THE FOSSIL FERNS OF PETRIFIED FOREST NATIONAL PARK, ARIZONA, AND THEIR PALEOCLIMATOLOGICAL IMPLICATIONS

SIDNEY ASH

Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, NM 87131


____________________

Abstract—Approximately one-fifth (ten species) of the plant megafossils in the Chinle Formation in Petrified Forest National Park, Arizona are the remains of ferns or fern-like leaves. Four of the species are assigned to living families, two are assigned to extinct families and the others are unassigned. The nearest living relatives and morphological features of these plants indicates that they lived under a humid tropical to subtropical climate that could have been monsoonal.

____________________
INTRODUCTION

Although, it was originally established to protect the large numbers of petrified logs that occur there, Petrified Forest National Park in east-central Arizona (Fig. 1) also contains significant deposits of compressed leaves, seeds, cones, and other plant parts, all of which occur in the Chinle Formation of Late Triassic age (Late Carnian and Early Norian Stages). Some of the most abundant of these compression fossils are the remains of ferns and fern-like foliage. At present this significant part of the flora consists of ten well-characterized species based on megafossils (Table 1). It will be useful now, at the beginning of the new millennium, to summarize what is now known about them and their paleoclimotological implications. The numerous dispersed fern spores found in the park that were described by Litwin (1984, 1986) and Litwin, et al. (1992) are beyond the scope of this article and are not considered here although they support the conclusions reached in this article.

map of four corners region
Figure 1. Map of part of the western United States showing the boundary of the Colorado Plateau and the location of Petrified Forest National Park and the other localities where significant numbers of Late Triassic ferns have been found. CL=Circle Cliffs, CR=Capitol Reef, FW=Fort Wingate, WI=White Canyon.

GEOLOGIC BACKGROUND

The Chinle Formation was deposited in a large continental basin that was situated a short distance north of the equator on the west coast of Pangea (Blakey, 1989). It is composed of strata deposited by northward-flowing streams and rivers and in lakes and on flood plains in a succession of incised paleovalley cut-and-fill complexes (Dubiel, et al., 1999). The unit is now exposed over broad parts of the Colorado Plateau and because of its environment of deposition is subdivided into several understandably discontinuous units (Stewart. et al., 1972). In Petrified Forest National Park the formation is about 300 m thick and is conventionally (see Dubiel, et al., 1999) divided into the Shinarump, Mesa Redondo, Monitor Butte, Petrified Forest and Owl Rock Members (Fig. 2). The Sonsela Sandstone Bed, a persistent and distinctive lithologic unit, occurs between the Monitor Butte and Petrified Forest Members and is of some correlative value in the eastern part of the Colorado Plateau. Another noteworthy unit, the Newspaper Sandstone Bed, is present locally in the Monitor Butte Member in the park.

In Petrified Forest National Park, petrified wood occurs in large quantites in the Sonsela Sandstone Bed and subadjacent strata of the Monitor Butte Member. It also occurs there in smaller quantites at several other horizons in both the Monitor Butte and Petrified Forest Members (Ash and Creber, 1992). The largest concentrations of petrified wood in the park are termed forests and given names that reflect some particular aspect of the wood (Ash, 1987). In the southern part of the park they are Rainbow Forest, Crystal Forest, and Jasper Forest (Fig. 1). The wood in them was derived from the Sonsela Sandstone Bed and the uppermost part of the Monirtor Butte Member. Another forest occurs in the Black Forest Bed in the Petrified Forest Member and because the wood is black it is called the Black Forest. . Most of the wood in the forests consists of long prostrate logs and short broken sections but in the Black Forest Bed and the Monitor Butte Member there are a few stumps that are in growth position. In contradst, the deposits of compressed plant remains are much less common and are smaller in extent than the forests. The largest of the compressed plant deposits are found in the Monitor Butte Member adjacent to the Newpaper Sandstone Bed. Smaller deposits occur elsewhere in the Monitor Butte Member and in the Petrified Forest Member. In many, if not all cases, the compressed plant deposits occur either in overbank deposits or in local swamp deposits. None of the deposits in the Petrified Forest cover more than a few hundred square meters and the ferns only occur in the one that is associated withthe Newspaper Sandstone Bed (Fig. 2). at other localities beyond Petrified Forest National Park. compressed plant fossils also occur only in the lower part of the Chinle Formation in the Shinarump and Monitor Butte Members mainly (Ash, 1989).

