Search

Coal balls are calcium carbonate accumulations that permineralized peat in paleotropical Permo-Carboniferous (∼320–250 Ma) mires. The formation of coal balls has been debated for over a century yet a widely applicable model is lacking. Two observations have been particularly challenging to explain: 1) the narrow temporal occurrence of coal balls and 2) their typical elemental (high Mg) and isotopic (low δ18O) composition that paradoxically indicate marine and freshwater origins, respectively. We evaluate a new model in which coal balls formed as CO2 escaped from the peat by diffusion through unusual air-filled spaces in the rootlets of lycopsid trees; critically, these trees were very common in paleotropical mires, and their evolutionary range matches the temporal range of coal balls. The episodic delivery of seawater and marine carbonate sediment to coastal mires is the first step in our model, although other pathways for the delivery of divalent cations are permissible. Subsequent dilution by freshwater and dissolution of these carbonates at the elevated CO2 of the mire subsurface is followed by the transport of CO2 through the rootlet airspaces and into the overlying water and atmosphere, which drives carbonate mineral precipitation in the sediment. We show that dilution by freshwater, because it minimally affects Mg/Ca ratios, can result in relatively low pore water δ18O values while allowing high-Mg calcite formation. This model thus explains the restriction of coal balls to the Permo-Carboniferous and resolves the discrepancy between elemental and isotopic compositions of coal ball carbonate minerals. Furthermore, we show with a 3D reactive transport model that CO2 could escape rapidly enough through the rootlets to fill 25 percent of the peat pore spaces with calcite before substantial burial (top decimeter of peat), explaining the exceptional preservation of coal swamp flora. Therefore, we suggest that coal balls are pedogenic in origin and that the disappearance of these pedogenically permineralized Histosols represents the first documented decrease in soil diversity on a vegetated planet. The rock record may thus provide important context for the modern loss and degradation of soils.

From the perspective of Phanerozoic time, coal balls are rare, apparently limited to a 24 m.y. interval (323–299 Ma) in the Pennsylvanian and earliest Permian. Yet within this interval, coal balls occur in many coals. Approximately 82 transgressive-regressive sedimentary cycles have been described for the Midcontinent, Illinois and Appalachian basins of North America during the mid-to-late Pennsylvanian. One third (27/82) have coal balls, including 57% of major cycles, 36% of intermediate cycles and 16% of minor cycles. Coal-ball occurrence in the Donets Basin is similar: over an interval of about 4 m.y. (~315–311 Ma, latest Bashkirian to mid-Moscovian), 39% (11/28) of transgressive-regressive cycles have coal balls. As North American paleoclimate became drier, tree ferns replaced lycopsids as the dominant plant in peat swamps at the Desmoinesian/Missourian boundary, and coal-ball occurrence declined. Overall, 47% of cycles (19/40) with lycopsid or cordaitean dominance have coal balls, whereas 19% (8/42) of cycles with tree fern dominance have coal balls (p This pattern may reflect the relative abundance of coal in the Late Pennsylvanian, with drier paleotropical climates in the Missourian-Virgilian leading to less paleotropical coal, fewer mines and a lower probability of discovering Missourian and Virgilian coal balls. However, average cycle duration decreased from 150 kyr in the Desmoinesian to ~100 kyr in the Missourian and Virgilian, and faster rates of relative sea-level rise could have affected coal-ball abundance, particularly if coal balls formed in marine swamps, with more rapid transgression leading to thinner coals, or to freshwater coals directly overlain by marine sediments. Rygel et al. (2008) suggested that Missourian-Virgilian cycles recorded more erosional relief than Desmoinesian cycles, consistent with increased amplitude and rates of sea-level rise during each glacial-eustatic cycle. Changes in Missourian-Virgilian cycles may have been augmented by stratigraphic attenuation as sediment filled the antecedent topography.

Petrographic analyses of 25 coal balls from well-studied paleobotanical profiles in the Middle Pennsylvanian Herrin Coal (Westphalian D, Illinois Basin) and five select coal balls from university collections, indicate that Herrin Coal-ball peats were permineralized by fibrous and non-fibrous carbonates. Fibrous carbonates occur in fan-like to spherulitic arrays in many intracellular (within tissue) pores, and are best developed in relatively open extracellular (between plant) pore spaces. Acid etched fibrous carbonates appear white under reflected light and possess a microcrystalline texture attributable to abundant microdolomite. Scanning electron microscopy, X-ray diffraction, and electron microprobe analysis demonstrate that individual fibers have a distinct trigonal prism morphology and are notable for their magnesium content (≈ 9–15 mol% MgCO 3 ). Non-fibrous carbonates fill intercrystalline spaces among fibers and pores within the peat as primary precipitates and neomorphic replacements. In the immediate vicinity of plant cell walls, non-fibrous carbonates cut across fibrous carbonates as a secondary, neomorphic phase attributed to coalification of plant cell walls. Dolomite occurs as diagenetic microdolomite associated with the fibrous carbonate phase, as sparite replacements, and as void-filling cement. Maximum dolomite (50–59 wt.%) is in the top-of-seam coal-ball zone at the Sahara Mine, which is overlain by the marine Anna Shale. Coal-ball formation in the Herrin Coal began with the precipitation of fibrous high magnesium calcite. The trigonal prism morphology of the carbonate fibers suggests rapid precipitation from super-saturated, meteoric pore waters. Carbonate precipitation from marine waters is discounted on the basis of stratigraphic, paleobotanical, and stable isotopic evidence. Most non-fibrous carbonate is attributable to later diagenetic events, including void-fill replacements, recrystallization, and post-depositional fracture fills. Evidence suggests that CO 2 degassing was important in coal-ball formation in the Herrin Coal, which mainly occurred sequentially upward with peat accumulation in the sites studied.

The mode of preservation dictates the preparation technique that will yield the most information about a specific fossil. Such considerations also include the time needed for preparation and degree of specimen destruction. Nowhere is this more clearly demonstrated than in the history of Carboniferous coal ball and chert research where the standard technique shifted from thin section to acetate peel preparations many years ago. Despite the ease and efficiency of acetate peels and the exponential increase in information they have provided about Carboniferous plants and ecosystems, we argue that there has been a concomitant decrease in attention directed at the microbial life also preserved in many cherts and coal balls. With this paper we endorse the use of thin sections, rather than peels, in order to study accurately the morphology and diversity of late Paleozoic microbial life.

