lutheran church. Six “other” or unidentified churches complete the list. Appendix G is the inventory.) summarizes these findings.

While it is clear the comb grave tradition is not tied to any particular denomination, the birth of the comb fashion coincides closely with the peak of the religious movement known as the Second Great Awakening15. Whether or not the combs are related in any way to this movement remains unclear, however Ball16 feels that “At a minimum, the time lag between the Great Awakening and the earliest known comb graves seems too close for mere coincidence.”

Comb graves in other states:

In addition to Tennessee, combs are known from ten other states, mostly Southern states: Alabama, Arkansas, Kentucky, West Virginia, Mississippi, Texas, Louisiana, and North Carolina. A few combs are also known in eastern Oklahoma and southernmost Missouri, both regions being culturally part of the Upland South (Jordan-Bychkov, 2003).

Alabama has a large number of cemeteries with comb graves, 69 of which were verified during the course of this study. In order to compare the Alabama and Tennessee comb grave features, 38 of these cemeteries were visited and photographs were obtained of the great majority of the remaining 31 comb cemeteries. These 69 Alabama comb cemeteries are found in the northern half of the state, in Blount, Cherokee, Cullman, Etowah, Fayette, Franklin, Jackson, Jefferson, Marion, Morgan, Tuscaloosa, Walker and Winston counties. Tuscaloosa and Walker counties seem to be areas of comb concentrations, having 30 and 18 comb cemeteries, respectively, of the total catalogued. Comb counts were made in 34 cemeteries with a total of 338 combs (292 extant combs plus at least 46 ex-combs). If the average of this sample is representative –which is by no means a given-- the total number of combs in the 69 comb cemeteries should be in the 680-690 range. However, these 69 graveyards represent less than a third of the comb cemeteries known in Alabama: cemetery researcher Dr. Ray Hutchison17 has documented 231 comb cemeteries containing 1251 combs. Furthermore, Hutchison has found comb cemeteries in two additional counties, Bibb and Lawrence.

It is assumed that a comb grave cover is erected at the time of the interment or not long after, so that the recorded death year is, in most cases, the year the comb was erected. Death year dates were collected from 179 combs, ranging from 1848 to 1932, with dates in the 1870s, ‘80s, and ‘90s accounting for 70% of all dates. Brown (2004) also found that “most Alabama combs date between the 1840s and World War I.”

The Gravlee-Nall Cem. in Fayette Co. is notable for a row of six virtually identical combs, constructed of finely cut grey limestone and marked with small flat granite headstones dated 1920, 1930, 1930, 1978, 1980 and 1984. The uniformity of the combs and headstones suggests that all six of these combs were erected at the same time, 1984 or later. A Gravlee family member confirmed that his grandfather, being an admirer of the comb grave tradition, had all six of these combs erected, at the same time, over the graves of these family members, some of whom had been deceased for many decades18. Thus the Gravlee-Nall graveyard shows that there are exceptions to the assumption that dates of death indicate the dates of comb construction. Furthermore, Gravlee-Nall, though something of an outlier in time, shows that the comb grave tradition did not completely die out in Alabama in the first half of the 20th century.

The Cargile Cem. (Jackson Co.), containing three limestone combs, may be the northernmost Alabama comb cemetery. It is the closest known Alabama comb cemetery to the main Tennessee comb range, lying just 23 miles from the southernmost cemetery in the main Tennessee range, the Perkins Cem. (Winchester quad, Franklin Co.).

The Roberts family cemetery in Morgan county contains several combs and is of particular interest as it is reported19 that the Roberts family moved to Alabama from White Co., Tennessee. It would appear they brought their burial customs with them, but inasmuch as the oldest dated comb in this graveyard is inscribed 1882, the Roberts clearly were not the first to bring comb graves to Alabama.

Of the 69 Alabama comb cemeteries recorded for this study, almost half (30) are found in the northern half of Tuscaloosa Co., where sandstone suitable for comb slabs occurs. In addition to this geologic influence, it is also interesting to note that local historians say that this portion of the county was settled by emigrants from Tennessee20.

As in Tennessee, the preferred stone for comb construction in Alabama was sandstone. The Hartselle Formation, present in the Alabama comb grave region (the type section is near the town of Hartselle, whence the name), is predominantly quartz sandstone, and probably was the source of some comb slabs. Another unit quarried for sandstone in Alabama is the Pottsville Formation of Pennsylvanian age21. The Pottsville contains sandstone of dimension stone quality, despite a tendency to spall upon weathering. It was quarried for the original state capitol in Tuscaloosa and for stone used in locks and dams on the Black Warrior River, and for some of the early tombstones in Tuscaloosa’s Greenwood Cemetery22. A comparison of the locations of the 69 comb cemeteries verified in this study with the geologic map of Alabama shows that the majority (46) are located on outcrops of the Pottsville Fm. or, if actually sited on other stratigraphic units, then very close to Pottsville outcrops (12). Only a single comb cemetery was found to be located on the Hartselle, with one additional cemetery being located close to Hartselle outcrops. Considering the transportation difficulties during the era of comb erection, it is only reasonable to assume that local availability determined which unit supplied the sandstone slabs used in the comb graveyards.

Regarding the geographic range of the Alabama comb cemeteries, all but two of the Alabama counties in which combs have been reported lie partially or wholly within the Cumberland Plateau physiographic province. (The exceptions are Cherokee Co., which lies in the Valley and Ridge province, and Bibb Co., which lies just outside but near the southern edge of the Cumberland Plateau province.) In Alabama the Cumberland Plateau is more dissected and discontinuous than in Tennessee, and Pottsville (and to a much lesser degree, Hartselle) outcrops are found over a broad region. If, as seems certain, these units supplied most of the comb slabs, this may explain why combs are found not in a linear belt as in Tennessee, but scattered over an irregular geographic area as broad from east to west as from north to south. Nonetheless, the Alabama comb range appears to be controlled by the distribution of suitable sandstone outcrops, just as in Tennessee.

Taking a multistate viewpoint, it could be argued that the Alabama comb cemeteries are in fact a continuation of the Tennessee comb range. However, there is a gap --albeit not a large one-- between the two collections of combs, both geographically and, it would seem, temporally. The use of comb graves in Alabama developed later than in Tennessee.

In Arkansas, comb graves have been documented in at least 17 cemeteries in 10 different counties, with a total of something over 50 combs. Almost all of these comb cemeteries are found in the Ozark Mountains region (northwest and north-central Arkansas), in Benton, Boone, Cleburne, Crawford, Franklin, Fulton, Newton, Searcy and Sharp counties. The single known comb cemetery lying outside of the Ozarks is found in Pike Co. in southwest Arkansas. Seven of the Ozark cemeteries were visited during the field work for this study. Most of the observed combs did not feature dates, but those that did ranged from 1855 to 1883.

Again, most combs were made of sandstone slabs, but a few were of limestone, including very finely dressed limestone (coarsely crystalline pink limestone, likely sold as marble) combs in Holmes Cem. in Boone Co.