 

Chinle Formation
Figure 2. Diagram showing the members and other units of the Chinle Formation in Petrified Forest National Park and the relative positions of the petrified forests, the stumps that are in the position of growth, and the major deposit of compressed ferns. Adapted from several sources, particularly Dubiel, et al., 1999 and Ash, 1987

PREVIOUS INVESTIGATIONS
map of Petrified Forest National Park

Figure 3. Map of Petrified Forest National Park showing the location of the forests and other features discussed in this article. Adapted from Ash, 1987

Fossilized remains of ferns were not discovered in the Petrified Forest or anyplace else in the American southwest until nearly a century after representatives of the conifers and cycadophytes were first found in the Chinle Formation (Ash, 1972b). Probably the principal reason that they remained undetected for so long is that fern fossils, like their living counterparts, are relatively delicate and are typically preserved in fine-grained rocks which had not been adequately investigated until the early part of the last century.

Fern fossils were first observed in the Chinle Formation during the early 1930s by the CCC crew that was building an all-weather road through Petrified Forest National Park (Ash, 1972b). The fossils occurred in grayish mudstones adjacent to the Newspaper Sandstone Bed in the Monitor Butte Member in a cut that the crew was excavating for the roadbed in the central part of the park near the Tepees (Fig. 3). The workers found compressions of fern fronds, some of which were complete and compressed leaves, cones and seeds of horsetails, cycadophytes, conifers, and other plants. Fortunately for science, Myrl Walker, the first naturalist appointed by the park, recognized the importance of the discovery and saw to it that many of the fossils were collected and preserved for exhibit and research. Later he arranged for their study by Lyman Daugherty, a botanist at San Jose State University in California. The results of his investigation were published a few years later (Daugherty, 1941). In the report, Daugherty described a total of 38 species based on megafossils, including eleven species of ferns and fern-like foliage assigned to nine genera (Table 1). Seven of them were new and three were known previously from the Late Triassic Newark Group in the eastern United States. Daugherty later (1960) described the trunk of a new tree fern from the Petrified Forest. Shortly afterwards, I began a reinvestigation of the Chinle flora, basing my work on new material collected from localities throughout the Southwest as well as from Petrified Forest. During those investigations I re-evaluated all of Daugherty's species of ferns and fern-like foliage (Ash, 1970, 1972a, 1973, 1994, 1999). Although most of his identifications were accepted, I redescribed several of his species using new techniques, and in some cases reassigned them to different and sometimes new genera. More recently, a new dimension was added to the biology of these plants when the evidence that some of them had been used by arthropods for food and shelter was discussed (Ash, 1997, 1999, 2000).






SYSTEMATIC REVIEW

Here, the salient features of each of the megafossils assigned to the ferns and the fern-like foliage found in the Chinle Formation in Petrified Forest National Park, Arizona are briefly described. The ferns are considered systematically under their assigned family and then the species based on fern-like foliage are discussed. Detailed descriptions and more illustrations of the fossils appear in the articles cited. Illustrations of each species based on leaves are given in Fig. 4. The key to the species given in Table 1 will assist with the identification of the leaves of fossil ferns or fern-like foliage discovered in the park by future investigators.

Family Osmundaceae

The only representative of this living family found in the park is Todites fragilis Daugherty emend. Ash, 1970. The leaf is small (largest specimen is 6 cm wide, 12.5 cm long), delicate, and bipinnate and has distinctly asymmetrical pinnae (Fig. 4E). On the lower side of the rachii of the pinnae the lamina is narrow and undivided, whereas on the upper side the lamina is broad and near the main rachis of the leaf is divided into several ovate pinnules, but it becomes progressively less divided apically and eventually becomes totally undivided. Venation consists of a midvein which gives off lateral veins alternately. Fertile and sterile pinnae occur on the same leaf. The sporangia are solitary on the lower surface of the pinnules and are round to oval with a subapical annulus. Small, round to slightly triangular trilete spores similar to the dispersed spore Osmundacidites parvus De Jersey (see Litwin, 1984) are produced in the sporangia on the distinctive leaves assigned to this species. Todites fragilis is relatively rare and is known only from the Monitor Butte Member of the Chinle Formation in Petrified Forest National Park and in the Fort Wingate area, New Mexico (Ash, 1970). Living members of this family typically inhabit humid subtropical to tropical parts of the world such as southeast Asia.

