Amer. J. Bot. 72(11): 1765-1774. 1985.

STUDIES OF PALEOZOIC FUNGI. IV. WOOD-DECAYING FUNGI IN CALLIXYLON NEWBERRYI FROM THE UPPER DEVONIAN1

SARA P. STUBBLEFIELD,2 THOMAS N. TAYLOR,2 AND CHARLES B. BECK3 2Departm.ent of Botany, The Ohio State University, Columbus, Ohio 43210; and 'Museum of Paleontology and Division of Biological Sciences, The University of Michigan, Ann Arbor, Michigan 48109

ABSTRACT An extensive fungal infection is present in silicified specimens ofCallixylon newberryi (Daw­ son) Elkins and Wieland from the Upper Devonian of Indiana. Fungi are represented by branched, septate filaments with both terminal and intercalary chlamydospores. Clamp con­ nections were not observed. Tracheids of the secondary xylem, decayed in varying degrees, show erosion troughs, cavities, and extensive lysis of tracheid walls. The pattern of decay is most similar to modem-day white rots, which are typically caused by basidiomycetes, while hyphal features are characteristic of either ascomycetes or basidiomycetes. The presence of wood decay coupled with a septate mycelium suggests the presence of higher fungi in the Devonianandindicatesa more diverse Devonian mycoflora than has previously been recognized.

THE DEVONIAN MYCOFLORA is poorly known. earliest known example ofwood decay caused Although a few fungi have been reported from by fungi, andalso suggests the presence ofhigher marine environments (e.g., Fry and McLaren, fungi in the Upper Devonian. 1959; Taylor, 1971), the mycoflora associated with early land plants is known almost entirely from silicified specimens in the Rhynie Chert MATERIALS AND METHODs-Fungi occur in (Kidston and Lang, 1921; Harvey, Lyon, and the secondary xylem of a block of silicified Lewis, 1969; Boullard and Lemoigne, 1971; Callixylon newberryi collected 5 mi northeast Illman, 1984) and similar calcified material ofHenryville, Clark County, IN. The fragment from the Gaspe Peninsula (Stubblefield and ofwood comes from the upper portion of the Banks, 1983). In the present paper an addi­ Blackiston Formation in the NewAlbany Shale. tional occurrence of Devonian fungi is docu­ This part ofthe formation is Upper Devonian mented. An extensive fungal infection is de­ in age and corresponds to the European Fa­ scribed in the tracheids and ray cells of the mennian stage (Collinson, 1967). The speci­ Upper Devonian progymnosperm Callixylon men is no. 13834 in the Museum of Paleon­ newberryi (Dawson) Elkins and Wieland. Hy­ tology, The University of Michigan-Ann phae were first noted in this material by Arnold Arbor. Peels and thin sections from this spec­ (1931), who published a single light micro­ imen bear nos. 13834-1 to 13834-20 in the graph of infected tracheids during the course same collection. of an anatomical investigation of Callixylon Material was prepared for light microscopy newberryi.Reinvestigation ofmaterial from this both as ground thin sections and as cellulose locality offers the opportunity to study both acetate peels in which concentrated hydro­ the earliest known fungi in wood as well as the fluoric acid was substituted for hydrochloric acid. In preparation for scanning electron mi­ croscopy (SEM), specimens were etched for 2­ I Received for publication 3 December 1984; revision accepted 14 June 1985. 4 min with concentrated hydrofluoric acid and The authors with to thank Dr. Roland L. Seymour of rinsed in water until neutral. Material was then the Department of Botany, The Ohio State University­ dried in a critical point dryer, mounted on Columbus, and Dr. Robert L. Shaffer of the Herbarium, standard stubs, and coated with both carbon The University of Michigan-Ann Arbor for much useful discussion. This research was supported in part by funds and gold. The use of carbon significantly re­ from the NSF (DEB-8213060). duced charging (Skog, 1983). We wish to dedicate this paper to our colleague, the late Dr. Charles E. Miller of Ohio University, Athens, OH who DESCRIPTION - The infected specimen is a worked closely with us in several studies on fossil fungi. Chuck Miller's cherished friendship, and his enthusiasm block of secondary xylem approximately 100 and in-depth understanding offungi will continue to serve cm ' in size from the progymnosperm Callix­ as an inspiration for our work. ylon newberryi. Due to the fragmentary nature 1765 1766 AMERICAN JOURNAL OF BOTANY [Vol. 72