The known comb grave cemeteries in Arkansas seem to be scattered over a wider area than the relatively narrow geographic belt containing the Tennessee combs. This may indicate that more comb grave cemeteries remain to be discovered in Arkansas. One might also infer from this distribution that the Arkansas combs result from the “Tennessee diaspora” described by Jordan-Bychkov (2003). One of the eight combs in Holmes Cem. covers the grave of a woman originally from Tennessee23 and Tennesseans from Overton Co. (including stonemasons from the Allred and Norris families) are known to have migrated into the Ozark region of northwest Arkansas24. In her book on burial customs of the Arkansas Ozarks, Burnett (2014) states that “older burials were protected by a variety of grave coverings, their styles [her list includes comb graves] having all been traced back to Tennessee.”

Combs are known from eight Kentucky cemeteries, one of which, Taylor Grove Cem., is included in the northernmost end of the main Tennessee comb grave range. Four of the other seven comb cemeteries visited during fieldwork for this study –Redbird, Steely, and Canada Church cemeteries (Wofford quad, Whitley Co.) and Ballou-Worley Cem. (Cumberland Falls quad, McCreary Co.)-- all lie 40 to 45 miles northeast of the easternmost Tennessee comb cemetery. Redbird Cem. was reported by Richmond (1998) and Ball (1999); it contains six combs, made of sandstone (as would be expected, considering the geology of the Cumberland Mountains). Three of the six combs are dated: 1862, 1863, and 1922(?). Nearby Steely Cem. once had three combs, now dismantled, and Canada Church cemetery has three combs. Ballou-Worley Cem. contains a single sandstone comb dated 1860.

The remaining three Kentucky comb cemeteries are found in western Kentucky, between Hopkinsville and Lake Barkley, well to the west of the main Tennessee comb range. Lander Cem., in westernmost Christian Co., holds four combs, three made of calcarenite (limestone of sand-sized grains) composed of fossil bits and ooids, and one made of white marble. The talented stonecutter who erected these combs did excellent work in dressing and shaping the side slabs, most of which, if not all, feature beveled upper edges so that they meet neatly at the crest of the comb. Unfortunately, these combs are not in good condition, for two reasons: the calcarenite tends to disaggregate and split apart as it weathers; furthermore, it appears that no gable stones were used to support the side slabs, leading to a tendency for the side slabs to spread and collapse with time. Three of these combs are dated: 1857, 1860 and 1860. Sinking Fork Cem., located a little east of Lander, holds a single comb, of the same type of stone and style as the Lander combs, and probably by the same stonecutter.

Just a few miles to the west, across the line in Trigg Co., lies Wall Cem., which has a single comb, made in the same style as the Lander and Sinking Fork combs, of the same calcarenite, and likely by the same stonecutter. In addition to deterioration of the comb by weathering, its headstone has recently (since 2012) been broken into several pieces, possibly by vandals. This comb is especially interesting due to its date, 1847. It has been demonstrated earlier in this paper that the practice of erecting combs over graves spread very rapidly throughout the main Tennessee comb range; this comb in Wall Cemetery shows that the practice did not require many years to be carried beyond its main range.

In West Virginia, combs constructed with roofing metal have been documented in two cemeteries in northern Mingo Co., not far from the Kentucky border. Newsome Ridge Cem., sited on a high and remote ridge, is a graveyard in which traditional grave customs are still strongly maintained. Five relatively recent combs made of corrugated roofing “tin” are found here. Four bear dates: 1937, 1947, 1948, and 1952, long post-dating the majority of the Tennessee combs. In addition to these five metal combs, nine gravehouses—including a new one in the process of being erected in Sept. 2014—were noted. Annually, on the third Sunday in August, local folk hold a cemetery decoration day that is attended by hundreds. This meeting is called the “Reunion” or “Basket Meeting” because participants bring baskets of food to be shared in a communal “dinner on the ground”, laid out on a long buffet table in the cemetery’s pavilion25.

In nearby Rose Town, at the Rose and Brewer Cem., a metal comb lies somewhat downhill and just outside of the graveyard proper. It is also 90 degrees out of alignment with the other graves (which run east-west, facing east). The corrugated “tin” comb is intact, but has been removed from the grave to which it once belonged, and discarded, possibly to make mowing easier. A local informant stated that there were formerly at least two combs in this cemetery.

No connection between the West Virginia combs and the Tennessee combs has been established. However, the use of “Tennessee” as a woman’s given name in another nearby Newsome family cemetery hints that a connection may exist.

Two Texas cemeteries are known to feature comb graves. Shiloh Cem., in Denton Co., contains an elegant marble comb (Plate 27), the ends of which are closed by large, matching head- and footstones much in the same manner as the truncated triangle combs of Overton Co., Tennessee. The name on this comb is Nancy Yeats and the date of death is 1910. How this lone comb came to be erected here is not known, but several possible connections with Tennessee exist. Although Nancy Yeats and her husband Dr. T. A. Ball came to Denton Co. from Missouri, Nancy was born in Tennessee; it is not impossible that Yeats might be the same surname spelled Yates in several comb cemeteries in Tennessee (e.g., Cold Hollow and Gravel Hill cemeteries in the Bald Knob quad, Van Buren Co.). Furthermore, the county and town of Denton were named for John B. Denton, a Methodist-Episcopal preacher and Indian fighter who was born in White Co., Tennessee in 1806. However, Denton was orphaned at age eight and moved with his adopted family to Arkansas around the same time the comb grave style was nascent in Tennessee. Hence it seems unlikely that Denton took any knowledge of comb graves with him when he left his natal state. In Arkansas he became a preacher, and in this profession would certainly have become familiar with burial customs, including, perhaps, comb graves. Yet if Denton brought the idea of comb graves from Arkansas to Texas, there should be other combs in this area, for Denton died in 1841, long before Nancy Yeats arrived around 1865. Other Dentons from White county emigrated to Texas around 187026 and the role of Tennesseans in Texas history is well known. It seems likely that the Nancy Yeats comb has Tennessee origins, whether through Yeats herself, John B. Denton or someone else.

Plate 27: Nancy Yeats’ 1910 marble comb, Shiloh Cem., Denton Co., Texas. (Image courtesy of Jim Fritze.)

The Nancy Yeats comb was the only Texas comb known to Jordan, a long-time researcher of traditional graveyards in Texas and the South27. Nonetheless, two sheet metal combs have been discovered in Walling Cem. in Cooke Co., one of which is a matrimonial comb covering the graves of a couple originally from Tennessee28. The dates of death are 1916 and 1917.

Mississippi has at least one cemetery that features comb graves. Pickle Cem., in Monroe Co., has four sheet metal combs ranging in age from 1946 to 199129. Thornton30 describes these combs as grave shelters and adds that he remembers others from his youth. It would not be surprising if other comb cemeteries still exist in Mississippi.

In Louisiana, one comb is found in the Old Bellwood Cem. (Natchitoches parish). It is a limestone comb for a child who died in 187831.

The lone North Carolina comb, dated 1834, in the City Cemetery of Raleigh (Wake Co.) has already been described.

