Serrated Edges in Rock Art

Certain rock art images prove to occur at numerous places in the world, like cupules and zigzags. In most cases this is a matter of parallel invention. However, a number of motifs may have travelled across the globe for short or even enormous distances. This study investigates the distribution of one of the enigmatic rock art features, the serrated edge and explores the possibility that this practice diffused from North America to South America (or vice versa).

By Maarten van Hoek



Serrated Edges in Rock Art – Diffusion or Parallel Invention?

Maarten van Hoek



In various publications I investigated the possibility that certain rock art phenomena diffused across the globe. In particular I focussed on motifs that may have diffused from South America to North America, or vice versa (Van Hoek 2018a; 2018b; 2018c). In most cases diffusion is only a remote possibility, although some rock art motifs may indeed have travelled over enormous distances. On the other hand, although occurring in two or more distant areas, certain motifs or customs (involving – for instance – rock art obliterated by grinding practices; Van Hoek 2018), are purely local inventions without having any cultural relationship. This paper deals with petroglyph rocks (boulders and outcrops) that have serrated edges that definitely are anthropic and intentional, although their function or meaning is still completely obscure. Remarkably, there are only very few instances (known to me) where rock edges have been serrated, and surprisingly they occur in both South America and in North America at sites that are roughly 7800 km apart (Figure 1).

In general the edges of boulders and outcrops are neglected during rock art manufacturing. The images and markings are generally placed on the surface of the rock without intentionally marking the edges. Of course images often touch the edge or are folded across the rock edge, like the mask-petroglyphs in New Mexico, USA (Van Hoek 2020). Although in those New Mexico cases the edge of the stone was intentionally used (i.e. incorporated into the mask with a special reason), it was not done to specifically mark the edge of the stone. In contrast, serrated edges intentionally mark and thus emphasise the edge of the stone. They may be considered to represent 3D-markings as there are always two 2D-panels involved.


South America

In South America only a very few instances where the rock’s edge has intentionally been marked are known to me. For instance, at Punta Picata, an extensive rock art site directly along the Pacific coast in southern Peru, a row of eight cupules intimately follows the edge of the rock. However, the edge was not serrated by this row of cupules (Umire Álvarez 2012: Fig. 6). As far as I know this is the only example at this site and thus it certainly is not even a (small scale) rock art tradition. Additionally, it may be important that most of the imagery at Punta Picata is noniconic, which means that abstract imagery predominates. The relatively many solar images at this rock art complex may be the result of its position directly on the shore of the Pacific Ocean, offering perfect spots to watch the sun set and to ritually manufacture motifs symbolising the importance of the (setting) sun.

Figure 1. Maps of the Study Areas showing the locations of some of the sites mentioned in the text. Inset 5: Rock panel at Red Canyon, California. Drawing © by Maarten van Hoek, based on a photo by Eugene Carsey.

About 235 km NW of Punta Picata is the largest rock art site of the Andes, Toro Muerto in the Majes drainage, southern Peru, with an estimated maximum of 3000 decorated boulders. However at only one (very large) boulder (TM-Fa-001) a small rock edge has intentionally been serrated, the notches clearly overlapping the edge (Van Hoek 2016: Fig. 21). However, the reason for this incidental type of marking is completely obscure. Again, this example does not at all reveal a local tradition to mark the rock’s edges. Moreover, the imagery on this boulder is mainly iconic (i.e. zoomorphic and anthropomorphic images predominate).


Quebrada de Algarrobos

No less than 1170 km NW of Toro Muerto is the important rock art complex of Chuquillanqui in the Chicama drainage, northern Peru. This complex comprises two significant rock art sites with altogether over 300 decorated surfaces: Quebrada de Algarrobos and (1400 m to the north) Cerro el Diablo. Those two sites share more or less the same imagery. However, it is remarkable that only at Quebrada de Algarrobos several rock edges have intentionally been serrated. My surveys at Quebrada de Algarrobos revealed no less than sixteen examples of serrated rock edges (Van Hoek 2016). This fact confirms that it definitely once was a very localised and intentional tradition to create serrated edges. The majority is found at Sector 2 of Quebrada de Algarrobos, where an impressive outcrop stack – comprising four large, almost vertical panels (CH-ALG-2-025 to 028) – has several serrated edges and also the great majority of (often noniconic) images (Figure 2A). Opposite this wall is another, lower and more sloping wall with several serrated edges (and again many other petroglyphs). Importantly, serrated edges at Quebrada de Algarrobos appear on horizontal rims as well as on vertical and sloping or diagonally arranged edges and also at small, almost horizontal boulders (Figure 3).

Figure 2. Outcrop stacks (Panels CH-ALG-2-025 to 028) at Quebrada de Algarrobos, Chuquillanqui, northern Peru (explanation in the text). Photographs © by Maarten van Hoek.

