Tree ferns 15 meters high along streams, and conifers in the dryer regions. In lesser quantity, plants related to the present-day horsetail (which resemble vertical drinking straws as much as 1.5 meters long) in both environments. This was the vegetation in an area near the municipality of Filadélfia, in the Brazilian state of Tocantins, during the early Permian Period nearly 300 million years ago. During that time, the blocks that make up South America were grouped very differently and were located farther to the south—the region now known as São Paulo, for example, was covered with ice fields. As these blocks migrated to warmer regions of the planet, the flora were also able to migrate. Closer to the end of the Permian, about 270 million years ago, vegetation already existed in what is now inland São Paulo State.
“I was convinced I would find more similarities between the fossils of that period found in the Parnaíba Basin in the Northeast, and the ones found in the Paraná Basin in the Southeast,” says paleobotanist Rosemarie Rohn Davies of the Universidade Estadual Paulista (Unesp) in Rio Claro. “But only the tree ferns are similar.” This picture of a distant past comes from evidence in the form of petrified stems and leaves, studied by Professor Rohn’s team and researchers from the University of Campinas (Unicamp), with funding from FAPESP, and from the Federal University of Rio Grande do Sul (UFRGS).
The analysis of stems from these giant tree ferns that still lie nearly whole at the Tocantins Fossil Trees Natural Monument, and of leaves that look like stone lacework, was the subject of the 2011 PhD dissertation of biologist Tatiane Marinho Vieira Tavares, now a temporary professor at the Federal University of Tocantins in Araguaína. She described the anatomy and morphology of this material belonging to the genera Psaronius and Tietea—the latter also found in the Paraná Basin, the subject of her master’s thesis. The tree ferns were apparently able to move from north to south over the course of millions of years as Gondwana, the supercontinent that contained many of the continents now in the Southern Hemisphere, was moving northward and becoming warmer.
The leaves were described as members of a new species because they were different from any seen before, but it is impossible to determine with which stems they formed a living plant. The reason is that the paleontological notion of species is very different from the concept used in biology—the source of endless discussions among specialists. Since paleontologists generally do not have to assemble the different parts of plant fossils—root, stem, leaves, etc.—into the jigsaw puzzle of a single plant, it is acceptable practice to describe each part as a different species. In the case of the fossil tree ferns in Tocantins, the leaves were much thicker than those of the present-day ferns, as shown in a forthcoming article in the Review of Palaeobotany and Palynology. “That characteristic has everything to do with environmental conditions,” explains Tavares, who was advised by Professor Rohn. “The thick leaf blade protects the plant’s reproductive structures and prevents excessive water loss in an arid or semiarid environment.” That is surprising because tree ferns depend on water during at least some phases of reproduction, so they are normally associated with moist environments, but what enabled these large trees to survive were the streams that they bordered.
Growing amidst the tree ferns, but not solitarily, were sphenophytes, whose only present-day representatives are horsetails. They are the subject of research being conducted by ecologist Rodrigo Neregato, who recently completed his doctorate with Professor Rohn. He described five new species of Arthropitys and found two different types: one having a very large pith, suggesting a habitat near water, and another with a more succulent stem, which probably lent the plants enough mechanical structure to live in firm soil a little farther from streams.
The analyses revealed plants that were very different from their descriptions in books, beginning with the unexpected ability to live in dry soil. They also look larger than expected. “We have 3-meter-high specimens that don’t even include the entire plant,” says Neregato. But what he expects will elicit more surprise are the vertical roots, instead of the horizontal rhizome hypothesized for this plant group. “We have a specimen with a root connected to the stem—the only one we know of to date,” he notes. He believes that this pattern holds true for the other sphenophytes of that era, as it improves water absorption and anchoring in unstable soil. “It was a fairly heavy weight; an inverted T structure would not be capable of sustaining it.”
Far from streams, the landscape was dominated by gymnosperms, which resemble modern-day pine trees. Because of this ecological specialization, the petrified fossils of these plants are much less abundant than tree fern fossils. Silica dissolved in water is what preserves their anatomical structures in three dimensions. When a fallen plant begins to decompose, its tissues release carbon dioxide that acidifies the alkaline water, precipitating the silica that penetrates into the plant walls. Farther from streams, the conifers tended to decompose more rapidly, and fossils are more scarce. That is why the group was less widely studied until Professor Rohn suggested that biologist Francine Kurzawe, then doing doctoral studies in Roberto Iannuzzi’s group at UFRGS, should research it. “Most of the time we only have access to small fragments that are already very rolled up, with their outer layers worn away,” says Kurzawe, who is now doing post-doctoral research at the University of London.
In two papers published this month in the Review of Palaeobotany and Palynology, she describes a series of new conifer species, as well as some unusual structures. “The piths of the fossilized gymnosperms have channels that denote adaptation to a dry climate,” she points out, corroborating the climatic conditions revealed by the leaves of the tree ferns. According to Kurzawe, today these channels exist only in young plants, which lose the pith as they grow. Adult pine trees, with a hollow trunk where the pith once was, do not have these specialized channels for water storage.
Kurzawe’s research indicates similarities between Gondwana flora and the flora of Euramerica, which now constitutes the northern portion of the planet. “The area that is now the state of Tocantins lay on the boundary between these two regions,” she explains. The gymnosperms from that area appear to have remained at latitudes characterized by milder temperatures and did not migrate southward. This has been evidenced by fossils—completely different from the ones from Tocantins—found in seven municipalities in inland São Paulo State and studied by paleobotanist Rafael Faria as part of his doctoral research at Unicamp under the guidance of Fresia Ricardi-Branco.