In Missouri, two cemeteries are known to feature comb graves. Though located in Missouri, these two comb cemeteries lie not far north of the state line with Arkansas and are part of the same Ozark cultural area as northern Arkansas, in which comb graves were a part of the cultural heritage brought in by immigrants from Tennessee.

In the very SW corner of Missouri, in McDonald Co., within the town of Noel, just 3-4 miles north of the line with Arkansas, lies Noel Cem., which features two comb graves. Further to the east, in Ozark Co. and just a few miles north of the state line lies Souder Cem. with four uninscribed --no dates, no names-- combs, along with what may be a collapsed former comb. A letter to the editor of "The Ozarks Mountaineer" (Oct. 1988 issue) suggests that at least one comb grave exists in Jasper Co., Mo.

While never officially a part of the Confederacy, southern Missouri is culturally part of the "Upland South" (Jordan-Bychkov, 2003), and received many settlers from Tennessee in the 1850-1880 period.

Eastern Oklahoma also lies within the cultural region of the Upland South (Jordan-Bychkov, 2003) and comb graves arrived here prior to statehood. At least three comb graves, extant in 2016, are documented by photographs in Old Cache Cem., near Keota, in Haskell Co. The combs seen in the images are very typical combs, and could easily pass for combs in a Tennessee graveyard. The stone slabs have the appearance of being sandstone, though it is difficult to be certain from photographs. One of the combs covers the grave of a child named Sarah Jackson and bears a date of death that appears to be 1891 (or possibly 1881).

According to local lore, the combs in Old Cache Cem. cover Native American graves. While it is beyond the scope of this paper to ascertain whether or not this fervently held belief is correct, the fact that the original land for the cemetery was donated by the Choctaw Nation makes this local history more plausible than it might otherwise seem. Furthermore, Jackson is a common Choctaw surname. The Choctaws were one of the Five Tribes (often referred to as the “Five Civilized Tribes”) that had adopted Anglo names and ways early in the 19th century. Further research would be needed to demonstrate that the combs at Old Cache Cem. are Native American graves, but the possibility is intriguing32.

In Muskogee Co., about 30 miles west from Old Cache is the 1889 grave of Belle Starr, notorious as the "Outlaw Queen." Photographs of her gravesite, which may be seen on the internet, show that her grave is marked by a structure that looks like a box grave made of cut stones and topped by a stone slab comb. This monument is very like the transitional comb structures found in Holmes Cem. in Arkansas (and described later in this paper).

On the basis of the Old Cache combs and the Belle Starr grave, it seems likely that a search of eastern Oklahoma traditional cemeteries would turn up more comb graves.

This survey of combs outside of Tennessee, while admittedly incomplete, indicates that Tennessee has by far and away a greater number of comb cemeteries and combs than any other state. The oldest known combs are also found in Tennessee. With the exceptions of the single North Carolina comb, the ten Kentucky combs lying to the east of the Tennessee comb range, and the six combs in West Virginia, all the other extant non-Tennessee combs are found westward or southward of the main Tennessee comb range. Westward, of course, was the general direction of historic migration. From this distribution pattern of numbers of comb cemeteries and combs, dates, and apparent westward spread, it is reasonable to infer that the comb tradition originated in Tennessee.

It should be noted here that the dispersion of the comb grave tradition from Tennessee into other Southern states is but one facet of a larger and more complex cultural transfer. Jordan-Bychkov (2003) defined the Upland South not only by its geography, physiography and agricultural style, but by a specific material culture. He argues persuasively that the Upland South culture first came to full flower in central Tennessee in the early 19th-century and was carried to the rest of the region by the “Tennessee diaspora.” The suite of cultural features Jordan-Bychkov found diagnostic of the Upland South includes graveyards and mortuary customs, and in particular, gravehouses (which he called “gravesheds”). While he did not mention comb graves per se, he was definitely aware of combs (as indicated by his reference to Finch, 1982). Possibly he considered combs a subclass of gravehouse (as did Ball in his 1977 paper). In the light of this general spread of material culture and customs from Tennessee westward and southward, the findings presented in the present paper regarding the origin and spread of the comb grave tradition are not unexpected.

Questionable comb graves in Georgia:

At least two Georgia cemeteries in the towns of Fairburn and Campbellton (west of Atlanta) feature a variety of cement or concrete grave covers that range in form from flat slabs, to rounded mounds, to round-crested comb-like structures, to a single instance of a triangular prism form that has the basic shape of a comb (Plate 28).

Plate 28: Comb-like cement grave cover beside flat cement grave covers in Baptist Church Cem., Campbellton, GA. (Image courtesy of Steven A. Hurd.)

The graveyard at the Baptist Church in the old community of Campbellton is of particular interest as featuring not only the one prismatic grave, but also a rounded grave cover that has been breached by weathering to the extent that the internal structure can be seen. This grave cover is constructed of a brick core that was plastered over with a relatively thin layer of cement. It is not a comb. A pair of comb-like (but round-crested) grave covers photographed in this cemetery in 198233, but no longer extant, were raised on brick bases, and may or may not have had brick cores. In contrast to the cement/concrete combs found in Tennessee, the molding of these former comb-like grave covers did not include any attempt to mimic the side slabs or triangular gable stones found in comb graves. Likewise, the single prismatic cement grave covering shows no attempt at mimicking the side slabs or gable stones of a traditional comb grave. Whether or not this prismatic grave is built over a brick core is not known, but it seems likely in view of the other brick cored structures and the fact that the structure would be much easier to build with a solid core under the cement.

The popularity and variety of cement grave covers in the Fairburn and Campbellton cemeteries, the use of brick sub-structures, and the failure to mimic the side slabs or gable stones in those structures that were comb-like in form all point toward the conclusion34 that these grave covers were not built by people aware of the comb grave style. Those grave covers that were comb-like were at best “coincidental combs”, resulting from the independent use of a very basic geometric form in the construction of a cement grave cover.

However, there is yet another old cemetery in Georgia which may feature actual comb graves. In western Spalding Co. there is an abandoned graveyard known locally as “Old Salem Church Cem.” It is populated mainly by burials marked by uninscribed fieldstones, but also includes a pair of small graves covered by long slabs that run the length of each grave. Judging from images of the two graves35, the covering slabs consist of weakly foliated metamorphic rock. One grave clearly has a pair of slabs running its length. Images of the second grave suggest that it too may be covered by a pair of slabs, with one slab being partially buried, making it impossible to see if it runs the full length of the grave. There are no supporting gable stones evident, and if these stones were originally arranged to form combs, they have since largely collapsed. Based on the evidence currently in hand, these graves appear to have been covered by crudely built combs. An on-site examination of these burials is needed to clarify this point. Also to be borne in mind, as pointed out in the preceding discussion of the “comb-like” grave coverings in the Fairburn and Campbellton cemeteries, comb-shaped grave covers may sometimes be erected by persons who likely had no knowledge of comb graves per se and were not intentionally participating in that tradition.