As far as I could check, there is no relationship between the serrated edges at Quebrada de Algarrobos and the subject matter depicted by the petroglyphs on the adjacent panel(s), except however for one exception. Panel CH-ALG-2-025 is special in this respect, as it not only has several 3D serrated edges, but also two 2D-petroglyphs (marked 3 in Figure 2B) that seems to mimic the parallel orientated serrated edges that can be seen only a short distance above these two petroglyphs (marked 2 in Figure 2B).

Figure 3. Part of Panel CH-ALG-2-015 at Quebrada de Algarrobos, Chuquillanqui, northern Peru. Photograph © by Maarten van Hoek.

Earlier (Van Hoek 2016) I suggested that the serrated edges at Quebrada de Algarrobos might mimic the indented edges of the numerous impressive cacti at Chuquillanqui (often more than 10 meters high). Nevertheless, despite my tentative suggestion, the true meaning of those serrated rock edges –  found only at Quebrada de Algarrobos – still remains a mystery. Finally, at Chuquillanqui several biomorphic petroglyphs definitely belong to the Andean Formative Period (roughly dating from 4000 B.C. to A.D. 0; Van Hoek 2011). But the dating of the serrated edges and many of the other petroglyphs is uncertain.


North America

As far as I know no instances of serrated rock edges have been recorded at rock art sites further north of Quebrada de Algarrobos, Peru, until the West of the USA is reached. However, it proves that serrated edges are often overlooked by rock art researchers in the field and/or have not been included in their publications or inventories. For example, none of the – in my opinion – important serrated edges at Quebrada de Algarrobos, Peru, was mentioned or illustrated in the inventory of Chuquillanqui by Peruvian archaeologist Daniel Castillo Benites (2006: 77-93). Therefore, in general there may well be more examples at more sites.

It now proves that the intentional practice to create serrated rock edges – despite being extremely rare – is rather surprisingly widespread across the West (this in sharp contrast with the heavily localised examples of Quebrada de Algarrobos, Peru). In this study it will not be attempted to culturally link all sites with serrated edges (known to me; there probably are [many?] more examples), as they are many kilometres apart and cultural differences will no doubt have resulted in different meanings.


Sites with Serrated Edges in the West

To (arbitrarily) start with, Malotki (2012: Fig. 6) illustrates a panel with a notched edge (also having several paired animal “tracks”) at an undisclosed rock art site “along the Rio Grande in New Mexico” (1 in Figure 1), which is said by Malotki to be a “Western Archaic Tradition site”. Importantly, the “Western Archaic Tradition” is mainly found some 1000 km NW of the Rio Grande in the area called the Great Basin and adjacent areas. In this area numerous examples of a very ancient rock art style are found, called the Great Basin Carved Abstract Rock Art Style, which clearly shows a preference for elementary noniconic, sheer geometric designs. A vertical petroglyph panel in the Great Basin Carved Abstract Rock Art Style at the Mazama site in Oregon, partially being sealed by a volcanic ash layer, is said to provide a minimal age of 7700 years (Malotki 2012: Fig. 3). Although I do not want to claim that all rock art sites with Great Basin Carved Abstract petroglyphs are that old, it is generally accepted that the Great Basin Carved Abstract Rock Art Style represents the oldest form of rock art in the west of the USA (Schaafsma 1980: 36). Therefore, the following examples of rock art panels with serrated edges may be equally old, although I do not want to claim that the serrated edges are always contemporary with the Great Basin Carved Abstract Rock petroglyphs on those panels, although serrated edges clearly occur in the context of early archaic rock art. Still, (some) serrated edges may well be a later addition.

At Grimes Point (2 in Figure 1), a rock art site on the banks of the ancient Lake Lahonton in western Nevada (1165 km NW of the Rio Grande) with over a thousand basalt boulders bearing petroglyphs, at least one boulder has a serrated top rim, but probably there are more (Malotki 2012: Fig. 5). In fact, the Grimes Point boulder (Figure 4) has its row of notches on the domed top of a rounded boulder. There is in fact no true edge. Therefore this example may exemplify a different tradition. A close but unrelated parallel is found at the Forsythe petroglyph boulder, Georgia, which has a row of cupules on its upper edge.

At Grimes Point noniconic, abstract designs predominate (including numerous very old cupules), while only a few biomorphic images have been reported. The petroglyphs at this site are called Iza?a thonnu, “Coyote’s writings”, indicating that it was made by the well-known trickster figure during “The Time When Animals Were People” (Woody 2000: 202), which actually means that they are very old. Also near Long Lake, Oregon (3 in Figure 1), roughly 100 km NNW of Grimes Point, Malotki reports petroglyph rocks with serrated edges (2012: 629). But as I have not seen photographs of those purported Long Lake serrated edges, I cannot guarantee that they fit into this category.