Fig. 1-7. Fungal hyphae in Callixylon newberryi. 1. Tangential section of tracheids with fungal hyphae. x 430 . 20 Tracheids containing fungal hyphae; note abundance of fungi. x 200. 30Terminal chlamydospores. x 2,600 . 40Septate hyphae (arrow) with terminal chlamydospore germinating at three points. x 2,600. 50 Radial section ofray cells on left and tracheid on right. Note continuity of hyphae between ray and tracheid (arrow) and small projections from hyphae . x 600. 6. Branched, septate hypha . Arrow at pseudoseptum formed by break . x 700. 7. Hypha with intercalary swelling x 2,600. November, 1985] STUBBLEFIELD ET AL. - WOOD- DECAYING FUNGI IN CALLIXYLON 1767 of the specimen, the original extent of the in­ grooves (Fig. 17 at arrows), oval cavities (Fig. fection cannot be determined. Fungal hyphae 14), and severely degraded wall surfaces (Fig. are abundant in both tracheids and ray cells, 12, 13). The walls of ray cells appear to be and in each case they are generally oriented unmodified. parallel to the long axis ofthe infected cell (Fig. An additional distinctive feature ofthis wood 1, 2, 5, 24-27). In some instances entire ray is the contents ofthe ray cells. Three types of cells may be occluded by hyphae. bodies are present. In transmitted light the most Hyphal morphology varies, particularly in obvious structures are dark brown or amber, respect to size, branching, and surface texture. and spherical, elongate, or irregular in shape Hyphae range from 1".5-6 /-tm in diam. In (Fig. 15,20,21). Spherical bodies range from ground sections they often appear flat and rib­ 1.8-50.0 /-tm in diam. They are often solid, but bonlike (Fig. 1). Twists in the filament create single hollow specimens and vesicular bodies the superficial appearance of changes in di­ are also prevalent. A smooth surface is most ameter, although hyphal swellings do occur(Fig. commonly observed, although rough spheres 9). Septae are present (Fig. 1, 6), although many are also present. These structures appear to be apparent septae are nothing more than breaks of the same material as closely associated ir­ in the filament (Fig. 4 and 6 at arrow). Fre­ regular meshlike deposits on the inner surfaces quently branched hyphae are common (Fig. 1, ofray cells. The second type ofcellular inclu­ 2,6) and sometimes intertwine forming a dense sion is distinguishable from the first only by mass within infected cells (Fig. 1, 2). Hyphal the presence of a clear surrounding zone branches are sometimes narrower than the par­ bounded by a layer with a rough outer surface ent filament (Fig. 6), but dichotomous branch­ (Fig. 15, 16, 18,21). The thickness ofthis trans­ ing is present. Other hyphae are unbranched parent layer varies, reaching 0.36 /-tm. This for a considerable length except for short, nar­ layer also encloses rods and irregularly shaped row lateral extensions (Fig. 5). In some cases pieces of the brown-amber material (Fig. 21). these hyphae produce a highly divided lateral The third type ofcellular inclusion consists of branch (Fig. 5 at bottom). In transmitted light, spherical bodies 10-23 /-tm in diam, which often hyphal surfaces are either smooth (Fig. 1-7) or appear collapsed (Fig. 22, 23). All three types rough (Fig. 19, 20). When scanned, even the are found primarily in the ray cells (Fig. 15, smoothest surfaces are uniformly rough (Fig. 16, 18,22), although some have been observed 8-10), while rougher hyphae bear