Former comb graves on the Delmarva Peninsula:

The former existence of a small number of wooden comb graves has been documented in a handful of graveyards on the Delmarva Peninsula, specifically, in southernmost Delaware and adjacent Maryland, a region with strong historical cultural ties to the South. I am much indebted to Chris Slavens36 of Laurel, Delaware, for calling these now vanished combs to my attention and for the factual information in the following paragraphs. Any conclusions drawn from this information I must take responsibility for myself.

In Sussex County, just north of the state line with Maryland, lies Bethel M. E. Church where the original church building was erected in 1841. At Bethel the writers of Delaware: A Guide to the First State (a Federal Writers Project publication issued in 1938) found “several old graves with shingle roofs…. ….examples of the once-popular local custom of placing a small pitched roof close over a grave to keep off the rain”. According to Zebley (1947), “in the old graveyard a few roofed-over graves can be seen, one of the few places in Delaware where any of these graves remain. These roofs are A shaped with the gables closed in, rest directly on the ground with the entire frame-work covered with shingles.” Zebley provides a photograph37 showing two combs in Bethel cemetery; the construction of the comb in the foreground of Zebley’s photo clearly fits the definition of comb grave, i.e., a grave cover in the form of a gable roof set directly on the ground over the length of an in-ground burial. As illustrated by the varied materials used for the construction of the Tennessee combs, it is the form, not the materials used, by which a comb may be defined. The Delaware combs were naturally made of wood inasmuch as stone suitable for quarrying is scarce or non-existent in this region.

Zebley goes on to state that “in the private graveyard on the farm of Ira West….there are several graves over which roofs have been built…. The most recent of these were over the graves of John C. West who died in 1858 and Mahala West who died in 1852.” In addition, he mentions that there “is one roofed-over grave” at King’s M.E. Church, where the first church building was erected in 1842. A third author, Pepper (1976), writes that “an old graveyard in back of Millard Johnson’s home near Bayard had some graves with a roof on top of each one. The A-shaped roofs were made of cedar shingles pointed on top exactly like a house roof, and each one covered an entire grave.” A 1947 photograph shows a comb over an 1891 grave near Bayard; this comb, like the Bethel and West combs has not survived to the present.

Immediately south of Sussex Co. lies Wicomico Co., Maryland, where Jacob (1971) reported that “it was the custom in the eastern section of Wicomico for many years to build a roof over a new grave. The roof was built on an ‘A’ frame, the peak about thirty inches high, with the structure covering the entire grave.” According to Jacob, only one of these grave roofs remained in 1971.

It is notable that the sources quoted above referred to the grave covers as “roofs”. The term “comb grave” was not used by these writers, and it seems likely that the term was not used locally. The overall number and time range of the Delmarva combs is unknown at present. The only dates currently known for Delaware combs are 1852, 1858 and 1891. An intriguing question is how long could wooden combs have survived? Pepper (1976) mentions “cedar” shingles, but a shingle-making industry using both cedar and bald cypress --the latter a very durable, rot-resistant wood-- was formerly important in Sussex county, and combs constructed of cypress would indeed have been long-lived for wooden structures. Although a number of combs still existed in the late 1930s and 1940s, and at least one as late as 1971, so far as is known, no Delaware or Maryland combs have survived to the present, presumably due to the ultimate perishability of the wood of which they were constructed.

What little is known of the dates for the Delmarva combs suggests that they post-date the earliest Tennessee combs, and therefore were not the source for the idea of combs in Tennessee. Furthermore, while it is unknown how many Delmarva graveyards once featured combs, it seems clear from the small number reported that the custom never became very widespread or strongly entrenched here, and therefore was less likely to have been exported. On the other hand, by reviewing census reports Slavens has found that dozens of residents from the Delmarva Peninsula emigrated to Tennessee. For example, Thomas West, born 1760 in the Delmarva region settled in DeKalb Co., Tennessee in 1804. The main Tennessee comb range extends into DeKalb Co. The possibility of a cultural exchange involving mortuary customs—either from Delaware to Tennessee or a cultural backflow from Tennessee to Delaware--- cannot be ruled out by the information currently in hand. Alternatively, the Delaware combs may simply have resulted from an independent solution to the problem of covering and protecting the graves of loved ones, employing one of the simplest –both in geometric form and in construction-- types of grave covers. It is a pity that neither the combs nor custom survived on Delmarva to present times.

For a more complete discussion of the Delmarva roofed graves, see Slavens (2020).

Comb graves outside the South:

Another venue in which comb graves once existed and may yet exist is in Native American cemeteries where grave covers --commonly known as “Spirit Houses”-- have been built by various Native American groups. Spirit Houses have been found in a number of states, mostly outside the South. A collection of early 20th century picture postcards belonging to John Waggoner, Jr.38 shows Native American graveyards in Michigan, Wisconsin, Minnesota, Iowa, Oklahoma, Alaska, British Columbia, Ontario and the Yukon Territory. In addition, Ball (1977) notes that Native American gravehouses have been documented in Louisiana and Washington State. The Spirit Houses pictured in Waggoner’s collection appear to be constructed of wood or wood with sheet metal. Among the many grave covers are a few that are simple gable roofs resting directly on the ground. They are in essence combs, though probably not by intent. No connection is known to exist between these Spirit House combs and the Tennessee comb grave tradition.

Noteworthy comb variants:

With approximately 5000 extant combs known in eleven states, produced by different local artisans, over nearly two centuries, it would be remarkable if there were no significant and interesting variations in the comb form. The aforementioned “truncated triangle” style of comb (Plates 8 and 9) was used in Overton Co. for over a century39, especially in the Allred area. It also occurs in Putnam Co., in lesser numbers. The following paragraphs will highlight some other noteworthy comb variants.

Concurrently with the “truncated triangle” combs, the Vaughn family of quarrymen also produced box graves consisting of four large sandstone slabs, held upright against each other by long iron cross bolts, and topped by an even larger flat-lying sandstone slab. The distinctive feature of these massive boxes was the necked discoid headstone that was integral with, and an extension to, the upright slab at the head end of the grave. Judging from two graves in Beaty Cem. (Moodyville quad, Pickett Co.), a few discerning comb customers wanted the best of both styles: the “truncated triangle” plus a necked discoid (Plate 29).

Plate 29: Truncated triangle comb with integral necked discoid headstone (1854), Beaty Cem., Moodyville quad, Pickett Co.

Remote Davis Cemetery, deep in the heart of Scotts Gulf (Lonewood quad, White Co.) hosts a unique mushroom-shaped headstone (Plate 30). This family graveyard also features several comb graves with inscriptions chiseled in cursive, something rarely seen and surely much more difficult than block lettering. The scrawled inscriptions likely mirror in no small degree the handwriting abilities of the family member who wrote the inscription for the stonecutter, or perhaps the stonecutter himself-- possibly one and the same person.

Plate 30: Unique mushroom shaped headstone (1873) with cursive inscription, Davis Cem., Lonewood quad, White Co.