Figure 4. Boulder at Grimes Point, Nevada, USA. Drawing © by Maarten van Hoek, based on a photo by Malotki (2012: Fig. 5).

About 200 km south of Grimes Point is the rock art site of Chalfant, California, located in the Owens (or Chalfant) Valley (4 in Figure 1). At the west side of the valley is an extensive cliff with numerous, mainly noniconic petroglyphs. At this site John Martinez photographed a projecting outcrop with two vertical panels with some petroglyphs (including some vulva-designs?), the corner of which was serrated by possibly up to 32 notches (Figure 5). Only 8 km north of Chalfant is the rock art site of Red Canyon (5 in Figure 1), where Eugene Carsey photographed a boulder that has a serrated edge featuring at least 18 rather long and irregular notches (Figure 1; inset 5).

Figure 5. Outcrop at Chalfant, California, USA. Drawing © by Maarten van Hoek, based on a photo by John Martinez (Facebook).

The major concentration of rocks with serrated edges in the west of the USA is found (as far as I know) at War Shield Canyon (6 in Figure 1), off Upper Arrow Canyon, in southern Nevada, 440 km SE of Grimes Point and 360 km ESE of Chalfant. No less than four slabs of stone with serrated edges have been photographed in this canyon by Southern Nevada Rock Art & Nevada Drone Archaeology. Somewhere in the canyon is a long triangular boulder slab with one edge that is decorated with a row of irregular notches. The decorated panel has a large number of cupules. Another slab (either with two stepped panels or a stone slab lying on another slab) has a row of possibly twelve crudely executed notches. The other, lower panel has a row of three rather complex circular designs, while its edge has possibly up to 17 crude notches that continue on the other side of the panel (Figure 6).

Figure 6. Boulder at War Shield Canyon, Nevada, USA. Drawing © by Maarten van Hoek, based on a photo by Southern Nevada Rock Art & Nevada Drone Archaeology (Facebook).

The most impressive example concerns an enormous triangular slab (estimated by me to measure at least 3 meters in height). It forms part of a group of three, closely positioned slabs located on the south bank and near the mouth of the canyon. It is suggested that they once fell down the cliff walls. They are known as the Calendar Rocks, although it seems that only the large triangular rock is actually labelled in this way (Carroll 2007: 300; Figs 5-41, 5-43). At least two of the slabs feature serrated edges. The upper ridge of the large triangular slab is decorated with a very long row of notches, each of which continues as a very short groove on the almost vertical side panel that faces the valley (Figure 7). There are several other, much weathered petroglyphs on this panel, including a long groove just below the notches. This groove may well be associated with the notches. Also the shorter edge on the right has some less-developed notches that are also associated with (and possibly even joined to) a long groove. Carroll (2007: 282) mentions the serrated edges as  “deeply incised lateral knife marks inscribed around circumference of a panel that has fallen from an adjacent canyon wall”.

Figure 7. Boulder at War Shield Canyon, Nevada, USA. Drawing © by Maarten van Hoek, based on a photograph by Peter Goin (1951), stored at the Special Collections and University Archives Department, University of Nevada, Reno.

The fourth boulder, which is found a very short distance behind the triangular slab, comprises a steeply sloping panel with a lot of petroglyphs, including a very distinct “Venus-Cross” design and – rather unusual for this type of rock art – an anthropomorphic petroglyph (Figure 8). The long, irregular upper ridge of this slab has numerous notches. It could not be established whether the notches continued as very short grooves on the other side of the slab (like at the triangular slab).

Figure 8. Part of a large boulder at War Shield Canyon, Nevada, USA. Drawing © by Maarten van Hoek, based on a photo by Southern Nevada Rock Art & Nevada Drone Archaeology (Facebook).

It may be significant that War Shield Canyon has small pools that periodically fill with water. It is therefore no surprise that, according to Kristen Jean Carroll, the Arrow Canyon site is located on an ancient, sacred pathway. Known as the Salt Song Trail, which was shared by the Paiute (the indigenous peoples of the Great Basin) and the Hualapai, this path covers a vast territory that corresponds directly to the traditional boundaries of the Numic homeland. Most importantly, she argues that, although there are multiple song trails in the area, there is only one trail to afterlife (Carroll 2007: 87, 289-299). Yet, the petroglyphs will probably be much older as “Some elders indicated that the petroglyphs in Arrow Canyon are very old and were there long before the Ghost Dance was conducted” (Carroll 2007: 309). If indeed the serrated edges are contemporary with the Great Basin Carved Abstract Rock Art Style they may be thousands of years old. However, there is no way to establish whether the notches are earlier, contemporary or later.