discrete in tracheids (Fig. 23). rounded knobs along their surfaces (Fig. 11). Both single intercalary swellings (Fig. 5, 7, 9 DISCUSSION - The discovery of a well-de­ at arrow, 26), and single (Fig. 4) and multiple veloped mycelium coupled with degraded tra­ terminal swellings (Fig. 3) occur. Typically, such cheids in Devonian wood raises a number of swellings branch at several points (Fig. 4, 7, interesting questions concerning the nature of 9). the interaction between the fungus and the vas­ The tracheid walls ofinfected Callixylon are cular plant, the nature ofthe fungus itself, and particularly interesting and show evidence of fungal evolution. In many instances involving fungal modification. Preservation is generally fossil fungi it is impossible to determine the excellent, and the inner walls are often nearly kind of relationship that existed between the smooth or slightly cracked (Fig. 8, 9, 12 at left). fungus and the plant with which it was asso­ However, many tracheids are extensively de­ ciated. However, information about the role cayed. Decay features include long, narrow ofthe fungus in Callixylon is suggested by the

Fig. 8-14. 8-11. Fungal hyphae. 12-14. Degraded tracheid walls. 8. Branching hyphae on inner wall oftracheid. Note relationship between hyphae and pits on tracheid wall. x 2,500. 9. Branching hyphae in two adjacent tracheids. Note intercalary swelling (arrow) and encrusting material on hyphae. x 1,200. 10. Hyphae with rough surfaces. x 600. 11. Hyphae with rounded bodies on surfaces. x 5,500. 12. Tracheid with degraded wall. Compare with intact wall containing longitudinal crack on left. x 2,000. 13. Degraded wall of tracheid showing irregular pattern caused by extensive lysis. x 2,000. 14. Tracheid with ellipsoidal regions of degradation. x 1,500.

Fig. 15-20. 15, 16, 18. Cell contents. 17. Degraded cell wall. 19, 20. Hyphal morphology. 15. Ray parenchyma with globular contents. Note hyaline regions around some spheres. x 660. 16. Note surface texture and variability in shape in contents of ray cells. x 800. 17. Inner surface of tracheid wall illustrating pits, crack, and narrow grooves (arrow). x 1,500. 18. Globules in ray cells. Note that globules contain an outer, clearly defined wall. x 900. 19. Rough fungal hyphae with closely associated spherical inclusions. x 2,000. 20. Rough fungal hyphae in ray parenchyma. x 2,000. 1768 AMERICAN JOURNAL OF BOTANY [Vol. 72 November, 1985] STUBBLEFIELD ET AL.-WOOD-DECAYING FUNGI IN CALLIXYLON 1769 1770 AMERICAN JOURNAL OF BOTANY [Vol. 72

Fig. 21-27. 21. Cell contents. 22, 23. Sporelike bodies and chlamydospores. 24, 25, 27. Penetration oftracheid walls by fungal hyphae. 21. Contents in ray parenchyma. Note hyaline regions around rodlets (arrows) and spheres. x 330. 22. Sporelike bodies in ray cells. x 525. 23. Sporelike body in tracheid. x 670.24. Fungal hyphae showing possible continuity (arrows) between ray cell and tracheid. x 650. 25. Fig. 24 in different plane of focus. Hypha at left (arrow) passes between ray cell and tracheid. x 650. 26. Hypha with chlamydospores. x 1,700. 27. Tracheid and adjacent ray cell. Hyphae in tracheid apparently radiate from a single point adjacent to ray cell. x 650. presence of degraded tracheids. Several types ofwood-decaying fungi involved, and the bio­ of wood decay effect extant woods, including chemical activity of the fungi (Montgomery, soft rots, white rots, and brown rots, each of 1982). Because chemical tests are not presently which differs in the pattern ofdecay, the types feasible with permineralized woods, morpho- November, 1985] STUBBLEFIELD ET AL.