Faria, now a professor at Pontifical Catholic University in Campinas, studied petrified wood—permineralized wood is the term preferred by specialists—from plants that lived some 270 million years ago. To do so he used both a traditional microscope and a scanning electron microscope, which enabled him get a better view of the cell structures. He defended his dissertation in April 2013, and the section describing the best-preserved fossils is about to be published in the Review of Palaeobotany and Palynology.
It was surprising to identify fungal hyphae in specimens that at first glance appeared to be dirty. “It’s the first record of a fossilized fungus in wood from that period in Gondwana,” says Faria, who interprets the finding as a sign of ecosystem collapse. “It’s as if there was a lot of organic material to be broken down, which was conducive to fungal proliferation.”
Faria also briefly described the ecology of these plants, based on his study of growth rings. In temperate regions, conifers generally produce wood with distinct properties geared to the season: in spring and summer the wood provides for water transfer to the crown, thus promoting growth, and in the fall it is more focused on subsistence. When comparing the growth rings of these fossils to those of present-day species, it is possible to infer whether or not the Permian conifers lost their leaves in winter. The analyses pointed to a community of mostly perennial trees that did not lose their leaves, especially in the Teresina Formation, where fossils can be found around the cities of Angatuba, Conchas and Laras. The other formation he has studied, Irati (around the cities of Piracicaba, Saltinho, Rio Claro and Santa Rosa de Viterbo), lies in slightly deeper—and therefore older—layers and hosted a higher proportion of deciduous trees, which shed their leaves in the winter. He believes that these observations corroborate earlier data indicating that during the Permian, this region of Brazil was located farther to the south than it is today.
Permian ecology of present-day Tocantins was the subject of the doctoral research conducted by ecologist Robson Capretz under Rohn’s guidance in Rio Claro. He studied fossils and their arrangement in an area of the Parnaíba Basin, and endeavored to reconstruct what the forest there must have been like. “I concentrated on the ecology of the fossils, rather than their anatomy,” he pointed out, distinguishing his research from the work of his colleagues. The principal conclusions, he says, indicate that the region was very flat and had a rainfall pattern similar to that of the monsoons of India, with very heavy rainstorms that periodically interrupted periods of drought and covered the region with a reasonably thick sheet of water. The torrents of water knocked down the stems and carried them over short distances. They ended up aligned in the same direction and buried in the sand, as shown in findings published this month in the Journal of South American Earth Sciences. “We don’t know how frequent these rains were,” says Capretz, “but the rest of the time it was almost desert-like.”
The arrangement of the plant fossils gives clues for reconstructing the characteristics of the streams—whether they were fast-flowing or gentle, narrow or wide, straight or winding. The resulting description contradicts the picture presented in geological studies that characterize the region as having dunes similar to the ones found today in Lençóis Maranhenses National Park. “But there are no tree ferns in Lençóis Maranhenses,” says Capretz, who follows the maxim that the present is the key to the past. Consequently, his findings have helped Tavares interpret what she saw in her fossilized leaves.
The movement of water is also responsible for the deposition of silica in the stems and the resultant petrification of the tree ferns. “Had they not been submerged and quickly buried in sand, they would have decomposed,” Capretz explains. These special conditions make Tocantins extremely important for paleontological studies. “There aren’t many places with petrified plants in Brazil, so not many studies of this kind exist,” he says.
The past in the present
Professor Rohn confirms that climate is essential for the proper preservation of fossils. A marked seasonal alternation raises the chances for occurrence of the type of fossilization found in Tocantins, where stems and leaves were preserved in all three dimensions. “In the Paraná Basin the fossils are two-dimensional,” she laments, and this makes it difficult to compare the two regions.
But anyone who walks often and attentively through the dry terrain at Fossil Trees Natura Monument has an excellent chance of finding fossils. That abundance is often the delight of those who sell fossils—an activity prohibited in Brazil. For this reason, much of the work on Brazilian fossil flora has been conducted in Germany, where researchers had acquired petrified material without knowing that it was obtained illegally. At least that material is available to Brazilians today through the collaboration of Kurzawe and Rohn’s group with Robert Noll and Ronny Rößler. The latter is director of the Chemnitz Museum, which has fossils that demonstrate the similarity of Permian flora from Tocantins and Germany.
Researchers involved in the study of petrified forests warn that fossil sellers are not the only threat to the preservation of that history. Overprotection that hinders access even to specialists is, in their view, an obstacle to the advancement of knowledge. “In order to study gymnosperms, we have to collect material and prepare slides to examine under the microscope,” Rohn says by way of illustration. “It’s impossible to identify anything with the naked eye.” Rafael Faria, whose research is being done outside of preservation areas, decided to publicize his work to obtain more material. He has already gotten calls from ranchers in interior São Paulo State offering fragments of “stone-wood” that they found on the ground.
CAPRETZ, R. L. & ROHN, R. Lower Permian stems as fluvial paleocurrent indicators of the Parnaíba Basin, northern Brazil. Journal of South American Earth Sciences. v. 45, p. 69-82. Aug. 2013.
KURZAWE, F. et al. New gymnospermous woods from the Permian of the Parnaíba Basin, Northeastern Brazil, Part I: Ductoabietoxylon, Scleroabietoxylon and Parnaiboxylon. Review of Palaeobotany and Palynology. v. 195, n. 1, p. 37-49. 16 Aug. 2013.
KURZAWE, F. et al. New gymnospermous woods from the Permian of the Parnaíba Basin, Northeastern Brazil, Part II: Damudoxylon, Kaokoxylon and Taeniopitys. Review of Palaeobotany and Palynology. v. 195, n. 1, p. 50-64. 16 Aug. 2013.