Family Guaireaceae

Itopsidema vancleaveii Daugherty (1960) is the petrified stem of a small treefern which is about 15 cm in diameter and when alive, probably was only about one meter tall. While the plant was living the stem would have been relatively soft and spongy and held upright by a mantle of numerous adventitious small roots. The sides of the stem were covered with the broken bases of many leaves and a crown of large leaves of unknown type was present on the top of the stem. Although, this fossil was originally assigned to the Osmundaceae by Daugherty (1960) recent investigations indicated that it should be assigned to the Guaireaceae, an extinct family of Mesozoic ferns (Ash, 1994). The only specimen of this species was collected from the Monitor Butte Member of the Chinle near the base of Agate Bridge Mesa (Daugherty, 1960; Ash, 2000).

Small coprolite-bearing cavities reported by Ash (2000) in this fossil indicate that the living plant provided shelter and food for mites. The cavities occur in and near the leaf petioles and in the root mantle, are round to irregularly shaped and range up to a few millimeters in diameter. Although the cavities are not lined with reaction tissue, a few masses of it are present in some of the cavities, indicating that the plant was alive when attacked. The coprolites are oval to weakly hexagonal in shape and the largest are slightly less than 0.1 mm in length. The mite responsible for the damage is unknown. Modern treeferns live in humid tropical and subtropical parts of the world.

Family Cynepteridaceae

Cynepteris lasiophora Ash (1970) is the only known representative of this extinct family of ferns. The fossil is a large (24 cm wide, possibly 1 m long) bipinnate leaf in which the pinnules are generally oblong and have reticulate venation (Fig. 4H). Fertile and sterile pinnules occur on the same pinna. The sporangia are solitary and are irregularly scattered on the lower surface of the pinnules or concentrated on just the basal parts. They are oval to pear-shaped and have a complete apical annulus. Spores of Cynepteris lasiophora are small and round to subtriangular and are comparable to the dispersed spore Verrucosisporites sp. C of Bharadwaj and Singh (see Litwin, 1984).

Some of the leaves of this distinctive fern show evidence of having been the source of food for phytophagous insects (Ash, 1997). This evidence consists of excisions along the margins of the pinnules where a portion of the lamina has been removed. In some places crescent shaped portions of the lamina were removed whereas in other places the excisions are more or less straight, paralleling the pinna rachis and cut across several pinnules. The margins of the excisions are sharply defined and unfrayed. Typically the excisions are bordered by a narrow zone of reaction tissue that includes both callus and necrotic tissue, indicating that the feeding occurred while the leaves were alive and still attached to the parent plant.

Cynepteris lasiophora is one of the most abundant and widespread of the ferns in the Upper Triassic of the western United States where it has been described from the Chinle Formation in Petrified Forest National Park, in the Fort Wingate area of New Mexico, and the White Canyon, Circle Cliffs, and Capitol Reef areas in southeastern Utah, as well as in the Dockum Group near Santa Rosa in eastern New Mexico and near Amarillo in west Texas (Ash, 1970; 1972c).

Family Gleicheniaceae (?)

Wingatea plumosa (Daugherty) Ash (1970) is tentatively assigned to this living family. The fossil is a large (up to 12 cm wide, 50 cm long), highly dissected tripinnate leaf with very small ultimate segments that vary in outline from oblanceolate to narrowly lobed (4F). Venation is dichotomous with a single vein in each lobe. Fertile and sterile pinnae occur on the same leaf. The sporangia are clustered in ring-like sori on the backs of the fertile pinnae and include 8-12 sporangia. The annulii on the sporangia are obliquely horizontal apical rings. The spores produced in the sporangia are rounded triangular and reportedly (Litwin, 1984) are similar to the dispersed spores assigned to Todisporites rotundiformis (Malyavkina) Pocock. This distinctive species is fairly common in the Monitor Butte Member of the Chinle Formation in Petrified Forest National Park, Arizona, the Fort Wingate area of New Mexico, and the White Canyon area in southeastern Utah (Ash, 1970, 1975).