-WOOD-DECAYING FUNGI IN CALLIXYLON 1771 logical features, especially ofthedegraded walls ery, 1971; Jutte and Zabel, 1974; Levy, 1974; and the fungi themselves, provide the only in­ Eriksson et aI., 1980). formation concerning the type ofwood decay. The means by which hyphae penetrate the In the past 20 yr, scanning electron micros­ tracheids and ray cells ofCallixylon is notclear. copy has been widely used to study decaying The fungal infection is extensive, and the ma­ woods and has contributed significantly to the jority ofcells contain fungi. In many instances, recognition ofdifferent types ofdecay in mod­ cells, particularly those ofthe ray parenchyma, em woods (e.g., Findlayand Levy, 1969; Brav­ are completely occluded. The degree of infec­ ery, 1971; Jutte and Zabel, 1974; Eriksson et tion is easily underestimated in SEM prepa­ aI., 1980). Soft rots, first recognized by Savory rations since fungi are often lost during the (1954), are typically caused by ascomycetes or etching process when anentire column ofsilica, imperfect fungi. The invading fungus forms which encases fungi, falls from a cell cavity. chains of cavities with pointed ends that par­ However, despite the fact that the fungi have allel the cellulose microfibrils in the S2 layer ramified throughout the entire block ofxylem, ofthe cell wall, and it penetrates adjacent cells it is unusual to see hyphae passing from cell by the formation offine hyphal strands termed to cell. There is no indication that they passed proboscishyphae (Corbett, 1965; Jutte and Za­ through pits, although this is a common route bel, 1974). White and brown rots are both for white rot fungi (Bravery, 1971). In fact, in caused by basidiomycetes. White rot fungi de­ all our SEM preparations, hyphae pass over, cay the secondary wall by the production of but not through the pits (e.g., Fig. 8). Bore holes erosion troughs along the path of the hyphae, are not obvious, but they are sometimes un­ which eventually coalesce and severely erode common in modem examples of white rot as the wall (Bravery, 1971). Brown rots produce well (Bravery, 1971). Several observations sug­ an amorphousdecay, which has been described gest that hyphae pass from rays to tracheids. as looking "more like expanded polystyrene or In the most convincing cases, a hypha in a foam rubber than the normal smooth texture tracheid seems to be continuous with a struc­ of the unattacked walls" (Levy, 1982). Phy­ ture in an adjacent ray cell (Fig. 5, 24, 25). In comycetes may also be found in wood as "pri­ other cases, the only hyphae in a tracheid ap­ mary moulds" that live from simple carbo­ pear to originate from a single point next to a hydrates derived from the ray parenchyma or ray cell (Fig. 5, 27). It is possible, therefore, soil, but they do not have the capacity to pen­ that the rays provide the main passageway for etrate wood except through natural openings the movement offungal mycelium throughout or cavities caused by other organisms and are the wood. Since all tracheids are probably in not regarded as wood destroyers (Levy, 1982). contact along their lengths with one or more In the present material the pattern of decay rays, all tracheids in a region of wood could is unlike that typically associated with brown contain the fungus without any hyphal pene­ rot. Furthermore, it differs from most soft rots. tration through the walls of contiguous tra­ Hyphae are not generally found within the tra­ cheids. cheid walls, and although pointed cavities are The fungus in Callixylon is interesting in present, they are produced only in the inner several respects. The infected wood contains surface layers ofthe wall and their orientation the most well-preserved hyphae presently differs from that produced by soft rot fungi. In known from the Devonian, and the oldest addition, the small cavities caused by probos­ wood-inhabiting fungus. The septate myce­ cis hyphae have not been found, although they lium is both extensively developed and well may be difficult to distinguish from small cracks preserved. Clamp connections were carefully resulting from processes associated with fos­ sought, but none were demonstrated to our silization. Hyphae