Marie Stopes (1880-1958) is chiefly remembered as a birth control pioneer and sexologist, but in her twenties and thirties she carved out a highly successful career as a palaeobotanist and coal geologist. This paper outlines her early geological research on coal balls—carbonate concretions found within the Carboniferous coal seams of northern England, which preserve the remains of the peat-forming plants in beautiful anatomical detail. Stopes worked on coal balls during three intervals of her career. In the first phase (early 1903), she was Francis Oliver's postgraduate research assistant at University College London, during the critical period leading up to the ‘discovery of pteridosperms’ with D. H. Scott. Stopes's role was to hunt down key specimens in coal ball collections scattered across Britain. In the second phase (late 1904-1907), which followed a year of doctoral research in Munich, she grappled with the more broad-ranging questions of the origin of coal balls, their stratigraphic distribution, and the taphonomy and ecology of the plants they contained. This work took place while she was a Demonstrator in Botany at the Victoria University of Manchester, and was undertaken in collaboration with David Watson. Their findings transformed understanding of coal ball origins and remain influential today. In the third phase (1907-1911), she searched for coal balls in other countries and other stratigraphic intervals. She explored Japan for coal balls of Mesozoic age (1907-early 1909), and although unsuccessful in this particular endeavour, later she became one of the first geologists to locate Carboniferous coal balls in North America in 1911.

Quantitative plant assemblage data from coal balls, miospores, megaspores, and compression floras from the Calhoun coal bed (Missourian) of the Illinois Basin (USA) are used to interpret spatial and temporal changes in plant communities in the paleo-peat swamp. Coal-ball and miospore floras from the Calhoun coal bed are dominated strongly by tree ferns, and pteridosperms and sigillarian lycopsids are subdominant, depending on geographic location within the coal bed. Although the overall composition of Calhoun peat-swamp assemblages is consistent both temporally and spatially, site-to-site differences and short-term shifts in species dominance indicate local topographic and hydrologic control on species composition within the broader context of the swamp. Statistical comparison of the Calhoun miospore assemblages with those from other Late Pennsylvanian coal beds suggests that the same basic species pool was represented in each peat-swamp landscape and that the relative patterns of dominance and diversity were persistent from site to site. Therefore, it appears that the relative patterns of proportional dominance stayed roughly the same from one coal bed to the next during Late Pennsylvanian glacially-driven climatic oscillations.

Coal balls were collected from four coal beds in the southeastern part of the Illinois Basin. Collections were made from the Springfield, Herrin, and Baker coals in western Kentucky, and from the Danville Coal in southwestern Indiana. These four coal beds are among the principal mineable coals of the Illinois Basin and belong to the Carbondale and Shelburn Formations of late Middle Pennsylvanian age. Vegetational composition was analyzed quantitatively. Coal-ball samples from the Springfield, Herrin, and Baker are dominated by the lycopsid tree Lepidophloios, with lesser numbers of Psaronius tree ferns, medullosan pteridosperms, and the lycopsid trees Synchysidendron and Diaphorodendron. This vegetation is similar to that found in the Springfield and Herrin coals elsewhere in the Illinois Basin, as reported in previous studies. The Danville coal sample, which is considerably smaller than the others, is dominated by Psaronius with the lycopsids Sigillaria and Synchysidendron as subdominants. Coal balls from the Springfield coal were collected in zones directly from the coal bed and their zone-by-zone composition indicates three to four distinct plant assemblages. The other coals were analyzed as whole-seam random samples, averaging the landscape composition of the parent mire environments. This analysis indicates that these coals, separated from each other by marine and terrestrial-clastic deposits, have essentially the same floristic composition and, thus, appear to represent a common species pool that persisted throughout the late Middle Pennsylvanian, despite changes in baselevel and climate attendant the glacial–interglacial cyclicity of the Pennsylvanian ice age. Patterns of species abundance and diversity are much the same for the Springfield, Herrin, and Baker, although each coal, both in the local area sampled, and regionally, has its own paleobotanical peculiarities. Despite minor differences, these coals indicate a high degree of recurrence of assemblage and landscape organization. The Danville departs dramatically from the dominance–diversity composition of the older coals, presaging patterns of tree–fern and Sigillaria dominance of Late Pennsylvanian coals of the eastern United States, but, nonetheless, built on a species pool shared with the older coals.

Large areas of concentrated coal balls (permineralized peat) up to 4 m thick obstructed longwall mining in the Herrin Coal at the Old Ben No. 24 mine. The largest coal‐ball area mapped contained >1500 m 3 ; several areas contained >400 m 3 of coal balls. In‐mine mapping established that there were two types of roof (freshwater and marine), and that the coal balls were spatially correlated with the marine roof units. Regional studies and local data revealed that the younger, freshwater Energy Shale (mud) originally covered all of the peat deposited at the mine, but the mud was locally removed during a period of erosion. The great majority of coal balls are found within the coal seam, where they were permineralized in situ by carbonates. Some coal balls are found exposed on the eroded coal surface and others are within the channel fill associated with the erosion, which predates any marine sedimentation. Thus, the mineralization of the coal balls was synchronous with the erosion of the Energy Shale mud. Data from C‐and O‐isotope, geochemical, and mineralogical analyses of coal balls and associated materials were used to refine a depositional model of coal‐ball formation. The concentrated coal‐ball areas were created by the triggered degassing of CO 2 from partially decomposed peat in the presence of cations from fresh waters; 13 C/ 12 C ratios in these coal balls average −23.9‰. 13 C/ 12 C ratios as low as −34‰ occurring in the most concentrated coal balls are consistent with CO 2 produced by anaerobic oxidation of methane. Top‐of‐seam coal balls were formed later and show slight to strong marine influence; their 13 C/ 12 C ratios average −10.8‰.

Coal balls are formed from the precipitation of carbonate minerals, predominantly calcite, from water during or immediately post-peat formation. Previous petrological studies suggest permineralization occurred at various stages during the peatification process by precipitation from a mixture of marine and meteoric freshwater. The geochemical studies reported here aimed to test previous theories of coal ball formation. Coal balls studied here are from Upper Carboniferous and Carboniferous/Permian coal seams in Britain, United States of America, Belgium and China. Stable carbon isotopic values of calcites range from δ13C of −5 to −35‰ indicating highly variable contribution of decayed organic matter to the carbonate cement. Oxygen isotopic values of δ18O range from −3 to −15‰ suggest both marine and meteoric freshwater sources. The range, in the British coal balls in particular, with δ13C ranging from −7 to −25‰ indicates the complexity of carbonate generation. Isotopic data support at least four theories of carbonate coal ball formation. This makes their absence from post-Permian coals the more puzzling and it is suggested that the overall structure of the peat may play a role.