Family Matoniaceae

This living family is represented by the pedately compound leaf Phlebopteris smithii (Daugherty) Arnold emended Ash, et al. (1982). This fossil was assigned to the obsolete genus Laccopteris when it was originally described by Daugherty (1941) but later it was assigned to the correct genus Phlebopteris by Arnold (1947). The leaf is composed of as many as 14 pinnae that are divided into narrow, oblong pinnules with rounded apices. The pinnae arise from the upper sides of two lateral arms that develop from the top of a long petiole (Fig. 4I). Each pinnule contains a strong midrib which extends most of the way to the apex of the pinnule and then dissolves into several veins that end at the margin. Lateral veins arise alternately from the midrib and usually fork once before reaching the pinnule margin. Both fertile and sterile pinnules occur in the same pinnae. Superficial ring-like sori occur on the lower surfaces of the pinnules of this species and consist of 5-14 oval sporangia that have a vertical to slightly oblique annulus. The spores are more or less triangular and trilete in polar view and according to Litwin (1984) are most similar to the dispersed spore species Dictyophyllidites harrisii Couper and D. mortonii Playford and Dettman. The remains of this species are very common in the Monitor Butte Member of the Chinle Formation in Petrified Forest National Park, Arizona, near Fort Wingate, New Mexico, in the White Canyon and Lake Powell areas of southeastern Utah (Arnold, 1956, Ash, 1975). It is also found in the Dockum Group of Late Triassic age in west Texas near Amarillo. Living members of this family are found in the humid monsoonal tropics of Southeast Asia.

Family Dipteridaceae

Clathropteris walkerii Daugherty emend. Ash (1970) is the only representative of this living family known in the park. It is a large palmate leaf up to 30 cm in diameter consisting of six to ten linear-lanceolate pinnae spreading outward from a broad basal web (Fig. 4J). The pinnae are up to 12 cm long and 3 cm wide and have acute apices. Numerous, short, acutely pointed, forward directed teeth are present along the pinnae margins. The leaf has net venation with many blind vein endings. Sori are irregularly distributed on the lower surfaces of the fertile leaves and consist of a round mass of 7-9 spherical sporangia which have complete oblique annulii. The spores of this species are rounded triangular in polar view and reportedly (Litwin, 1984) are most similar to the dispersed spore species Granulatisporites infirmus (Balme) Cornet and Traverse.

This species has been found in the Monitor Butte Member of the Chinle Formation in Petrified Forest National Park, in the Fort Wingate area of New Mexico and the White Canyon area in southeastern Utah as well as in the Dockum Group near Amarillo in west Texas (Ash, 1970; 1972c). Living members of this family inhabit the humid monsoonal tropics of Southeast Asia.

Family Uncertain

A few species of fern-like leaves that can not be assigned to a particular family of ferns because their fertile structures are not known occur in the Chinle Formation in Petrified Forest National Park. It is possible that some of these fossils are in fact the leaves of pteridosperms (commonly called seed ferns), a group of extinct plants that had fern-like leaves but reproduced by means of seeds, not spores. By convention fern-like leaves are usually treated as ferns until additional information is forthcoming.

One of the most common of the fern-like leaves in Petrified Forest National Park is assigned to Cladophlebis daughertyi Ash (1970). It is a fairly large bipinnate leaf which when complete was about 30 cm wide and probably 60 cm or more in length when alive. The pinnules are small (typically 2 mm x 4.5 mm), delicate and somewhat falcate in outline with smooth margins (Fig. 4A). Each pinnule contains a midrib from which narrow lateral veins arise in an alternate manner Fig. 4B). Cladophlebis daughertyi occurs in the Monitor Butte Member of the Chinle Formation in Petrified Forest National Park and the Fort Wingate area of New Mexico.

A second species of Cladophlebis, C. yazzia Ash (1973), has been described also from the Petrified Forest where it is represented by just a few small fragments. The species is somewhat similar to C. daughertyi but the leaf is only about 5 cm wide and the pinnules are typically long and falcate with dentate margins particularly in the upper parts (Fig. 4G). Each pinnule contains a midrib from which narrow lateral veins arise in an alternate manner.