typically extend along the satisfaction, although occasional bulges along inner wall surface of the tracheid and pass hyphal margins (see Fig. 6) may sometimes be through the cell lumen. Thepattern ofprobable suggestive of clamps. Hyphae are also char­ erosion troughs on decayed tracheid walls is acterized by terminal and intercalary chla­ consistent with the coalescing ofthese troughs. mydospores. Germination of single terminal Severe lysis is also evident in some tracheids. spores, sometimes from more than one point Thesefeatures suggest thatthe Callixylon wood (Fig. 4, 7), may result in the continuation of was infected by a fungus ofthe white rot type, the filament. Similar structures are known from and the involvement ofbasidiomycetes is im­ extant ascomycetes and basidiomycetes. The plied. Several authors have provided excellent form of the hyphae and the presence of wood scanning electron micrographs ofwhite rot de­ decay suggests that the mycelium is not phy­ cay in several different extant woods, which comycetous but rather is that ofa higherfungus may be examined for comparison (e.g., Brav- (ascomycete or basidiomycete). However, in 1772 AMERICAN JOURNAL OF BOTANY [Vol. 72 the absence ofclamp connections it is not pos­ tissue. These bodies are spatially associated sible to differentiate ascomycete and basidio­ with the Hartig net. Among other possibilities, mycete on the basis of hyphal morphology Wills and Cole (1978) suggest that the bodies (Roff, 1964), and neither ascocarps nor basid­ may be an adcrusting deposit similar to that iocarps have been observed. reported in wheat caryopsis (Zee, 1975), al­ The surface texture ofthe hyphae is another though they are ultimately uncertain about their interesting feature. Considerable variability in origin and function. The accumulation of res­ surface morphology exists in extant hyphae. inous material in tracheids and ray paren­ For example, among hyphae with a rough tex­ chyma has also been observed in other in­ ture are the terminal hairs of thermophilic stances of wood decay. Blanchette (1979) species of Chaetonium Kunze ex Fr. (Hawks­ describes this in the degradation ofseveral types worth and Wells, 1973; Ellis, 1981). Hawks­ of gymnosperms by Phellinus pini (Thore ex worth and Wells (1973) described 10 types of Fr.) A. Ames and suggests that this appears to ornamentation, several ofwhich are similar to be a host response to infection. The conspic­ that seen among the roughest hyphae in Cal­ uous nonfungal bodies in the ray cells of Cal­ lixylon, but it appears that in Chaetonium this lixylon newberryi can also be hypothesized to ornamentation is not a taxonomically useful be a response to fungal infection. However, feature (Ellis, 1981). Similar excrecences can observations ofapparently uninfected wood of also be found on conidia of several imperfect this species from three different localities do fungi (Reisingerand Kiffer, 1974) andon perid­ not support this hypothesis since the rays in ial hyphae and appendages of Pseudogym­ these wood samples also contain similar in­ noascus roseus Raillo (Tsuneda, 1982). In each clusions. These observations do not nullify the case they are producedby the fungus. However, hypothesis, however, since a fungal infection hyphae mayalso be encrusted with degradation some distance away might have had a phys­ products in decaying wood, and it is interesting iological effect in the region ofwood sectioned. that the roughest hyphae occur in the ray cells The third type of structure is found in both that contain large amounts ofpossible ergastic the ray cells and tracheids but is much less material. Although there is always the possi­ common than the first two. They are generally bility of more than one type of fungus in a spherical, but are often wrinkled, as though piece ofdecayed fossil wood, the variability in they may have shrunken or collapsed during