This paper reports on a new genus and a new species of the Order Cardiocarpales, Parataxospermum taiyuanensis gen. et sp. nov., which is described from coal balls of the Upper Carboniferous (or Lower Permian) in the Shanxi Province, China. The taxon is described and its taxonic position and affinities are discussed in detail. In addition, a three-dimensional reconstruction technique using computer graphics is utilized to reconstruct external images of the seeds.

Insight into the origin of fusinite was gained using Late Carboniferous polished coal balls (permineralized peat) from the New Castle coal bed (Alabama, U.S.A.), the Union coal bed, (Lancashire, England) and the Calhoun coal bed (Illinois, U.S.A.). Two types of fusinite were studied; primary fusinite (fusinite formed without undergoing charring or oxidation) and degradofusinite (fusinite formed due to decay or degradation). Evidence for primary fusinite in vascular plants previously relied on color in coal-ball thin sections for maceral identification. Color is shown to be an unreliable guide to maceral identity in coal-ball thin sections. Other evidence for primary fusinite from vascular plants is scant. Abundant fungi with vitrinitic reflectivity were found in coal balls whereas none with fusinitic reflectivity were found. Although genuine fungal sclerotinite does exist, the occurrence of fungal sclerotinite does not prove that fungal sclerotinite is primary fusinite but only that fungi can be fusinitized. Both lycopod periderm with fungal-decay features and coprolites have vitrinitic reflectivity in coal balls. These tissues should have had fusinitic reflectivity if decay or degradation results in fusinitization. These results, therefore, cast doubt on the existence of degradofusinite. Charcoal and decayed tissues show features that may allow them to be distinguished in some cases. In charcoal, the secondary cell walls of adjacent cells together with the middle lamellae are usually fused into a single homogeneous unit. In decayed tissues, the middle lamellae is preferentially decayed which can result in the separation of adjacent cells. Some prominent illustrations of degradofusinite show fusion of the cell walls of adjacent cells indicating an origin by charring. This indicates that presently accepted criteria for recognizing degradofusinite are inadequate even if degradofusinite does exist. Some evidence for degradofusinite does exist but the key illustration, as published (Teichmuller, 1950, 1989), does not demonstrate the existence of degradofusinite.

Possible fossil cocci, both solitary and diplococcus types, have been discovered in coalified tissue (xylem) from coal balls and bituminous coal of Carboniferous age. Most of the presumed bacteria are microspheroids 1–2.5 μm in diameter and were found in partly degraded humic tissue (xylem) from a medullosan seed fern preserved in a coal ball (Herrin No. 6 coal bed, Westphalian D), an indication that they may have been saprotrophs (decay causing). The same kind of bodies were also found in coalified xylem from the Pittsburgh coal bed (early Stephanian). Whatever the origin of the bacteria-like bodies, they provide direct evidence of secondary organic components in tissue that gives rise to vitrinite and fusinite, major macerals of bituminous and anthracitic coals.

In 2002 & 2003, the Ameren UE Meramec plant carried out complete coal pulverizer design upgrades and installations on the plants two largest operating units. These units were designed and constructed in the late 1950’s and early 1960’s. The coal pulverizer replacement project involved the removal of six rotary coal ball mills and the installation of twelve new state of the art coal pulverizers. This technical paper examines the resulting operating experience demonstrated and achieved as a result of the new coal pulverizer installation on the plants two largest generating units.Copyright © 2003 by ASME

Carbon allocation to different tissues and organs of Late Pennsylvanian-aged coal-swamp plants was determined for the first time through study of coal balls from the Calhoun coal of central Illinois. Coal balls are concretions that preserve the original peat fabric of the coal seam; they occur within the coal matrix and generally are permineralized with calcium carbonate. Coalified plant cell walls are preserved within the mineral matrix of the coal ball, entombed within the calcium carbonate. Relative carbon allocation was determined by separation of the mineral matrix and organic carbon. Carbon content was determined for replicates of tissues or organs of Psaronius tree ferns, Medullosapteridosperms, the lycopsid Sigillaria approximata, the small fern Botryopteris forensis, and "whole peat'. Unit volume results for tissues and organs were normalized for trunk sections 45 cm in diameter and 1 m long. Psaronius tree ferns were significantly "cheaper" than all other plants due to the large mantle of aerenchymatous roots that made up the bulk of the trunk. Medullosans and small ferns were most "expensive", with the arborescent lycopsid in between. Relative expense of construction closely parallels the inferred ecological role of each plant within ancient coal swamps, deduced from distributional patterns and indicators of habitat conditions. Tree ferns and lycopsids are colonists, medullosans are site occupying forms.

Late Pennsylvanian coal swamps of the Illinois Basin were dominated by Psaronius tree ferns with a spatially heterogeneous distribution of medullosan pteridosperms (subdominant), calamites, sigillarian lycopsids, and cordaites. Miospore and coal-ball plant assemblages from the Missourian-age Bristol Hill Coal Member (Bond Formation) and Friendsville Coal Member (Mattoon Formation) of southeastern Illinois were quantified to analyze vegetational patterns in Late Pennsylvanian peat swamps and to compare vegetational composition of the coals. In the Bristol Hill Coal (95 coal balls, 2613 cm 2 coalball peels), tree ferns are dominant (75.6%), followed by pteridosperms (11.5%), lycopsids (10.6%), sphenopsids (0.6%), small ferns (1.6%), and cordaites (0.1%)