Sphenopteris arizonica Daugherty emend. Ash (1999) is the third fern-like leaf to have been described from the park. The leaves that have been found are about 10 cm wide and 40 cm long. They are tripinnate below becoming bipinnate above. The ultimate segments (pinnules) are oval to lobed and range up to 6 mm tall and 4 mm wide (Fig. 4C). There is the possibility that these leaves actually are parts of a large bipartite leaf. If this were to be confirmed it would indicate that S. arizonica is a seed fern as are at least some of the other fossils referred to the genus. The pinnules on some of the leaves attributed to this species show evidence of having provided nourishment for phytophagous insects (Ash, 1999). The evidence is similar to that reported on the fern Cynepteris lasiophora and demonstrates that the feeding occurred while the leaves were alive and still attached to the parent plant.

Marcouia neuropteroides (Daugherty) Ash is another example of a fern-like frond found in the Chinle Formation in Petrified Forest National Park and elsewhere in the Colorado Plateau region. This leaf is a palmately compound leaf that was perhaps as large as one meter in diameter according to Daugherty (1941). The pinnae are linear-lanceolate and are divided into oval to linear pinnules which have wavy to lobed margins (Fig. 4D). Numerous lateral veins arise from the pinnule midrib at a high angle and divide and anastomose one or more times with adjacent veins. Usually the anastomosing takes place near the margins and the lateral veins are rarely free at the margins. Originally Daugherty (1941) attributed this species to the genus Ctenis but that assignment was found unacceptable by Ash (1972d) and the fossil was reassigned to the new genus Marcouia. Small oval holes found in several of the pinnules of M. neuropteroides appear to have been caused by phytophagous insects (Ash, 1997).
sketches of the pinnules, venation and morphology of the fern leaves and the fern like fronds found in the Chinle Formation

Figure 4. Sketches of the pinnules, venation, and morphology of the fern leaves (E-F, H-J) and the fern-like fronds (A-D, G) found in the Chinle Formation. A,B, Cladophlebis daughertyi. C, Sphenopteris arizonica. D, Marcouia neuropteroides. E, Todites fragilis. F, Wingatea plumosa. G, Cladophlebis yazzia. H, Cynepteris lasiophora. I, Phlebopteris smithii. J, Clathropteris walkerii. (A, D, x1 1/2; B, C, F, G, H, x3; E, I, J, x1/4) Adapted from several sources including Ash, 1970, 1972a, d, 1973, 1980, 1999.



PALEOCLIMATOLOGICAL IMPLICATIONS

Of all the fossil plants found in the lower part of the Chinle Formation, the ferns seem to be particularly useful for reconstructing the climate under which it was deposited. Such reconstruction using fossil ferns and/or other plants is based on two assumptions: first, that the climatic preferences of fossil plants were similar to that of their nearest living relatives and second that the morphological features of fossil species are similar to those of extant plants inhabiting a particular existing climate.

Before discussing the paleoecological implications of the Chinle ferns it should be noted that most (about 80%) living ferns are restricted to the tropics and subtropics where the regional climate ranges from consistently wet to monsoonal (Lellinger, 1985). In those regions, the climate is humid throughout the year, although it may be slightly less so during the dry part of the monsoonal cycle. Also, it should be noted that the uplands in those regions are as humid as the river valleys, if not more so (Tryon, 1986). The relatively specialized, so called desert ferns live in more arid areas, but they are easily distinguished from the more common ferns that inhabit the humid tropics and subtropics.

The nearest living relatives of all the Chinle ferns live in the humid tropics and subtropics. Two of them, Clathropteris walkerii and Phlebopteris smithii, belong to families that formerly had a worldwide distribution (Tidwell and Ash, 1995) but now only occur in areas of southeast Asia that have a monsoonal climate (Bower, 1928). The other Chinle ferns are not as closely related to any living ferns but nevertheless their relatives always inhabit the humid tropics and subtropics where some may even live under a monsoon climate. None of the Chinle ferns are at all closely related to the desert ferns.