surface morphology among the hyphae does fossilization. It is possible that some of these not necessarily imply a variety offungal types bodies are fungal reproductive structures. Un­ in the present material. fortunately, they are always discrete, and there The contents of ray parenchyma cells, first is no evidence of germination. Nevertheless, noted by Arnold (1931), is probably the most chlamydospores have been found in attach­ distinctive and puzzling feature of this wood. ment to hyphae, and it would not be unrea­ Ofthe three types ofbodies found in these cells, sonable to find isolated spores in eitherthe rays the first two types are highly variable in size or tracheids. Alternatively, these structures may and shape and, therefore, are probably non­ be related to the hyaline regions surrounding fungal despite the resemblance of the smaller many spheres, although they are typically thin­ spherical bodies to spores. On the basis ofcom­ ner walled than these structures, apparently parison with permineralized Carboniferous smooth, and more numerous in the tracheids fossils, this material may be an ergastic product than dark spheres. or resinous compound. The origin ofthe hya­ The fossil record has traditionally provided line layer around certain bodies is unclear. little information concerning the evolution of However, because it encases material of all fungi (Bartnicki-Garcia, 1970). Several authors sizes and shapes it is unreasonable to regard it have suggested an early, possibly Precambrian, as a fungal spore or a hyphal sheath. origin for primitive fungi (e.g., Tyler and Barg­ The search for modem analogs is inconclu­ hoorn, 1954; Pflug, 1964; Schopf and Barg­ sive. However, similar bodies have been re­ hoorn, 1969; Timofeev, 1970; Pflug and von ported from root cells of mycorrhizal Pinus Klopotek, 1978), while others question these mugo (Wills and Cole, 1978). These structures, reports (Tiffney, pers. comm.). However, by referred to as "vesicular bodies" (Wills and the end of the Paleozoic major fungal groups Cole, 1978), are highly variable in size and are represented, but the changes that occurred shape. Large vesicles sometimes appear to be between the Precambrian and the end of the formed by the coalescence ofsmaller ones. As Paleozoic are poorly documented. Our present in the present material, some are smooth while knowledge ofthe Devonian mycoflora suggests most are rough, closely resembling the warty the existence of oomycetes and zygomycetes appearance of the fungal hyphae in the same (Kidston and Lang, 1921; Butler, 1938), which November, 1985] STUBBLEFIELD ET AL.-WOOD-DECAYING FUNGI IN CALLIXYLON 1773 may be endomycorrhizal (Boullard and Le­ LITERATURE CITED moigne, 1971; Pirozynski and Malloch, 1975; ARNOLD, C. A. 1931. On Callixylon newberryi (Dawson) Wagner and Taylor, 1982) as well as possible Elkins et Wieland. Contrib. Mus. Paleontol. Univ. chytrids or hyphochytrids (Illman, 1984). How­ Mich. 3(12): 207-232. ever, the higher fungi are not customarily as­ BANKS, H. P. 1980. Floral assemblages in the Siluro­ sociated with the Devonian. The first generally Devonian. In D. Dilcher and T. N. Taylor [eds.], accepted basidiomycetes are known from Biostratigraphy of fossil plants, pp. 1-24. Dowden, Pennsylvanian hyphae with clamp connections Hutchinson and Ross, Stroudsburg, Pa. BARTNICKI-GARCIA, S. 1970. Cell wall composition and (Dennis, 1970), while there has been little other biochemical markers in fungal phylogeny. In J. agreement concerning the time oforigin ofthe B. Harborne [ed.], Phytochemical phylogeny, pp. 81­ ascomycetes. While some argue for the pres­ 103. Academic Press, London. ence of Carboniferous ascomycetes (Stubble­ BLANCHETTE, R. A. 1979. Wood decomposition by Phel­ field and Taylor, 1983a, b; see also Tiffney and linus (Fomes) pini: a scanning electron microscopy study. Can. J. Bot. 58: 1496-1503. Barghoorn, 1974), others are skeptical of past BOULLARD, B., AND Y. LEMOIGNE. 