Improved regional and interregional stratigraphic correlations of Pennsylvanian strata permit comparisons of vegetational changes in Euramerican coal swamps. The coal-swamp vegetation is known directly from in situ coal-ball peat deposits from more than 65 coals in the United States and Europe. Interpretations of coal-swamp floras on the basis of coal-ball peat studies are extended to broader regional and stratigraphic patterns by use of coal palynology. Objectives of the quantitative analyses of the vegetation in relation to coal are to determine the botanical constituents at the peat stage and their environmental implications for plant growth and peat accumulation. Morphological and paleoecological analyses provide a basis for deducing freshwater regimes of coal swamps. Changes in composition of Pennsylvanian coal-swamp vegetation are quire similar from one paralic coal region to another and show synchrony that is attributable to climate. Paleobotany and paleogeography of the Euramerican province indicate a moist tropical paleoclimate. Rainfall, runoff and evapotranspiration were the variable climatic controls in the distribution of coal-swamp vegetation, peat accumulation and coal resources. In relative terms of climatic wetness the Pennsylvanian Period is divisible into five intervals, which include two relatively drier intervals that developed during the Lower-Middle and Middle-Upper Pennsylvanian transitions. The climate during Early Pennsylvanian time was moderately wet and the median in moisture availability. Early Middle Pennsylvanian was drier, probably seasonally dry-wet; late Middle Pennsylvanian was the wettest in the Midcontinent; early Late Pennsylvanian was the driest; and late Late Pennsylvanian was probably the wettest in the Dunkard Basin. The five climatic intervals represent a general means of dividing coal resources within each region into groups with similar botanical constituents and environments of peat accumulation. Regional differences in basinal geology and climate were significant variables, but the synchronous control of paleoclimate was of primary importance.

Resin rodlets, sclerenchyma strands and woody splinters, which are collectively called rodlets, were studied by chemical, optical petrographic, and scanning-electron microscopic (SEM) techniques. A study was made of such rodlets from the bituminous coal beds of the central Appalachian basin (Pennsylvanian; Upper Carboniferous) of the United States. Comparisons were made with rodlets from coal beds of the Illinois basin, the Southern Anthracite Field of Pennsylvania, the St. Rose coal field of Nova Scotia, and European and other coal fields. In order to determine their physical and chemical properties, a detailed study was made of the rodlets from the Pomeroy coal bed (high volatile A bituminous coal; Monongahela Formation; Upper Pennsylvanian) of Kanawha County, West Virginia. The origin of the rodlets was determined by a comparative analysis of a medullosan (seed fern) stem from the Herrin (No. 6) coal bed (high volatile C bituminous coal; Carbondale Formation) from Washington County, Illinois. Rodlets are commonly concentrated in fusain or carbominerite layers or lenses in bituminous coal beds of the central Appalachian basin. Most of the rodlets examined in our study were probably derived from medullosan seed ferns. The three types of rodlets are distinguished on the basis of cellularity, morphology and fracture. The resin rodlets studied by us are noncellular and appear to be similar in properties and origin to those found in coal beds of the Middle and Upper Pennsylvanian of the Illinois basin. The resin rodlets extracted from the Pomeroy coal bed exhibit high relief and high reflectance when polished and viewed in reflected light; they are opaque in transmitted light. In cross section, the resin rodlets are oval to round and have diameters ranging from 60 to 450 μm. Many are solid, but some have vesicles, canals or cavities, which are commonly filled with clay, probably kaolinite. Typically, they have distinct fracture patterns (“kerfs”) in longitudinal and cross sections and many are characterized by dense (probably oxidized) rims. The orientation and amounts of void space and mineralization of resin rodlets in coal have resulted in much confusion in their recognition and classification. The resin rodlets are petrographically recognized as sclerotinites of the inertinite maceral group. We here propose that resin rodlets be assigned to the maceral variety of sclerotinites termed “resino-sclerotinite” because of their presumable resinous origin. Other investigators have confused some fusinitized resin rodlets with fungal masses, which have different morphological properties and which probably have different chemical properties. We here propose that such fungal masses be assigned to the maceral variety of sclerotinites termed “fungo-sclerotinite.” The sclerenchyma strands examined in our study are cellular, thick-walled, and crescent-shaped in cross section. They exhibit high reflectance and high relief and belong to semifusinite and fusinite of the inertinite maceral group. Sclerenchyma strands are commonly associated with resin canals in Medullosa and related seed-fern genera, which are common in coal balls of the Illinois basin. We here propose adoption of the maceral varietal terms “sclerenchymo-fusinite” and “sclerenchymo-semifusinite” for these bodies. The woody splinters in the Pomeroy coal bed are cellular and thin-walled and have scattered pits as much as a few microns in diameter. They are dark brown to black in transmitted light and commonly have a lower reflectance than the resino-sclerotinite and sclerenchymo-fusinite of the Pomeroy coal. The woody splinters belong to semifusinite and fusinite of the inertinite maceral group. The maceral varietal terms “xylemo-semifusinite” and “xylemo-fusinite” are here proposed for these bodies. Elemental chemical data for the resin rodlets of the Pomeroy coal bed of the central Appalachian basin indicate that resin rodlets have significantly lower atomic H/C and O/C ratios than do sclerenchyma strands and woody splinters. The lower atomic H/C and O/C ratios of the resin rodlets correlate with the highest reflectance. In the coal ball medullosan seed-fern stem from the Herrin (No. 6) coal bed of the Illinois basin, the reflectances of the resin rodlets, woody splinters and sclerenchyma strands are similar and comparable to those of associated vitrinite in the coal ball stem and in the attached coal. However, resin rodlets and sclerenchyma strands in the attached coal have significantly higher reflectances, similar to those of the Pomeroy coal.

SUMMARY Rodlets from fusain and carbominerite bands or lenses in bituminous coal beds of the central part of the Appalachian coal field in the eastern United States were studied in three dimensions using both light microscopy and scanning-electron microscopy (SEM). Infrared studies and chemical analyses complemented reflectance measurements. The rodlets are found strati-graphically from the Pocahontas no. 3 coal bed (Pocahontas Formation, Lower Pennsylvanian or Namurian B & C) to the Washington coal bed (Washington Formation, Lower Permian). The most common rodlets are noncellular, range in diameter from 60 to 440 μm, and, when polished, have the relief and high reflectances typical of the inertinite maceral group. They have characteristic notches, ovoid shapes, distinctive fracture patterns, dense (oxidized) rims, vesicles, cavities and canals, some of which contain minerals (tentatively identified by SEM semiquantitative X-ray and infrared analyses). Some rodlets show a cellular cast on their longitudinal surfaces. Due to different exposures in variously-oriented polished sections, this type of rodlet is classified as sclerotinite of the inertinite maceral group. Comparison of the sclerotinites from the central part of the Appalachian coal field with the fusinitized resin rodlets from coal and coal balls of the Illinois basin (Kosanke & Harrison, 1957) and with the ‘sclerotioids’ of the St Rose no. 5 coal of Nova Scotia indicates that they represent the same bodies. All are interpreted to be rodlets of resinous origin belonging to one or more of the genera of medullosan seed ferns of Carboniferous and Permian age. We suggest using the existing terms ‘resino-sclerotinite’ and ‘fungo-sclerotinite’ for two distinct maceral varieties of the inertinite maceral group because of their distinguishing properties and their different paleobotanical origins. Two types of cellular rodlets, woody splinters and sclerenchyma strands, are also documented in this study because of their close association with the resino-sclerotinites of the central part of the Appalachian coal field.