Although gravehouses not infrequently are constructed to cover multiple graves, with comb graves the rule is one grave per comb. Two exceptions are known. In Pine Hill Cem. (Cookeville West quad, Jackson Co.) a somewhat crudely constructed matrimonial comb covers the graves of John and Elizabeth Cummins (Plate 31), who died in 1868 and 1864 respectively (Ball, Waggoner and Finch, 2016). And in Walling Cem. (Cooke Co., Texas), a broad sheet metal comb covers the graves of a Squire Umsted (1916) and his wife Mary Adams (1917).

Plate 31: Matrimonial comb for John and Elizabeth Cummins, with marble headstones, Pine Hill Cem., Cookeville West quad, Jackson Co.

Alabama cemeteries contain a number of unusual combs, with Old Liberty Hill Cem. (Walker Co.) being a special treasure trove of oddities. Several combs here feature bas relief geometric designs on the side slabs. One of these (Plate 32) includes the deceased’s three initials, J D E, in the centers of the three design panels. The side slabs of this comb rest on a sand or earth mound that apparently fills the interior of the comb and gives support to the side slabs. Several other combs in this cemetery and in other Alabama cemeteries were seen to be fill-supported40, but triangle gable stones remain the rule.

Plate 32: Sandstone comb, side slabs supported by fill and decorated with bas relief geometric designs containing the initials, J D E, of the deceased. Old Liberty Hill Cem., Walker Co., AL.

A single comb in Old Liberty Hill Cem. with a “jig-saw puzzle” comb support system was photographed in 1982 (Plate 33). In this comb, the end stones are larger than normal gable stones, and notched in such a way that the modified lower corners of the side slabs fit into the notches to form an interlocking, self-supporting structure. Sadly, this unusual comb was not found when this cemetery was re-visited in 2012. However, as of 2019, three combs built with this interlocking type of structure could still be seen at Madison Cem., in Fayette Co.

Plate 33: Unusual sandstone comb with notched end stone interlocking with modified side slabs to make a self-supporting structure. Photo taken 1982. Old Liberty Hill Cem., Walker Co., AL.

As described earlier, a few Tennessee graveyards feature combs constructed of poured concrete. An even more unusual concrete comb is that of Mary Sparks (1840-1932) found in Steadman Cem. (Walker Co.). Rather than being poured in place, this comb consists of individual "stones" --side slabs, gable stones, head- and footstones-- all poured separately, allowed to set, and then assembled over the grave in just the same manner as a normal comb made of sandstone slabs (Plate 34). A delightful touch is the dog footprint left in one side slab by some errant pup prior to the setting of the concrete.

Plate 34: Comb grave for Mary Sparks, made of poured concrete "stones"; note track left by dog in side slab while the concrete was still soft. Steadman Cem., Walker Co., AL. The adjacent comb, covering Mary's husband, is constructed of traditional sandstone slabs.

Also in Alabama are perhaps the most unusual combs seen during this study: two rather skeletal-looking combs covering the graves of Moses Barton (1890) and his wife, Margaret (1898) in Sardis Cem. (Walker Co.). This pair of combs is constructed of sections of ornate iron railing or fencing (Plate 35), such as might normally be found enclosing a family plot. But here they have been deliberately arranged to form symbolic, if not practically functional, combs.

Plate 35: Symbolic combs made of ornate iron fencing, Sardis Cem., Walker Co., AL (Photo by Janie C. Finch.)

In Arkansas, two of the eight combs in Holmes Cem. (Boone Co.), sit atop stone boxes (Plate 36). In both cases, the comb and box are of finely dressed limestone and evidently built as a unit. These structures are, essentially, composites of comb graves and box graves. One might also call them gravehouses, but the presence in the same cemetery of six normal --i.e., directly on the ground-- combs, of similar limestone construction suggests that these two graves should be deemed unusual comb variants.

Plate 36: Combination box – comb grave, Holmes Cem., Boone Co., AR.

Finally, to return to Tennessee, a concrete or cement comb in Curlee Cem. (Readyville quad, Cannon Co.) seems to begin to blur the lines between comb and gravehouse. The key difference between a comb grave and a gravehouse is that the gravehouse has a roof that sits atop walls, or at least corner posts, whereas the comb has no walls, but rather is a gable roof set directly on the ground or (rarely) on a very low base. This comb at Curlee Cem. (Plate 37) features low but distinct walls. While more of a comb than a gravehouse, it is something of a hybrid, which serves to emphasize that the basic purpose of both types of grave covers is the same.

Plate 37: Cement or concrete comb grave with low supporting walls, Curlee Cem., Readyville quad, Cannon Co.

Transitional grave covers:

As noted above, the cement comb at Curlee Cem. has very low supporting walls, a feature more appropriate to a gravehouse than a comb grave. Two concrete combs at Rock Springs Cem. (Buffalo Valley quad, Putnam Co.) (Plate 38) also deviate from the normal comb pattern with each being set on a concrete base and having a flattened apex rather than a sharp ridgeline. Nonetheless, they have been counted as cement combs in this survey, inasmuch as they are more like combs than unlike. Note that the molding mimics the side slabs and triangular end stones.

Plate 38: Concrete combs showing non-standard features: concrete bases and flattened ridge crests; Rock Springs Cem., Buffalo Valley quad, Putnam Co.

On the other hand, grave covers such as several graves covered with mounded and sculpted cement in Reed Cem. (Oneida South quad, Scott Co.) (Plate 39) clearly are not combs, though they serve the same purposes and probably were erected for the same reasons of protecting the graves. Nor are the stone structures at Clear Creek Cem. (Lancing quad, Morgan Co.) (Plate 40) combs, although they have sloping side slabs of sandstone. Observations made during this survey show that there are transitional forms between combs and gravehouses, combs and box graves, and a variety of unusual grave covers that may have some attributes in common with comb graves. Nonetheless, comb graves represent a distinctive style of grave cover, normally featuring two planar surfaces inclined to meet at a sharp crest like a gable roof over the grave, and normally set directly on the ground rather than on foundations or walls.

Plate 39: Grave covers made of mounded and shaped cement in Reed Cem., Oneida South quad, Scott Co.

Plate 40: Grave with mixed box and comb grave characteristics in Clear Creek Cem., Lancing quad, Morgan Co. Original slab top cover has been broken.

An additional class of comb grave:

Near the woods along the back edge of an Overton Co. comb cemetery lies a very special comb. It is small –just 33 inches long—and uninscribed except for the name “Zip” neatly engraved on one gable stone. It is the grave of a much beloved dog that died in Dec. 2012. Zip’s owner is an admirer of the comb grave tradition and chose to memorialize Zip in this manner (Plate 41).

Zip’s comb is made of sandstone slabs from a quarry in Rhea Co. The stones were cut, Zip’s name engraved, and the comb erected by a close friend of Zip’s master. Inasmuch as Zip’s comb was erected in 2013, the earlier statement regarding the Kathleen Rodger’s 2012 comb as being the most recent comb must be revised as the most recent comb for a human being.

Plate 41: A sandstone comb for Zip, a beloved dog, Overton Co.