1971. Les champi­ reports and have suggested that undisputed as­ gnons endophytes du "Rhynia gwynne-vaughanii" K. comycetes have not been found in the Paleo­ et L. Botaniste 14: 49-89. zoic (Pirozynski, 1976). Recently, however, BRAVERY, A. F. 1971. The application of scanning elec­ Sherwood-Pike and Gray (1985) have docu­ tron microscopy in the study oftimber decay. J. Inst. mented the existence ofprobable ascomycetes Wood Sci. 5: 13-19. BUTLER, E. J. 1938. The occurrences and systematic po­ from the Silurian ofGotland. The fossil record sition of the vesicular-arbuscular type ofmycorrhizal ofthe Devonian mycoflora, then, consists pri­ fungi. Trans. Br. Mycol. Soc. 22: 274-301. marily ofa very few reports oflower fungi. We COLLINSON, C. 1967. Devonian of the north central re­ believe that this this is an inaccurate reflection gion of the United States. In D. H. Oswald [ed.], offungal diversity during the Devonian and a International symposium on the Devonian system, poor indication of the geological range of the pp. 933-971. Can. Soc. Pet. Geol., Calgary, Alberta. higher fungi. The morphological and taxo­ CORBETT, N. H. 1965. Micro-morphological studies on the degradation of lignified cell walls by ascomycetes nomical complexity exhibited by the late De­ and fungi imperfecti. J. Inst. Wood Sci. 14: 18-29. vonian vascular flora suggests the availability DENNIS, R. L. 1970. A Middle Pennsylvanian basidio­ of a variety of substrates and a high potential mycete mycelium with clamp connections. Mycologia for plant/fungi interactions. Vascular land 62: 578-584. plants arose by the end of the Silurian, sec­ ELLIS, D. H. 1981. Ascocarp morphology and terminal ondary xylem had evolved by the Eifelian, and hair ornamentation in thermophilic Chaetomium species. Mycologia 73: 755-773. arborescent plants with extensively developed ERIKSSON, K., A. GRUNWALD, T. NILLSON, AND L. VAL­ secondary xylem and phloem were present by LENDER. 1980. A scanningelectron microscopy study the Givetian (Banks, 1980). Furthermore, of the growth and attack on wood by three white-rot spores, broad megaphylls, periderm and seeds, fungi and their cellulase-less mutants. Holzforschung all substrates for fungi today, existed by the 34: 207-213. close of the Devonian. In light of the varied FINDLAY, G. W. D., AND J. F. LEVY. 1969. Scanning electron microscopy as an aid to the study of wood interactions that exist between fungi and vas­ anatomy and decay. J. Inst. Wood Sci. 23: 57-63. cular plants today, it would not be surprising FRY, W. L., AND D. J. McLAREN. 1959. Fungal filaments to find parasites, pathogens, leaf-inhabiting in a Devonian limestone from Alberta. Can. Geol. fungi, mycorrhizae, or saprophytes, including Surv. Bull. 48: 1-9. wood-decaying fungi far earlier than the mid­ HARVEY, R, A. G. LYON, ANDP. N. LEWIS. 1969. Afossil fungus from Rhynie chert. Trans. Br. Mycol. Soc. 53: dle Pennsylvanian. 155-157, PI. 16. Speculation on the origin ofhigher fungi has, HAWKSWORTH, D. L., AND H. WELLS. 1973. Ornamen­ of necessity, been based on modern forms. tation on the terminal hairs in Chaetomium Kunze However, the fossil record may yet provide ex Fr. and some allied genera. Mycol. Pap. 134: 1­ valuable data concerning the time of origin of 24. major groups and their eventual diversifica­ ILLMAN, W. I. 1984. Zoosporic fungal bodies in the spores tion.The present material is particularly en­ of the Devonian fossil vascular plant, Horneophyton. Mycologia 76: 545-547. couraging in suggesting the existence of De­ JUTTE, S. M., AND R A. ZABEL. 1974. Initial wood decay vonian fungi, probably ascomycetes or stages as revealed by scanning electron microscopy. basidiomycetes, which interacted with early Scanning electron microsc. 