The first sufficient quantitative analysis of vegetation preserved in coal balls from the Allegheny Formation, late Middle Pennsylvanian age, was made in the present study by measuring 13,203 cm2 from middle peels of over 300 coal balls or pieces of coal balls. The distinctive features of this flora are the dominance of lycopods, the absence of cordaitean plants, and the extremely low shoot/root ratio (0.74 for total plant remains, 0.24 for lycopods). It is suggested that the climate at the Derringer Corners coal swamps was a drier tropical climate, compared to that of some earlier coal-forming environment of the Appalachian Basin, and to those of coal swamps in the Illinois Basin during the Middle Pennsylvanian. The discrepancy between the abundance of lycopods (as an indicator of a wet environment or habitat) and the low shoot/root ratio (inferring relatively dry conditions) in the coal-forming peat swamps at the Derringer Corners indicates that the actual set of factors causing massive reduction of arborescent lycopods at, or near, the boundary between Middle and Late Pennsylvanian would be much more complex than simply a “drier climate”. Some local factors, such as edaphic, perhaps geomorphic conditions, and local events, might play an important role in creating a favorable environment for the growth of lycopod trees.

Coal balls (exceptionally preserved calcareous permineralized peat), widely described from tropical Carboniferous Euramerian coal seams, have yielded diverse data on the biology, ontogeny and ecology of swamp plants and ecosystems. Probably over 75% of the swamp taxa may have been preserved, in contrast to probably less then 10% in other contemporaneous environments, but the assemblages are species-poor and represent an evolutionarily conservative assemblage. The in situ nature of coal ball peats allows ecological changes to be identified from vertical profiles. Major changes in plant communities both within individual seams and between seams appear to reflect both local and widespread climatic changes, particularly in rainfall. The preservation of cell contents, plant apices, gametophytes, etc., demonstrates the contemporaneous or early diagenetic formation of coal balls. Their common association with marine sediments has supported arguments for a marine source for the carbonate. Stable oxygen and carbon isotope studies suggest a meteoric origin for some carbonate. No current model for the formation of coal balls completely explains their occurrence and rarity outside the Upper Carboniferous of Eurameria.

Three methods were used to collect data from the same coal-ball peels and were compared for data comparability and for relative time efficiency. Although the data produced by the standardly used cm2 grid method and by the newly devised random mm2 method were of equivalent accuracy, the newly devised method is estimated to be much more time efficient for abundant samples of coal balls containing numerous small plants. In contrast, the accuracy of the subdivided cm2 method is equivocal, and the time efficiency of this method is minimal.

The amount, kind, distribution, and genesis of pyrite in the Lower Bakerstown coal bed in a 150 × 15 m area of the Bettinger mine, Castleman coal field, Maryland, were studied by various analytical techniques. The mined coal, which had a nonmarine roof rock, contained 1.4–2.8 wt.% total sulfur, generally much lower than the high-sulfur coal (> 3.0 wt.% total S) to the north, which is associated with marine roof rocks. Small-scale systematic and nonsystematic variations in total sulfur and pyrite distribution were found in the mined area. In the column sample, most of the pyrite was found in the upper 9 cm of the 69-cm-thick mined coal and occurred mainly as a pyrite lens containing cell fillings in seed-fern tissue (coal ball). As-bearing pyrite was detected by laser microprobe techniques in the cell walls of this tissue but not elsewhere in the column sample. This may indicate that the As was derived from decomposition of organic matter in the cell walls. The sulfur isotopic composition and distribution of pyrite in the coal are consistent with introduction of marine sulfate shortly after peat deposition, followed by bacterial reduction and pyrite precipitation. Epigenetic cleat pyrite in the coal is isotopically heavy, implying that later aqueous sulfate was 34S-enriched.

Criteria are provided for identification of certain Pennsylvanian-age plant megafossils directly from coal based on their characteristic anatomical structures as documented from etched polished coal surfaces in comparison with other modes of preservation. Lepidophloios hallii periderm, Diaphorodendron periderm, an Alethopteris pinnule, and a Cordaites leaf were studied in material in continuity with adjacent permineralized peat (carbonate coal-ballas). Calamites wood in attachment to a pitch cast and a Psaronius stem in coal in attachment to a fusinitized Psaronius inner root mantle were studied. Sigillaria was identified in coal by comparison to its structure in permineralized peat. Other plant tissues with characteristic structures were found but could not be attributed to specific plants.

Shell-armored wood cobbles occur on detrital-peat beaches along the seaward edge of the Mississippi Delta. Shell material consists exclusively of Mulinici lateralis, a dwarf surf clam. Soft, heavy, waterlogged wood fragments are abraded and become armored by hard shells in response to wave activity on the beach. Although their preservation potential is suspect, fossilized shell-armored wood clasts would probably be recognized as a type of coal ball and might indicate an allochthonous origin for the host coal.