The morphological features of the Chinle ferns are comparable to those of ferns that now live in consistently wet to monsoonal tropical to subtropical climates. Just like their living counterparts from those climatic regimes, the Chinle ferns have superficial, unprotected stomata on both sides of the leaves. The leaves of most, including Todites fragilis, Cynepteris lasiophora, and Wingatea plumosa as well as the two species of Cladophlebis were originally thin and delicate. These ferns, like their living counterparts, probably inhabited shady areas under larger plants such as cycadophytes and conifers and perhaps other ferns such Phlebopteris smithii and Clathropteris walkerii, which had somewhat more robust leaves. These latter two ferns, just like their nearest living relatives, probably lived in an open environment and provided shelter for the more delicate ferns and other plants, as is characteristic of their descendants in the monsoonal tropics (Bower, 1928).

The Chinle ferns have none of the special morphological adaptations that permit ferns to survive in areas where the water supply is severely limited. For example, the leaves in the Chinle ferns are generally broad (always narrow in desert ferns) and remain flat (curl up when the weather gets extremely dry in desert ferns), and the stomata in the Chinle ferns occur on both sides of the leaves (restricted to the lower side in desert ferns) and are virtually unprotected (protected by hairs and scales in desert ferns). A desert fern that shares these later characters is Notholaena parryi. It inhabits the Colorado Desert and has small leaves that curl up when the weather becomes extremely dry. Nearly all of the stomata occur on the lower surface of the leaves in this species and they are protected by hairs and scales ( Helvy, 1963). Similar characters also are present in the several species of the desert fern Cheilanthes which inhabit the arid parts of India (Nayar(1962) and elsewhere (Marsh, 1914).

Table 1. Key to the fern leaves and fern-like foliage of the Chinle Formation. The morphology and venation of the leaves are used in this key to differentiate the species illustrated in Figure 4.

1. Leaf palmate ……………………………………………………………………………..…..…. 2

1.Pinnae margins toothed..........................................……......................... Clathropteris walkerii

2. Pinnae margins wavy to lobed................................…........................ Marcouia neuropteroides

1. Leaf not palmate..............................................................................................................................2

2. Leaf pedate............................................................................................... Phlebopteris smithii

2. Leaf pinnate.............................................................................................................................3

3. Venation reticulate.................................................................. Cynepteris lasiophora

3. Venation not reticulate (open)...........…....................................................................4

4. Pinnules rounded.....................................................................................5

5. Basiscopic pinnules fused..............….....................Todites fragilis

5. Basiscopic pinnules not fused.....….….......Sphenopteris arizonica

4. Pinnules not rounded (angular)..................................................................5

5. Pinnules narrow............................................... Wingatea plumosa

5. Pinnules not narrow (broad)........................................................6

6. Pinnule margins toothed.....................Cladophlebis yazzia

6. Pinnule margins not toothed (smooth).........C. daughertyi


The morphology and nearest living relatives of many of the non-fern species found in the Chinle Formation indicate that they too lived under a regional humid tropical to subtropical climate. Although most of them obviously lived along streams and around lakes, a few obviously were transported some distance from uplands to where they were deposited, and provide data on the paleoclimate of those areas. Particularly instructive in this respect are the compressions of leaf shoots and isolated leaves of the conifers in the flora. They are generally rare and fragmentary, indicating that they were transported some distance before burial. These fossils do not have the epidermal features that are said (Parrish, 1998) to be indicative of a strongly arid climate, such as deeply sunken, strongly protected stomata, stomata in grooves, exceedingly thick cuticles, or stomatal plugs. Rather, their epidermal features are close to those of the plants, including the ferns, of the contemporary riparian communities, indicating that they too inhabited a similar climate. Thus, when all of the evidence provided by the plants is considered it is apparent that both the riparian and upland communities lived under a humid subtropical to tropical climate that could have been monsoonal or consistently wet.

In conclusion, the theory of Demko, et al. (1998) that the ferns and other compressed plant fossils in the lower part of the Chinle Formation reflect only the humid climate along streams and around lakes and not the climate of the surrounding region ignores important facts. First, the climate along modern rivers is little different than that of the region they pass through and this was most probably true in the past. Second, humid climates simply could not and do not exist in narrow corridors along water courses or around lakes under an otherwise arid or strongly seasonal climate as visualized by Demko, et al. (1998). Although large trees may grow along the banks of some rivers in arid regions such as the Sonoran Desert the climate along such corridors is closely similar to that of the surrounding region and the trees obtain water from the water table. Furthermore, ferns can not and do not survive along the Rio Grande, the Colorado, the Nile, the Darling, and other rivers passing through arid lands. Not even desert ferns live along the water courses. Instead they inhabit certain specialized microhabitats. For example, the desert fern Notholaena parryi typically grows in crevices and under rocks in the Colorado desert (Nobel, 1978) and Cheilanthes mysurensis and related species grow on gravelly or rocky substrata in the deserts of India (Nayar, 1962).