A second comb for a pet exists in Sumner Co., well to the west of the main comb range. This comb was erected in 2012 by an historian with an interest in Tennessee’s cultural history, in memory of his cat, Stubby. He explained, “I thought Sumner County needed one, even though the form was historically expressed further east, on the Highland Rim. I bought two longer than wide slabs of sandstone from Lowe's or Home Depot. Not Hartselle Formation, I know, but that is not easily available in Sumner County.”41

A third pet comb exists in a backyard in Monterey (Putnam Co.), where it was constructed several years ago over the grave of a pet dog. The gentleman who erected it formerly worked in kitchen counter installation and had some spare slabs of stone. Although he was unfamiliar with the term “comb grave,” he most likely was inspired by combs seen in local cemeteries.

While combs over pet graves are a rarity, they are nonetheless legitimate combs. They are erected out of affection for and in tribute to the departed friends, just as regular combs have been erected.

Loss of combs:

Both Cantrell (1981) and Ball (1997) comment on the destruction and disappearance of comb graves. For the present survey, some 45 cemeteries were noted that once had combs, but which no longer do (see Appendix H). In many cases the presence of a former comb is recorded by a pair of triangular gable stones left behind when the side slabs were removed. In the case of 19 former comb cemeteries, the combs were present when this study was initiated in the mid-1970s, but not when revisited in later years.

The reasons for the loss of combs are numerous. Cantrell (1981) notes that some cemetery committee members feel combs do not fit in well with “well-ordered rows” of modern headstones, that they are a “hindrance to the proper upkeep of a cemetery”, that they are “snake harbors,” and that they “are being removed to make graveyard cleaning easier.” In the mid-1960s the trustees of Dodson Chapel cemetery decided to have all the comb grave covers removed (along with a couple of box graves) to "modernize" the cemetery. Ball (1997) comments “even markers made of stone are not impervious to the long term deteriorating effects of wind, rain, and freezing weather or destruction by unthinking (or uncaring) individuals bent on vandalism.”

During the present study the following reasons for damage to or destruction of combs were catalogued:

Plate 42: Bench and patio beside Walnut Grove Cem. (Livingston quad, Overton Co.) which no longer has any extant comb graves.

Natural weathering processes may rapidly destroy combs made of inferior stone, such as Chattanooga shale, and obliterate inscriptions on limestone grave markers in little over a century, but combs made of well-indurated quartz sandstone, such as the Hartselle sandstone, should last for many centuries. The relatively rapid loss of combs in the latter half of the 20th century is largely due to misguided human activities.

To respect the wishes of both the departed and of their loved ones who memorialized them with combs, we should preserve the remaining combs. The combs provide an interesting glimpse of a nearly bygone funerary custom, one that apparently is original with Tennesseans. In the words of Lynwood Montell, a well known folklorist, “these traditional prismatic grave structures more than any other single cultural feature lend uniqueness to the folk cemeteries in the Upper Cumberland” (Montell, 1993). The comb graves of Tennessee, and other states as well, constitute an interesting and valuable part of our material cultural heritage and should be cherished.


A comb grave is an interment over which a protective cover having the form of a gable roof, resting directly on the grave (not raised above it on walls or corner posts), has been erected. Most combs are made of sandstone slabs, with other materials being used to a much lesser degree: limestone, marble, cement or concrete, sheet metal, foliated granite, and even shale. In Tennessee, sandstone slabs quarried from the Hartselle Formation are the traditionally preferred comb-building material.

Tennessee combs are found in a geographic belt that parallels the western escarpment of the Cumberland Plateau from Winchester northward to just across the state line into Kentucky. The majority of combs are found in small cemeteries on the Eastern Highland Rim, but combs are also found on the Hartselle Bench and on the western portions of the Cumberland Plateau itself. This geographic distribution strongly correlates with the outcrop trace of the Hartselle Formation along the western flank of the Plateau and in erosional outliers of the Plateau. Evidently, the presence of appropriate sandstone in the Hartselle Formation exerted a major control on the use of combs to cover graves.

Within the main comb range there are two notable local concentrations of comb graves in the Caney Fork Valley (White and northern Van Buren counties) and in Overton Co.

Although admittedly incomplete, the data presented in this paper, augmented with anecdotal information, point toward the conclusion that the style of covered graves known as comb graves originated in Tennessee, in the White Co. area, in the 1815-1820 period (possibly slightly earlier, certainly no later). From Tennessee, the comb style and custom was carried, mainly westward and southward, into other states, including Alabama, Arkansas, Louisiana, Mississippi, Texas, Missouri and Oklahoma, and also into Kentucky, West Virginia and North Carolina. The relationship, if any, between the Tennessee comb tradition and the former wooden combs of the Delmarva Peninsula remains at present enigmatic.

Historic picture postcards show that at least a few Native American Spirit Houses were, by form and construction, combs; however, it seems probable that these grave covers were unrelated to the Tennessee comb grave tradition.

More definitively documented by the data, the comb style spread rapidly throughout the full length of its main Tennessee range by the late 1830s. The style achieved its greatest popularity during the latter half of the 19th century and was still commonly used, especially in Overton Co., up through the 1930s. By the 1950s the practice of erecting combs over graves had nearly been abandoned, with a few rare exceptions. Nonetheless, the custom has survived into the 21st century, the most recent two combs for human interments being erected in 2001 and 2012.

Combs serve the same functional purposes as any other form of grave cover, primarily to offer some protection to the grave from the elements and, to some extent, from animals. The erection of a comb over the grave of a loved one no doubt also helped fulfill emotional needs, showing that the deceased was cared for. Combs also serve to make graves more permanent, less likely to disappear when graveyards cease to be kept up.

During the course of this study comb graves were documented in 520 Tennessee cemeteries; however, the actual number of comb cemeteries is certainly higher. As of 2023 there were 3668 extant comb graves known in Tennessee, far more than any other state. Unfortunately, combs are being dismantled or destroyed for a variety of reasons, none of them very good. At least 45 former comb cemeteries are known within the main Tennessee comb grave range; of these, 19 have lost their combs since this survey was initiated in the mid-1970s.

In recent years, combs have been erected over at least three pet graves.

Comb graves, an original Tennessee folk custom, represent an unusual and charming part of Tennessee’s material cultural heritage. Comb graves should be appreciated and conserved, yea even cherished.

Concluding musings:

In Overton county, on the south flank of Alpine Mtn., lying on the Hartselle Bench at a spot reached via several miles of very rough 4WD road, is an old house site, the house now entirely disappeared except for some ornamental plantings and a pile of stones that may have been a chimney. A keen eye is needed to recognize this as the erstwhile location of a home. In contrast, a hiker passing by the right spot in the woods a few yards to the north of the house site could not fail to notice a lone comb grave. Made of Hartselle sandstone, beautifully inscribed, and almost certainly a Vaughn family product from their nearby quarry, this comb is in nearly perfect condition. The inscription reads “Daughter of W. T. & Z. D. Gore Born Mar. 17. 1902. Died The Same. At Rest.” Although the infant may not have lived long enough to receive a name, her parents nonetheless cared sufficiently to provide her with a lasting memorial.