An evaluation was made of the degree of coalification of two coal balls from the Illinois Basin of the Pennsylvanian (upper Carboniferous) of the United States. Previous interpretations are mainly misleading and contradictory, primarily because of the assumption that the brown color and exceptional cellular and subcellular preservation typical of American coal balls imply chemical preservation of cellulose and lignin, the primary components of peat. Xylem tissue from a medullosan seed fern contained in a coal ball and the coal attached to the coal ball from the Calhoun coal bed, Mattoon Formation, Illinois, was analyzed by elemental, petrographic, and nuclear magnetic resonance (NMR) techniques to determine the degree of coalification. The NMR and elemental data indicate the lack of cellulose and lignin and a probable rank of high-volatile C bituminous coal. These data corroborate data for a coal ball from the Herrin (No. 6) coal bed (Carbondale Formation, Middle Pennsylvanian) and support our hypothesis that the organic matter in coal balls of the Pennsylvanian strata of the United States is coalified to about the same degree as the surrounding coal. Data presented show a range of lower reflectances for xylem tissue and vitrinite in the analyzed coal balls compared with vitrinite in the attached coal. The data reported indicate that physical preservation of organic matter in coal balls does not imply chemical preservation. Also our study supports the hypothesis that compactional (static load) pressure is not a prerequisite for coalification up to a rank of high-volatile C bituminous coal. A whole-rock analysis of the Calhoun coal ball indicates a similarity to other carbonate coal balls from the United States. It consists primarily of calcium carbonate and 1–2% organic matter; silica and alumina together make up less than 0.5%, indicating the lack of minerals such as quartz and clays.

Plant tissues were traced from coal balls to coal to identify and distinguish their characteristic structures in coal and to determine compaction ratios for specific tissues. Comparison of average cell sizes of lycopod aerial periderm in coal-ball peat and coal gave compaction ratios ranging from 7:1 to 19:1. Some of this compaction took place by three-dimensional contraction of the periderm so that the shapes of the cells in cross section were not greatly changed. Compaction ratios of other tissues ranged from 3:1 to 60:1. Tissues traced into vitrinite include lycopod stem periderm, Stigmaria periderm, “ Diaphorodendron ” cortex, Myeloxylon, Pennsylvanioxylon wood and bark, marattialean fern leaves, Psaronius outer roots, and Stigmaria rootlets. Tissues traced into fusinite or semifusinite include lycopod stem periderm, Myeloxylon , and Psaronius inner roots. Cordaitean leaves, Cordaianthus , and marattialean fructifications were identified in coal and coal balls from the same layer. The coal balls were from the Herrin, Springfield, and Houchin Creek Coal Members of the Carbondale Formation, Desmoinesian Series, Pennsylvanian System and an unnamed coal in Indiana equivalent to the Murphysboro Coal Member in Illinois, Staunton Formation, Desmoinesian Series, Pennsylvanian System. Thin sections and etched polished blocks were characterized by light and SEM microscopy.

MEDULLOSA DISTELICA iS the third species of this genus to be discovered in America. Some years ago Thiessen (1920a, 1920b) illustrated a pyritized tristelar Medullosa in connection with his studies of bituminous coal. This form was identified as Medullosa anglica, but after reexamination of Thiessen's specimen, it seems best to class it as a distinct variety. The origin of Thiessen's specimen is unknown beyond the fact that it had been collected on one of the earlier reconnaissance trips in the western coal fields of this country.2 Medullosa noei, the second species, was recently described by Steidtmann (1937). It is from coal balls found in a thin locally mined coal a few miles east of the town of Calhoun, Richland County, southeastern Illinois. The limestone above the coal at this locality has been described (as LaSalle) bv Newton and Weller (1937) and is placed at about the middle of the Mcleansboro formation. This formation is considered roughly equivalent to the Conemaugh of eastern United States. In Richland County the Herrin (No. 6) coal at the base of the McLeansboro is about 800 feet below the Calhoun coal.3 The third species, here described as Medullosa distelica, is from coal balls out of the upper part of the Herrin (No. 6) coal bed (Cady, 1936) in the mine of the Clarkson Coal and Mining Co. at Nashville, Washington County, Illinois. The Nashville coal balls were discovered in April, 1935, by G. H. Cady, L. C. McCabe, and C. G. Ball of the Illinois Survey. These calcareous concretions have provided a variety of fossil plant material representative of that which contributed to the formation of the upper part of this coal bed. New plants, such as the one treated in this paper, will be described from time to time independent of studies on the botanical constitution of coal, since such fossils have special botanical and stratigraphic significance. Medullosa distelica sp. nov.-Stem oval, of medium size, measuring ca. 8 X 4 cm. when slightly compressed. Two highly asymmetric protostelic cylinders each ca. 7 X 18 mm. in diameter with mixed "pith" occur within the bandlike internal periderm. Secondary wood strongly developed on 1 Received for publication December 23, 1938. Published by permission of the Chief, Illinois State Geological Survey, Urbana. The author is much indebted to Dr. Gilbert H. Cady, Senior Geologist of the Survey and Head of the Coal Division, for his encouragement and constructive criticism during the preparation of this paper. Dr. J. T. Buchholz of the University of Illinois Botany Department has been kind enough to read the manuscript and suggest improvements. Mr. Oliver of the Classics Department of the University aided by suggesting certain of the new names proposed. 2 Reinhardt Thiessen, personal communication to the author, October, 1936. 3 Records of the Division of Subsurface Geology, Illinois State Geological Survey. each stele toward the center of stem, much less developed on external sides of each stele; little or no secondary xylem associated with outgoing leaf traces. Internal periderm 11/2-3 mm. broad and originally probably continuous around the stelar tissues. The periderm band is not broken where the outgoing leaf traces pass through. Three extremely decurrent leaf bases occupy the outer 1-2 cm. of stem, two of which are demarcated internally by a sclerenchyma zone. Internodes long. The sclerotic rind is of Myeloxylon landroitii type, 2-4 mm. thick, with its internal boundary not strictly definable. Secretory ducts enclosing "resin rodlets" numerous and commonly associated with sclerotic strands. Leaf trace bundles are chiefly of the typical centripetal collateral type; they are a little more massive when entering the leaf base but soon bifurcate to assume their tvpical configuration. The holotype was derived from an approximately spherical coal ball. The stem was situated somewhat excentrically within the nodule and exposed in slightly oblique section by the first exploratory cut. About 5 inches of the stem was available for study. Figures 1 and 2, plate 1, are taken from opposite extremities about 4 inches apart, where in each case the specimen is fairly complete. Figure la is a tracing taken from the photograph in lb. No leaf traces are indicated in this or other drawings. The stelar structures, similar to those shown in plate 1, figure la, have been drawn at successive levels approximately an inch apart and are shown in their normal succession in text figure ic, d, and e; figure 2, plate 1, is thus subjacent to text figure le. The phyllotaxy merits no special comment at this time except to note that the same leaf bases are present throughout, and thus a long internode is indicated. Internal sclerenchyma setting off two of the leaf bases is present, making it more like Medullosa centrofllis (De Fraine, 1914) than M. anglica in this respect. One of the distinct features of this species is the extreme asymmetry of the xylem cylinders. By asymmetry is meant deviation (in this case bilateral) from the radial development common to most steles of vascular plants. In Medullosa distelica asymmetric stelar growth is most prominently displayed by the secondary wood which is narrow on the external side of the individual stele and much thicker internally. Medullosa anglica (Scott, 1899) shows similar asymmetry of the steles but in a lesser degree; thus far this feature has attracted little attention. In M. anglica the asymmetry of growth might be explained, at least in part, as due to depletion of the peripheral wood by outgoing leaf traces. Since the leaf traces are considerably less massive in M. distelica than in M. anglica, they are