CONCLUSIONS

Ten species of ferns and fern-like fronds are now known from the lower part of the Chinle Formation in Petrified Forest National Park and elsewhere in the southwestern United States. The ferns represent nearly one-fifth of the plant species based on megafossils in the Chinle flora and are assigned to four living and two extinct families; several are unassigned. Using both their morphology and the climatic preferences of their nearest living relatives it is evident that the Chinle ferns lived under a humid subtropical to tropical climate. The climate could very well have been monsoonal since the nearest living relatives of two of the ferns now inhabit monsoonal southeast Asia and more distant relatives of the other ferns also live there, in addition to other parts of the tropics. These conclusions are also corroborated by many of the other plant fossils found in the Chinle Formation in the park and elsewhere.

ACKNOWLEDGEMENTS

I acknowledge with thanks all of the support I have received over the years from the superintendents and staff of Petrified Forest National Park. Also, I am grateful to the Petrified Forest Museum Association for financial support.

REFERENCES

Arnold, C.A., 1947. An introduction to paleobotany. New York, McGraw-Hill Book Co., 433 p.

____, 1956. Fossil ferns of the Matoniaceae from North America: Journal of the Palaeontological Society of India, 1: 118-121.

Ash, S.R., 1970. Ferns from the Chinle Formation (Upper Triassic) in the Fort Wingate area, New Mexico. U. S. Geological Survey Professional Paper 613-D: 1-40.

____, 1972a. Plant megafossils of the Chinle Formation, in, W.J. Breed and C.S. Breed, (eds), Investigations in the Triassic Chinle Formation. Museum Northern Arizona Bulletin, 47: 23-43.

____, 1972b. The search for plant fossils in the Chinle Formation, in, W.J. Breed and C.S. Breed, (eds), Investigations in the Triassic Chinle Formation. Museum Northern Arizona Bulletin 47: 45-58.

____, 1972c. Upper Triassic Dockum flora of eastern New Mexico and Texas. New Mexico Geological Society, 23rd Annual Field Conference, east-central New Mexico, Guidebook, p. 124-127.

____, 1972d. Marcouia gen. nov., a problematical plant from the Late Triassic of the southwestern U.S.A. Palaeontology, 15: 423-429.

____, 1973. Two new Late Triassic plants from the Petrified Forest of Arizona. Journal of Paleontology, 47: 46-53.

____, 1975. The Chinle (Upper Triassic) flora of southeastern Utah. Four Corners Geological Society Guidebook, 8th Annual Field Conference, Canyonlands, p. 143-147.

____, 1980, Upper Triassic floral zones of North America, in, D.L. Dilcher and T.M. Taylor, (eds), Biostratigraphy of fossil plants. Dowden, Hutchinson and Ross, Inc., Stroudsburg, PA., p. 153-270.

____, 1987. Petrified Forest National Park, Arizona, in, S. S. Beus, (ed), Centennial Field Guide, Volume 2, Rocky Mountain Section of the Geological Society of America, p. 405-410.
____, 1989. A catalog of Upper Triassic plant megafossils of the western United States through 1988, in, S.G. Lucas and A.P. Hunt, (eds), Dawn of the Age of Dinosaurs in the American Southwest. New Mexico Museum of Natural History, Albuquerque, p. 189-222.

____, 1994. Donwelliacaulis chlouberii gen. et sp. nov. (Guaireaceae, Osmundales) one of the oldest Mesozoic plant megafossils in North America. Palaeontographica, 234B:1-17.

____, 1997. Evidence of arthropod-plant interactions in the Upper Triassic of the southwestern United States. Lethaia, 29: 237-248.

____, 1999. An Upper Triassic Sphenopteris showing evidence of insect predation from Petrified Forest National Park, Arizona. International Journal of Plant Sciences, 160: 208-215.