Nameless, forgotten, her home long abandoned and now almost entirely obliterated, the short existence of the Gore infant is still memorialized 121 years after her death. Indeed, the fine-grained, quartz-cemented Hartselle sandstone will easily endure long after marble has weathered to illegibility; the sandstone will likely outlast granite. Barring its destruction by a falling tree or vandals, this comb should easily attest to her memory for several centuries more. Hikers wandering through the woods will readily see her grave long after the house site is completely unrecognizable. This solitary comb in the wilderness bears witness to the remarkable permanence of the Hartselle sandstone comb graves.



1OED definition 6e further states that this usage is dialect. states that this architectural definition of comb is primarily a Midland and Southern usage. OED cites an 1870 instance of its use by Mark Twain. Many Tennessee combs predate this cited use, some by more than 50 years, and in all likelihood they have always been known as combs.

2Rector Cemetery is transected by the Kentucky-Tennessee border, and whether this comb lies in Tennessee or Kentucky or is actually bisected by the line has not been determined.

3Nearly 50 cemeteries could not be located. Around 11 seem to have been moved: possibly as many as seven for the impoundment of Cordell Hull reservoir, two for I-24, one for mining activities near Elmwood, and one for the building of a home. Three apparently were bulldozed into oblivion; one or two likely were plowed under; and 30+ others have simply disappeared through time, weathering, erosion and neglect.

4When sandstone slabs are pried out of the quarry, they tend naturally to break across the bedding at right angles. To cut a beveled edge would likely be difficult in the thin beds preferred for comb slabs, though an example can be seen in Beaty Cem., Moodyville quad, Pickett Co.

5White (2002), who studied a variety of types of grave covers --including combs-- concluded that the "period of use, [sic] for comb covers seems to be from about 1840 to 1900." To draw this conclusion, he had to exclude from his data set the dates of the bolted combs, which he referred to as "a newer version of comb covers". Thus he called them combs, and mentioned their having "death dates well into the 1900s" and "from 1900 to 1950", yet excluded them from consideration in determining the temporal range for the use of comb covers. Cantrell (1981) included these bolted combs in his set of comb grave dates, a fact White mentioned while failing to explain his own reason(s) for excluding them. He stated that this "newer version" was centered on Falling Spring [sic] Cemetery and "spread to only a few other cemeteries". In truth, White's sample was too small to justify these conclusions. He documented various grave covers in 288 cemeteries (many outside the known comb grave range); how many of these 288 cemeteries contained comb covers he did not report, but he did note a total of 311 comb graves. In actuality, the bolted combs are widespread in Overton Co. and are found as far west as central Putnam Co. White's writings on this particular subject are confusing, but it appears that he does not believe the "newer version" of comb covers are legitimate comb graves. While they involve taller than usual head- and footstones, the bolted combs have the same basic shape as all combs; they are made of the Hartselle sandstone as are most combs; and they serve the same purpose(s) as other combs. The principal difference is the iron rod support system in place of gable stones. The bolted combs represent a local variant in comb construction style, developed at a later date than the early combs in the White Co. area, but they are indeed legitimate comb graves, and their time range is part of the over all comb grave time range.

6A few miles to the south, Christian Cem. (Silver Point quad, Putnam Co.) features another grave covered with Chattanooga Shale slabs. The head half of this grave is covered by two slabs leaned together like a comb, but the foot half at present appears to have only a single flat-lying slab. It is unclear whether or not this grave cover was originally constructed as a comb.

7Both these 5-Vee roofing metal combs were constructed by Clio Webb at the request of his sister, Myrtle Webb Judd. The first was constructed and emplaced over the grave of Myrtle’s husband, Joe Judd, who died in 1983. When interviewed on this subject in 2012, Clio stated that Myrtle wanted a comb “to keep the grave dry”. The metal comb constructed by Clio closely matched the metal combs that once covered Joe’s parents (which Clio did not construct) nearby in the same cemetery (but which have since been removed). When Myrtle died in 2001, Clio constructed a second comb, matching that of Joe’s, again in accordance with his sister’s wishes. This was the first comb to be constructed in the 21st century, (and one of only two known thus far).

As of 2012, Clio was still maintaining these two combs when they needed repair; a few years later, as his health declined, Clio apparently was not able to continue to look after them. Clio died in Feb. 2021, and as of Dec. 2022 both combs had deteriorated, with Myrtle’s comb being completely collapsed. No doubt the remains of these combs will be removed from Bear Creek Cem. in the not-too-distant future, and the number of 21st century comb graves will be reduced to a single example.

8The stratigraphic generalizations stated in this paper are based on data from 7 ½ minute geologic quadrangle maps published by the Division of Geology, Tennessee Dept. of Environment and Conservation.

9Kathy Stockton Williams, personal communication, January 2014.

10Both the author of this paper and Michael McNerney originally thought this stone was cut at a date later than 1817. McNerney, who has studied the evolution and spread of the necked discoid style, estimated that this headstone was carved sometime within the years 1825-1835. (McNerney, personal communications, Mar. 2013 and Jan. 2014). We now doubt this earlier interpretation.

11Cliff Owens, personal communication, Apr. 2014.

12Cantrell, 1981, reported that in the 1930’s a comb set cost $30, whereas a box set cost $60.

13Emma Herbert-Davies, University of Leeds, sent a description and images of the Otley comb; personal communications, Jan. 2016.

14Dr. Ray Hutchison, Prof. of Urban and Regional Studies, Univ. of Wisconsin, Green Bay, alerted me to the existence of these grave structures, personal communication, Jan. 2016.

15This time relationship was first called to my attention by graveyard enthusiast Dr. Ray Hutchison.

16Donald B. Ball, personal communication, Mar. 2014.

17Dr. Ray Hutchison, personal communication, May 2021.

18Dr. Ray Hutchison, personal communication, May 2021.


20James Ezell, P.E., personal communications, May 2016.

21Dr. Eugene M. Wilson, Prof. of Geography, Univ. of South Alabama, suggested that sandstone from the Pottsville Formation of Pennsylvanian age might also be a source of comb slabs. Personal communications, Oct. 1981, Apr. 2014.

22James Ezell, P.E., personal communications, May 2016.

23John Waggoner, Jr., personal communication, Mar. 2014.

24Michael McNerney, personal communications, May 2013.

25Dr. Alan Jabbour, personal communications, Sept. 2014 and Apr. 2015.

26Beth Moore Farmer (a Denton descendant), personal communication, Feb. 2014.

27The late Dr. Terry G. Jordan, personal communication, Aug. 1981.

28John Waggoner, Jr., personal communication, Nov. 2014.

29John Waggoner, Jr., personal communication, Feb. 2014.

30Terry Thornton,

31John Waggoner, Jr., personal communication, Jan. 2014.

32My thanks to Dagmar Anne Cole of Ft. Worth, TX, for calling the combs in Old Cache Cemetery to my attention, and to Jerry Smith of Cartersville, OK, for providing images.