To botanists and geologists alike coal and the beds associated with it afford a rich field for investigation. Geologists have long been particularly fascinated by the theories which attempt to account for the mode of origin of the coal, and as a result a vast literature has been built up in support of the various views of the authors who have dealt with the problem. Nevertheless, for our purpose, it is enough merely to name the two main theories which now hold the field, as the “growth in situ ” and the “drifted” origin for the plants which compose the coal. It will, perhaps, be well to express clearly at the beginning of this paper that we do not propose to enter into a critical discussion of the views of other writers on the subject, except where they have a very direct bearing on our own investigation. It will be found that the results of our work have a certain bearing on the wider subject of the origin of coal in general, but in this paper we wish to confine ourselves to the consideration of questions suggested by the “coal balls” in particular. The conclusions drawn from the special aspect of the case are not applicable to all seams of coal, though they illustrate fully the history of certain seams. The calcareous masses which occur actually embedded in the coal of some districts have proved of vital importance to students of botanical anatomy and phylogeny, because they contain plant tissues so perfectly petrified that the complete structure of the plants can be discovered by microscopic investigation. Yet although these calcareous masses or “coal balls” have been the source of so much valuable information, little is to be found in the literature, and one gathers also that but little is actually known to scientists about their mode of occurrence and the many interesting phenomena presented by their relation to the beds in which they are found. We have endeavoured to collect and preserve information about these structures, which are being daily destroyed, information which may prove of service to both botanists and geologists.

1. American coal balls showing preserved plant structure have been sectioned. 2. From an Illinois coal ball from the Carbondale formation were obtained sections of a monocotyledonous stem, with collateral closed bundles scattered throughout the solid cross-section. 3. The vessels were of the spiral type exclusively, with transverse end walls. The phloem consisted of sieve tubes and what appeared to be companion cells. 4. Around the periphery of the stem, arranged in four or five irregular concentric rows, were groups of hexagonal, small, heavy-walled sclerenchyma cells. 5. This is the first angiosperm described from the Paleozoic, and the name Angiospermophyton americanum is proposed. 6. No attempt is made as yet to correlate this with any living group. 7. This announcement throws no light on the phylogenetic origin of the angiosperms, but does remove the problem from the Mesozoic to the Carboniferous or pre-Carboniferous.

MAMAY, SERGIUS H. (U. S. Geological Survey, Washington, D. C.) Litostroma, a new genus of problematical algae from the Pennsylvanian of Oklahoma. Amer. Jour. Bot. 46(4): 283-292. Illus. 1959.-Litostroma oklahomense, gen. et sp. nov. is described on the basis of fossil plant material found in a Pennsylvanian marine limestone from the vicinity of McAlester, Oklahoma. Litostroma is a simple plant consisting of a small, irregularly shaped thallus 1 cell thick. Some thalli have small perforations and filament-like marginal outgrowths. The reproductive organs are not known. A marine fauna is intimately associated with Litostroma, and includes epiphytic Foraminifera preserved in actual growth positions on surfaces of the plants. The evidence indicates a marine habitat for Litostroma and, accordingly, algal affinity. Without information bearing on its reproductive organs or pigmentation, however, Litostroma cannot with assurance be assigned to any known group of algae. It possibly represents a group of green, brown, or even red algae. AN INVESTIGATION of marine faunas associated with land plants in American coal balls led to examination of samples of a fossiliferous marine limestone associated with coal balls from Oklahoma. The limestone contained some curious thallose plant remains, the unusual structure and excellent preservation of which warrant detailed description and an attempt at interpretation. The material was collected in 1939 by C. B. Read, U. S. Geological Survey, at a slope mine owned by Joseph Lemont, near McAlester, Oklahoma. The productive seam, the Secor coal, lies in the Boggy shale of lower Des Moines age (Hendricks, 1937). Hendricks mentions no limestone above the Secor coal, but during my visit to the locality in 1955 Mr. Lemont informed me that the limestone overlay the coal directly, at the site of the coal ball occurrence. The limestone was likely of very local development. The exposure has since been filled and is no longer accessible. MATERIAL AND TECHNIQUES.-The plants were richly concentrated in an ovoid chunk of nodular limestone, 8 cm. in greatest dimension; they also occurred sparingly in a few other nodules in the collection. Since it was evident that the thalli contained in the several nodules were identical, only the contents of the richest specimen (U.S.G.S. 8764-9) were investigated in detail. Saw cuts of the nodule revealed a diverse assemblage of marine animal remains associated with the plant material. Most of the rock matrix is dark 1Received for publication September 18, 1958. Publication authorized by the Director, U.S. Geological Survey. I wish to express my appreciation to Dr. W. C. Steere, New York Botanical Garden, and to Dr. Rufus H. Thompson, University of Kansas, for their interest and advice in this investigation. Thanks are also due to Mr. L. G. Henbest, U.S. Geological Survey, for the photographs shown as fig. 29-31 and his paleozoological and paleoecological contribution to this paper, and to Mr. N. W. Shupe, U.S. Geological Survey, for the remainder of the photographic illustrations. brown, but there also are roughly cylindrical, intertwined areas of light gray color. The distinct color and sharp contacts of these light areas with the surrounding matrix suggest that they are filled animal burrows, made while the limestone was still a calcareous mud. A faint lamination of the dark part of the matrix is visible on smoothed surfaces, roughly parallel to the plane defined by the long axes of the nodule. This is due to a concentration of small, bedded plant thalli. In section, the plants appear as numerous chainlike groups of cells (fig. 3, 4), the chains lying parallel to each other. In places, groups of several are bent and contorted (fig. 1) but they are mostly flat and parallel, imparting the laminated appearance to the matrix. Etching the rock surface with weak formic acid showed the plants to be well preserved and undistorted by compaction; it also showed that they were not actually simple cellular chains 'but small, flat platelets a single cell thick. Sectional views showed no diagnostic features, so a chip of the rock was broken off and slowly dissolved in weak formic acid with the hope of isolating specimens whose surface features could be observed. The resultant sludge contained literally hundreds of fragmentary thalli. The process was repeated with additional samples. Tap water was then added to the sludge and decanted after settling of the organic matter; this was repeated until all acid had been removed, and all specimens except those selected for permanent mounts were placed in 30% ethyl alcohol for storage. Attempts to dry specimens were unsuccessful, as the cell walls collapsed with desiccation. This rendered the preparation of permanent mounts difficult, and the delicacy of the thalli necessitated caution in both chemical and mechanical treatment. The following technique produced excellent results. The aqueous residue was scanned in a dish under a binocular microscope. When a suitable specimen