____, 2000. Evidence of oribatid mite herbivory in the stem of a Late Triassic tree fern from Arizona. Journal of Paleontology, 74: 1065-1071.

____ and G. Creber, 1992. Palaeoclimatic interpretation of the wood structures of the trees in the Chinle Formation (Upper Triassic) in the area of Petrified Forest National Park, Arizona, U.S.A. Palaeogeography, Palaeoclimatology, Palaeoecology, 96: 299-317.

____, R. Litwin, and A. Traverse, 1982. The Upper Triassic fern Phlebopteris smithii (Daugherty) Arnold and its spores. Palynology, 6: 203-219.

Blakey, R.C., 1989. Triassic and Jurassic geology of the southern Colorado Plateau, in, J.P. Jenny, S.J. Reynolds, (eds), Geologic evolution of Arizona. Arizona Geological Society Digest, 17: 369-396.

Bower, F.O., 1928. The ferns (Filicales), Volume III. The Leptosporangiate ferns. Cambridge University Press, Cambridge. 306 p.

Daugherty, L.H., 1941. The Upper Triassic flora of Arizona. Carnegie Institution of Washington Publication, 526:1-108.

____, 1960. Itopsidema, a new genus of the Osmundaceae from the Triassic of Arizona. American Journal of Botany, 47: 771-777.

Demko, T.M., R.F. Dubiel,and J.T. Parrish, 1998. Plant taphonomy in incised valleys: Implications for interpreting paleoclimate from fossil plants. Geology, 26: 1119-1122.

Dubiel, R.F., S.T. Hasiotis, and T.M. Demko, 1999. Incised valley fills in the lower part of the Chinle Formation, Petrified Forest National Park, Arizona: Complete measured sections and regional stratigraphic implications of Upper Triassic rocks, in, V.L. Santucci and L.
McCelland, (eds), National Park Service Paleontological Research, 4: 78-84.

Helvy, R.H., 1963. Adaptations of cheilanthoid ferns to desert environments. Journal of the Arizona Academy of Sciences, 2: 164-175.

Lellinger, D.B., 1985. A field guide of the fern and fern-allies of the United States and Canada. Smithsonian Institution Press, Washington, D.C. 389 p.

Litwin, R.J., 1984. Fertile organs and in situ spores of ferns from the Late Triassic Chinle Formation of Arizona and New Mexico, with discussions of associated dispersed spores: Review of Palaeobotany and Palynology, 42:101-146.

____, 1986. The palynostratigraphy and age of the Chinle and Moenave Formations, southwestern U.S.A. PhD thesis, Pennsylvania State University, 256 p.

____, A. Traverse and S.R. Ash, 1992. Preliminary palynological zonation of the Chinle Formation, southwestern U. S. A., and its correlation with the Newark Supergroup (eastern U. S. A.). Review of Palaeobotany and Palynology, 68: 269-287.

Marsh, A.S., 1914. The anatomy of some xerophilous species of Cheilanthes and Pellaea. Annals of Botany, 28: 671-684.

Nayar, B.K., 1962. The morphology of some species of Cheilanthes. Journal of the Linnean Society (Botany), 58: 449-460.

Nobel, P.S., 1978. Microhabitat, water relations, and photosynthesis of a desert fern, Notholaena parryi. Oecologia, 31: 293-309.

Parrish, J.T., 1998. Interpreting pre-Quaternary climate from the geologic record. Columbia University Press, New York City. 338 p.

Stewart, J.H., F.G. Poole, and R.F. Wilson, with a section on Sedimentary petrology by R.A. Cadigan, and a section on Conglomerate studies by W. Thordarson, H.F. Albee, and J.H. Stewart, 1972. Stratigraphy and origin of the Chinle Formation and related Upper Triassic strata in the Colorado Plateau region. U.S. Geological Survey Professional Paper, 690:1-336.

Tidwell, W.D. and S.R. Ash, 1995. A review of selected Triassic to Early Cretaceous ferns. Journal of Plant Research, 107: 417-442.

Tryon, R.M., 1982. Pteridophytes, in, H. Leith and M.J. H. Werger (eds), Tropical Rain Forests of the world. Ecoobelsystems of the world, v. 14B. Elsevier, Amsterdam. 346 p.

____, 1986. The biogeography of species, with special reference to ferns. Botanical Review, 52:117-155.