Cellulose digestion by wood roaches may have removed enough mulch (detritus) to have caused hiatus of coal, aridity, and some of the temperature rise, as well as increasing conifers across the Permian. The largest part of the temperature rise may have been from overturn of an anoxic ocean by a huge comet.
There is also a difficulty to perceive that rates of evolution must proceed at drastically different rates. A simple change such as change in size or loss of an existing organ should be able to occur with crisp speed, as we have seen with domesticated organisms. On the other hand, complicated neural patterns such as nest building and intricate organs such as termite poison squirting apparatus must surely require long periods of time. Progenitors of traits like these must surely go back a considerable span from when they first appear on the fossil record, especially when associated with a long lived or sparsely populated organism or when a sudden appearance of diverse forms imply an undiscovered source.
Sometime in the vicinity of the Cambrian period multicelled plants evolved which could synthesize cellulose and live on dry land [Axelrod, 1969]. This, unlike animals, is believed to have happened only once [Kenrik and Crane 1997]. The advantage cellulose gave them was tremendous. Cellulose is a long chain carbohydrate polymer derived from sugar. Not only are there very few organisms which can break the cellulose linkages, but even those which can, do so slowly, because enormously large molecules can form. Its tensile strength is very great, and therefore large stemmed plants were made possible for competition for light. Hydrogen bonding of the polymer chains allowed for considerable shock stress before rupture which prevented brittle fracture in high wind or animal blows. Also important was its lack of palatability and indigestibility to the primitive worms, millipedes, snails and other animals which came out on land then or soon after. As a result, descendants of these plants spread across the land in the form of forests of ferns and other plants by the Devonian. These forests resulted in thin coal seams, which became thicker as the carboniferous went on.
One reason why thick coal deposits were able to form was that those early plants made extensive use of lignin. They had bark to wood ratios of 8 to 1, and even as high as 20 to 1. This compares to modern values less than 1 to 4. This bark, which must have been used as support as well as protection, probably had 38% to 58% lignin [Robinson p608]. Lignin is insoluble, too large to pass through cell walls, too heterogeneous for specific enzymes, and toxic, so that few organisms other than Basidiomycetes funguses can degrade it. It can not be oxidized in less than 5% oxygen atmosphere. It can linger in soil for thousands of years and inhibits decay of other substances [Robinson p608]. Probably the reason for its high percentages is protection from insect herbivory in a world containing very effective herbivores, but nothing remotely as effective as modern insectivores and probably much fewer plant poisons than currently.
ROACHES
It was not until the Carboniferous that large numbers of insects appeared on the fossil record. It is thought that they may have evolved on some as yet unknown island from worm like ancestors similar to the caterpillar like Onychophora, and then became inoculated fully developed into the known continents. Primitive silver fish, collembola, and dragonflies became numerous. The Insect orders became the most important of the herbivorous orders. By the Pennsylvanian, by far the most successful were ancestors of the cockroach. Six fast legs, two well developed folding wings, fairly good eyes, long, well developed antennae (olfactory), an omnivorous digestive system, a receptacle for storing sperm, a chitin skeleton that could support, protect, as well as form a gizzard, and efficient mouth parts gave it formidable advantages over other herbivorous animals. Present day roaches can eat bark, leaves, pith of living cycads, paper, wool, sugar, cheese bread, oil, lemon, flesh, fish, leather, dead roaches [Miall], and earth [author's observation]. About 90% of insects were roaches [Zimmerman,]. The Pennsylvanian is sometimes known as the age of roaches.
Especially significant was their ability to lay their eggs on dry land. The excretion of nitrogen in the form of insoluble uric acid is said to be an important factor in their successful invasion of dry land. [Prosser et al, 1950]. Spiders, centipedes, scorpions, and amphibious vertebrates could take a toll on the ground and ancestors of the dragonfly could catch them in the air at least, but these predators lacked the efficiency of our best modern predators. Catching them, equipped as they were with efficient sensory and locomotive apparatus, may have been something other than easy. Even where the amphibians and dragonflies were numerous, roaches were probably an effective group, for they grew to great size.
Their attack on plants must have been severe, and they also must have had an affect on disease transmission. They were the most important part of the fauna until well into the Triassic [Anderson, et al, 1996]. The fast turnover of plant life helped to build up the soil organic content. The large roaches probably could not venture out onto twigs, so they probably caused even more leaf fall than what they managed to eat. Bite marks on leaves are rare, less than 4% [Chaloner]. I suspect that their discards and excreta were a considerable part of the thick layers of litter which later turned into coal. Insects were probably the dominant herbivores then [Shear]. Spore ferns, horse tail rushes, and lycopods or club mosses, were the plants present [Gastaldo, with colored graphs].