1. Structurally preserved seeds were found in coal ball material collected under the auspices of the Illinois State Geological Survey by Dr. A. C. Noe of the University of Chicago. The source of the coal balls was from southern Illinois and Indiana in coal no. 5, which is in the Allegheny group of the Upper Pennsylvanian. 2. The seeds described are Conostoma oblongum Will. of the order Lagenostomales, and Pachytesta gigantea Br. of the order Trigonocarpales. This is the first description of Pachytesta from American coal balls. 3. Illustrations of Lagenostoma ovoides from the British Coal Measures, and of Pachytesta gigantea, Conostoma oblongum, Cardiocarpon, and Trigonocarpon hookeri from the American Coal Measures, are used to demonstrate similarity in development of Paleozoic seeds with those of modern gymnosperms, and to show variation in ontogeny at the time of petrifaction.

Chemical analyses of morphologically preserved organic matter in a Carboniferous coal ball reveal that the material is coalified to a rank approximately equal to that of the surrounding coal. Hence, the plant tissues in the coal ball were chemically altered by coalification processes and were not preserved as peat.

A simple method of treating cobalt-60 and iodine-131 production liquid waste can be disposed completely after alkalized by sodium hydroxide and then run through the soil wastes of coal ball column; while the iodine-131 production liquid waste can be disposed completely after reduced by sodium bisulfite, alkalized by sodium hydroxide and finally run through the soil wastes of coal ball column. The present method is better than the adsorption method through soils, for the liquid wastes can go through the soil wastes of coal ball more freely. The decontamination factor of this method is very high. Under the experimental conditions, one hundred grams of the soil wastes of coal can be used to dispose more than 15 liters of cobalt-60 liquid waste and 10 liters of iodine-131 production liquid waste respectively.

This new genus of fossil plants was first described and introduced to the scientific world so recently as the year 1878, by Professor W. C. Williamson, F.R.S., in his ninth memoir “On the Organization of the Fossil Plants of the Coal Measures.” Fragments of Astromyelon, however had been previously found and described as Calamitean, on account of their close resemblance to similar structures in Calamites, both by Mr. Binney and Professor Williamson. Since the publication of the above memoir, our knowledge of this interesting genus has been largely increased by the discovery of a large series of additional specimens by myself and other workers in this branch of science. I purpose in this paper to give a short sketch of what is known about Astromyelon and also of its nearest relations. The Astromyelons have been named and described solely from the microscopical examination of specimens obtained from the coal-balls found in the Halifax Hard Bed Coal, and in a similar coal in the neighbourhood of Oldham. Very little is known respecting the external appearance, but they appear to have been for the most part small herbaceous plants, like Asterophyllites and the smaller kinds of Calamites. I have recently found specimens of Astromyelon both as casts in sandstone and as impressions on shale. In the flagstone quarries at Ringby, near Halifax, I saw the impression of a longitudinal section of the stem of Astromyelon on the face of a flagstone, which shewed the structure almost as perfectly as a coal ball ...

Carboniferous petrifactions known as "coal balls" have long been known from Europe, and much of our knowledge of Carboniferous plants has been derived from a study of these petrifactions. Coal balls have been collected in America, however, but for a comparatively short time. The purpose of this paper is to contribute to our knowledge of this coal ball flora by adding certain new members to the growing list of American Carboniferous plants with structure preserved, with brief anatomical descriptions of the species listed. The plants included in this paper are described from sections of coal balls collected from various localities in Illinois, either by the writer, or by Dr. A. C. Noe, under the authority of the Illinois Geological Survey. All were found in the McLeansboro formation, the approximate time equivalent of the Conemaugh of Pennsylvania.

THE method described in this letter was devised in collaboration with Dr. R. G. Koopmans of Utrecht. A flat surface must first of all be cut or ground on the coal ball or other petrified mass parallel to the planes in which the sections are required: this surface is ground smooth, but a fine polish is not necessary. The surface is then immersed for a definite length of time in an etching solution of hydrochloric acid, the optimum concentration being found by trial. The acid dissolves away a film of the carbonate of which the mass is largely composed, and the plant substance contained in the film is left behind, standing in relief above the surface. The surface is washed carefully and dried. A solution or fluid, which on drying or hardening forms a tough film, is then poured over the surface and allowed to dry or harden. Several solutions have been used for this purpose; the best as regards consistency is the trade preparation ‘Durofix,’ but other solutions containing cellulose compounds may be used. When quite hard this film of cellulose-compound may be peeled off, as it has considerable tensile strength. The plant substance which was left in relief by the etching process is embedded in the film and is detached with it from the surface of the petrifaction. The film is then washed, first in the acid and then in water, dried, cleared, and finally mounted in Canada balsam between a slide and coverslip. The surface of the petrified mass is gently rubbed down on fine carborundum paper and is then ready for the removal of another section.

The present description concerns a lepidophyte sporophyll and sporangium discovered in coal ball M22 of the University of Kansas collection of fossil plants. The fossil material was collected by the author at the Pittsburg-Midway Coal Mine about 8 miles southwest of West Mineral, Kansas. The coal mined at the locality is assigned to the Fleming and Mineral seams. The Fleming seam lies about 12 feet above the Mineral and contains the majority of the coal balls although occasional pockets are known to occur in the Mineral coal as well. Stratigraphically both these seams are of approximately lower middle Pennsylvanian age.