33Marion O. Smith, personal communication, Apr. 2014.

34I am indebted to Steven A. Hurd, Ray Hutchison and John Waggoner, Jr. for insightful comments regarding these debatable comb graves in the Campbellton Baptist Cemetery.

35I thank Dennis and Mike Scott of the organization Monumental Preservation for informing me of these two graves, providing images, and sharing their conclusions regarding the significance of these graves.

36Chris Slavens, personal communications, Sept. 2016; also Slavens (2017, 2020).

37Zebley (1947), p. 349; see also the Zebley Collection Churches of Delaware, Delaware Public Archives for a better image.

38I am grateful to John Waggoner, Jr. for making his collection of postcards picturing Spirit Houses available to me.

39In the Allred area a truncated triangle comb covers an 1847 grave in Shiloh Cem., several dated in the 1850s are known, and a 1963 date occurs on a truncated triangle comb in Cub Cem. (all in Crawford quad, Overton Co.).

40Scott Cem. (Walker Co., Alabama) contained 51 combs when visited in May 2019. Many, perhaps the majority, are earth-filled combs. Many are also in a state of partial to complete collapse. Although no data were collected to document the % of the total that are earth-filled, nor any data collected to document a positive correlation between earth-filled and collapsed combs, general observation and photography left a strong impression that earth-fill leads to collapse. Many of the collapsed combs have clearly spread laterally, the side slabs moving away from and at right angles to the center line of the comb. As they spread apart, the slabs rotate to angles much flatter (lower) than when the comb was originally set up, with a gap between the upper edges of the slabs that commonly exceeds a foot. It seems likely that the slabs have actually been pushed apart by expansion of the earth fill during many years of freeze/thaw cycles. If this interpretation is correct, then the earth-filled combs contain within themselves the seeds of their own destruction.

41Dr. Joseph C. Douglas, personal communication, Mar. 2014.

References Cited:

Ball, Donald B., 1977, Observations on the Form and Function of Middle Tennessee Gravehouses: Tennessee Anthropologist, v. II, n. 1, Spring 1977, p. 29-61.

Ball, Donald B., 1997, Types of Early Grave Decoration in Middle Tennessee: Tennessee Folklore Society Bulletin, v. LVIII, n. 3, p. 117-127.

Ball, Donald B., 1999, Comments on the Distribution and Chronology of Comb Graves in the Upland South: Ohio Valley Historical Archaeology, v. 14, p. 58-66.

Ball, Donald B., 2005, Further Observations on Gravehouse Origins in the Upland South: Tennessee Folklore Society Bulletin, v. LXI, n. 2, p. 17-30.

Ball, Donald B., 2008, An Alternate Hypothesis on the Origin of Upland South Gravehouses: Ohio Valley Historical Archaeology, v. 23, p. 101-118.

Ball, Donald B., John C. Waggoner and Richard C. Finch, 2016, Note on an Unusual Double Comb Grave in Jackson County, Tennessee: Ohio Valley Historical Archaeology, v. 26, p. 142-144.

Brown, Ian W., 2004, Some Grave Houses in West-central Alabama: AGS Quarterly, Bulletin of the Association for Gravestone Studies, v. 28, n. 2, p. 14-15.

Burnett, Abby. 2014. Gone to the Grave: Burial Customs of the Arkansas Ozarks, 1850-1950: The University Press of Mississippi, 327 p.

Cantrell, Brent, 1981, Traditional Grave Structures on the Eastern Highland Rim: Tennessee Folklore Society Bulletin, v. XLVII, n. 3, p. 93-103.

Crissman, James K., 1994, Death and Dying in Central Appalachia: Urbana, University of Illinois Press, 247 p.

Federal Writers’ Project, 1938, Delaware: A Guide to the First State, Compiled and Written by the Federal Writers’ Project of the Works Progress Administration: The Viking Press, New York, 549 p.

Finch, Richard C., 1982, Unique Grave Houses in Tennessee: Monterey, TN, The Standing Stone Press, 4th year, n. 4, Spring 1982, p. 1-3.

Finch, Richard C., 2004, Ashes to Ashes, in Birdwell, Michael E. and Dickinson, W. Calvin, 2004, Rural Life and Culture in the Upper Cumberland: Lexington, KY, The University Press of Kentucky, p. 66-72

Finch, Richard C., 2013, Tennessee Comb Grave Tradition Survives into the 21st Century: Tennessee Folklore Society Bulletin, v. LXIX, n. 1 (Spring 2013), p.17-24.

Jacob, John E., 1971, Graveyards and Gravestones of Wicomico: Salisbury, MD, Salisbury Advertiser, 131 p.

Jordan-Bychkov, Terry G., 2003, The Upland South: Santa Fe, NM, Center for American Places, 121 p.

McNerney, Michael J., 2017, A Shape in Time and Space: The Migration of the Necked Discoid Gravemarker: Tuscaloosa, AL, Borgo Publishing, 246 p.

Montell, William Lynwood, 1993, Upper Cumberland Country: The University Press of Mississippi, Print-on-Demand Edition, 187 p.

Oxford English Dictionary, The Compact Edition, Vol. 1 A-O, 1971: New York, Oxford University Press, p. 472.

Pepper, Dorothy W., 1976, Folklore of Sussex County, Delaware: Sussex County Bicentennial Committee, (no publisher given), 116 p.

Pollard, Geraldine Elrod, ed., 2003, The Fred Clark Book of Cemeteries of White County, Tennessee, v I and II: Published by The White County Genealogical – Historical Society, 564 p. and 524 p.

Richmond, Michael D., 1998, An Archeological Survey of the Proposed Lake Cumberland Debris Management System Near the Community of Redbird in Whitley County, Kentucky: Lexington, KY, Contract Publications Series 98-09 prepared by Cultural Resource Analysists, Inc., 28 p.

Slavens, Chris, 2017, The Roofed Graves of Delmarva: Laurel Historical Society Newsletter, Winter 2017, p. 4.

Slavens, Christopher, 2020, The Roofed Graves of Delmarva: Laurel, Delaware, Bald Cypress Books, 91 p.

White, Vernon, 2002, Grave Covers: Our Cultural Heritage: Berea, KY, KANA Imprints, 119 p.

Zebley, Frank R., 1947, The Churches of Delaware: (no publisher given), 363 pages.

Special thanks to Janie C. Finch for accompanying me on many of my cemetery survey outings and for her superb proofreading and editing abilities which have greatly improved this manuscript.

Any reader of this paper who knows of the existence of a Tennessee comb cemetery not listed in Appendix A1 or A2 is urged to send this information to the author at


Appendix A-1 Comb Cemetery Inventory by Quadrangle Map.

Appendix A-2 Comb Cemetery Inventory by County.

Appendix B Limestone Comb Graves.

Appendix C Marble and Shale Comb Graves.

Appendix D Concrete and Concrete Comb Graves.

Appendix E Sheet Metal Comb Graves.

Appendix F Early Comb Grave Dates.

Appendix G Comb Grave Church Associations.

Appendix H Former Comb Cemetery Inventory.