It is surprising to see that serrated edges occur in northern Peru and in the West of the USA (see Figure 1). The salient difference that in Peru the practice occurred heavily concentrated only at one site (Chuquillanqui, disregarding the single example at Toro Muerto in southern Peru), whereas in the West of the USA serrated edges are found on at least six sites that are often separated by hundreds of kilometres. What connects those USA sites is that they probably all belong to the ancient Great Basin Carved Abstract Rock Art Style. However, this does not mean that all the instances of rocks with serrated edges have the same meaning, symbolism or perhaps the same practical function. Especially the meaning or symbolism of the notches at the rock art site of Quebrada de Algarrobos, Chuquillanqui, Peru, may be completely different to those in the West of the USA. Yet it must be mentioned that one of the serrated slabs at War Shield Canyon, USA, also has a very distinct petroglyph of a “Venus-Cross”. Although it most likely is sheer coincidence, it is remarkable that two similar (yet less complex) “Venus-Cross” petroglyphs are also found very near the serrated edges on Panel CH-ALG-2-025 (one indicated with an orange arrow; the other only faintly visible and indicated with an orange + in Figure 2A).

The fact that at least one boulder with serrated edges (the large triangular stone at War Shield Canyon) is called the Calendar Stone, does not imply at all that serrated edges have anything to do with measuring time. For that reason I also do not believe that those notches are tally marks or tallies. Tally marks usually comprise rows of short, vertically oriented markings presumed to be an ancient accounting of something or some events. Tally marks (mainly pictographs) or markings looking like tally marks are quite numerous in the West of the USA (see for example Keyser 1992), but are also said to occur much further east, for example at Pigeon Cave (34Ci-48), Black Mesa, Cimarron County, Oklahoma (Figure 9), but at Pigeon Cave these vertically arranged 2D-grooves do not occur on the rock’s edges.

Figure 9. Outcrop at Pigeon Cave, Oklahoma, USA. Drawing © by Maarten van Hoek, based on a photo by “Meriword”.

Likewise I do not think that those notches have had an astronomical function. However, I cannot rule out that in some cases the notches were used to count or to commemorate something. Perhaps (for example in the case of the examples at Chuquillanqui, Peru) the serrated edges (only?) served to emphasise the importance and sacredness of the site. Nevertheless, despite all attempts to interpret and explain the phenomenon of the serrated edges, it will remain completely obscure what function they served or what symbolism they expressed. Hence it will also be clear that there is no question of serrated edges having been diffused from Chuquillanqui to the West of the USA or vice versa, even when symbolism or function will prove to have been analogous after all.



Carroll, K. J. 2007. Place, Performance, and Social Memory in the 1890s Ghost Dance. PhD. Dissertation. The University of Arizona.

Castillo Benites, D. S. 2006. Arte Rupestre en La Cuenca del Chicama. Primera edición, Ediciones Sian, Trujillo.

Keyser, J. D. 1992. Indian Rock Art of the Colombia Plateau. University of Washington Press.

Malotki, E. 2012. Precursors of figurative depictions in the noniconic Western Archaic Tradition rock art of the American West. In: Pleistocene art of the world. Actes du Congrès IFRAO, Tarascon-sur-Ariège. Symposium “Art pléistocène dans les Amériques”.

Schaafsma, P. 1980. Indian Rock Art of the Southwest. University of New Mexico Press.

Umire Álvarez, A. 2012. Los petroglifos en la costa norte de Tacna: los complejos de Punta Picata e Ite. Investigaciones Sociales. Vol.16-28, pp. 389 – 395. UNMSM-IIHS. Lima, Perú.

Van Hoek, M. 2011. The Chavín Controversy – Rock Art from the Andean Formative Period. Oisterwijk, The Netherlands.

Van Hoek, M. 2016. Los Bordes Aserrados de las Rocas de Chuquillanqui, Perú / The Serrated Rock-edges of Chuquillanqui, Peru. In: TRACCE – On-line Rock Art Bulletin, Italy.

Van Hoek, M. 2018a. The Pipette-Design in Desert Andes Rock Art. In: ADORANTEN-2017; pp. 103 – 115. Underslös, Sweden.

Van Hoek, M. 2018b. Indifferent Obliteration of Petroglyph Art. TRACCE – Online Rock Art Bulletin, Italy. PDF available at ResearchGate.

Van Hoek, M. 2018c. Possible Indications for Long Distance Diffusion of Rock Art Motifs in the Americas. In: TRACCE – Online Rock Art Bulletin, Italy.

Van Hoek, M. 2020. Rock Art Images Wrapped Around the Corner – The Three Rivers 3D Masks. In: Patina; Utah Rock Art Research Association (URARA). Salt Lake City, Utah.

Woody, A. J. 2000. How To Do Things With Petroglyphs: The Rock Art of Nevada. PhD. Thesis, University of Southampton, Department of Archaeology. Pp. 333.

Leave a Reply

Your email address will not be published. Required fields are marked *

3 × 5 =