1974 (Part II): 445-452. land plants in a manner similar to modern-day KIDSTON, R., AND W. H. LANG. 1921. On Old Red Sand­ fungi. This material also leads us to expect stone plants showing structure, from the Rhynie Chert greater diversity in the Devonian mycoflora Bed, Aberdeenshire. V. The Thallophyta occurring in the peat-bed; the succession of the plants through a and suggests that the earliest land plants may vertical section of the bed, and the conditions of ac­ be a fruitful source of information concerning cumulation and preservation ofthe deposit. Trans. R the origin of fungi adapted to life on land. Soc. Edinburgh 52: 855-902, PI. I-X. 1774 AMERICAN JOURNAL OF BOTANY [Vo1. 72

LEVY, J. F. 1974. Fungi in wood. Scanning electron mi­ SKOG, J. E. 1983. Callixylon wood from the Greenland erose, 1974 (Part II): 461-468. Gap Group (Upper Devonian) of southwestern Vir­ ---. 1982. The place ofbasidiomycetes in the decay ginia. Va. J. Sci. 34: 58-64. ofwood in contact with the ground. In J. C. Frankland, STUBBLEFIELD, S. P., AND H. P. BANKS. 1983. Fungal J. N. Hedger, and M. J. Swift [eds.], Decomposer remains in the Devonian trimerophyte Psilophyton Basidiomycetes, pp. 161-178. Cambridge University dawsonii. Amer. J. Bot. 70: 1258-1261. Press, Cambridge, U.K. ---, AND T. N. TAYLOR. 1983a. Studies of Paleozoic MONTGOMERY, R. A. P. 1982. The role of poly sacchar­ fungi. I. The structure and organization of Traquairia idase enzymes in the decay of wood by basidiomy­ (Ascomycota). Amer. J. Bot. 70: 387-399. cetes. In J. C. Frankland, J. N. Hedger, and M. J. Swift ---, AND ---. 1983b. Studies of Carboniferous [eds.], Decomposer Basidiomycetes, pp. 51-65. Cam­ fungi. II. The structure and organization ofMycocar­ bridge University Press, Cambridge, U.K. pon, Sporocarpon, Dubiocarpon, and Coleocarpon PFLUG, H. D. 1964. Niedere Algen und ahnliche Klein­ (Ascomycotina). Amer. J. Bot. 70: 1482-1498. formen aus dem Algonkium der Belt-Serie. Ber. Ober­ TAYLOR, B. J. 1971. Thallophyte borings in phosphatic hess. Ges. Nat.-u. Heilk. 33: 403-411. fossils from the Lower Cretaceous ofsoutheast Alex­ ---, AND A. VON KLOPOTEK. 1978. Eucaryonten im ander Island, Antarctica. Palaeontology 14: 294-302. Archaikum? Oberhess. Naturwiss. Z. 44: 19-28. TIFFNEY, B. H., AND E. S. BARGHOORN. 1974. The fossil PIROZYNSKl, K. A. 1976. Fossil fungi. Annu. Rev. Phy­ record of the fungi. Occas. Pap. Farlow Herb. Cryp­ topathol. 14: 237-246. togam. Bot. Harv. Univ. 7: 42. ---, AND D. W. MALLOCH. 1975. The origin ofland TIMOFEEV, B. V. 1970. Une decouverte de phycomycetes plants: a matter ofmycotrophism. BioSystems 6: 153­ dans Ie Precambrien. Rev. Palaeobot. Palynol. 10: 79­ 164. 81. REISINGER, 0., AND E. KJFFER. 1974. Contribution to the TSUNEDA, A. 1982. Scanning electron microscopy of fungal microftora of the Congo. IV. Polyschema con­ Pseudogymnoascus roseus. Mycologia 74: 844-847. golensis sp. nov.: Taxonomy and ultrastructure. Trans. TYLER, S. A., AND E. S. BARGHOORN. 1954. Occurrences Br. Mycol. Soc. 62: 289-294. ofstructurally preserved plants in Precambrian rocks ROFF, J. W. 1964. Hyphal characteristics ofcertain fungi of the Canadian Shield. Science 119: 606-608. in wood. Mycologia 56: 799-804. WAGNER, C. A., AND T. N. TAYLOR. 1982. Fungal chla­ SAVORY, J. G. 1954. Breakdown of timber by Asco­ mydospores from the Pennsylvanian ofNorth Amer­ mycetes and Fungi Imperfecti. Ann. Appl. Bot. 41: ica. Rev. Palaeobot. Palynol. 37: 317-328. 336-347. WILLS, B.J., ANDA.L. J. COLE. 1978. A scanning electron SCHOPF, J. W., AND E. S. BARGHOORN. 1969. Microor­ microscopy study ofvesicular bodies' in mycorrhizal ganisms from the late Precambrian ofSouth Australia. roots of Pinus mugo (Turra). New Phytol. 80: 579­ J. Paleontol. 43: 111-118. 582. SHERWOOD-PIKE, M. A., AND J. GRAY. 1985. Silurian ZEE, S. Y. 1975. Scanning microscope observations of fungal remains: probable records of the class Asco­ the pigment strand in the caryopsis ofwheat. Aust. J. mycetes. Lethaia 18: 1-20. Bot. 23: 107-110.