THE WOOD ROACH
Within the moist upper Pennsylvanian or Permian soil there dwelt a ciliated protozoa capable of digesting particles of cellulose and leaves which descended from above in a never ending rain of insect droppings, etc. and converting them into energy. This microorganism lived under anaerobic conditions since its descendants in wood roaches are anaerobic. There are no such organisms known to be living in the soil or swamps today, but they must have existed at one time.
Eventually a roach or roaches evolved, which, instead of digesting that protozoa, formed a symbiosis with it. This meant that now for the first time, a higher animal could make use of the vast stores of cellulose which littered the earth and perhaps even much of that incorporated in the plants, especially the trunks of trees. During the Permian plants started to dominate which made use of dead material high in cellulose in their interiors, such as conifers. According to Kirby the use of cellulose by insects took place in the Permian or Triassic [Steinhaus, 1947; p531]. Fossil wings have been discovered in the Permian of Kansas which have a close resemblance to wings of Mastotermes, which is the most primitive living termite and which is thought to be the descendant of Cryptocercus genus, the wood roach. This fossil is called Pycnoblattina. It folded its wings in a convex pattern between segments 1a and 2a. Mastotermes is the only living insect that does the same [Tilyard RJ 1937]. This is strong circumstantial evidence that the wood roach did evolve in the Permian. If the wood roach is the ancestor of Mastotermes, analysis of its mitochondrial ribosomal DNA for RNA genes [Kambhampati 1995], indicate that a large number of descendants are missing from the fossil record. I suspect that it was the primary change which brought, by the end of the Permian, a close to the carboniferous coal deposits, except in Australia [Tilyard, 1917] at least. Cellulose digestion is inefficient and very rare in insects [Martin], and is probably a recent development in the few other species that have it.
Not until the third molt does the nymph complete its symbiosis, possibly because of high pH in the gut [Nalepa, 1990]. As a result, wood roaches have a strong imperative to be social in order to transmit the microbes. This symbiosis probably predated sociality because parental care is rare in Blabeoidea to which family the wood roach is said to belong [Grandicolas 1996, p523]. However, Kambhampati places it neither there nor in Polyphaginae subfamily, but more closely related to Blattidae, so he assigns it to its own family, Cryptocecidae. All this last on the basis of mitochondrial DNA testing [Kambhampati 1996].
The Permian was a transition period in the earth's history. If the wood roach evolved in this period, it did so in a harsh environment. The Pennsylvanian may have been much different in some areas than the damp coal forests would indicate, but the Permian was definitely characterized by widespread aridity and climatic extremes. Indeed, the effects which wood roaches had on soil organic matter may have caused a considerable part of the climatic extremes. Mulch and organic matter permit considerable infiltration of water into soils by virtue of decreased runoff and increased porosity and also are capable of considerable water storage themselves. As a result moisture can be transferred far inland by air moistened by those stores. In addition, the oxidation of these vast stores of carbon must have caused an increase of the greenhouse effect. This could have caused a small part of the variable increase of carbon dioxide believed to have occurred in the Permian [Mora et al, 1996] [Berner & Canfield, 1991]. It could not have caused a major rise because the oceans have a tremendous capacity to absorb carbon dioxide. It was probably a triggering affect. The ocean contains fifteen times as much carbon as all terrestrial biotic content and sixty tines the atmosphere. [Walker and Drever 1988, p63][Gregor, et al]. This may be the reason why the oxygen started to drop rapidly at mid Permian while the carbon dioxide only started to rise rapidly starting about two thirds through the Permian [Huey and Ward 2005]. I suspect the bulk of the rise was from oxidation of cellulose in coal exposed by the rise of the Appalachian and Ural Mountains coupled with a lessened withdrawal by marine shellfish (because of phosphorus ocean surfeit to be discussed in a later article). Initially dissolution of the vast deposits of terrestrial limestone and dolomite may also have contributed, although eventually such dissolution would actually subtract carbon dioxide from the atmosphere because of making the ocean more alkaline, although half the carbon would be released again when shellfish skeletons settled out. The net affect would have increased temperatures especially in high latitudes [Creber & Chaloner, 1985; p47] and therefore evaporation. While increased atmospheric carbon dioxide would increase temperature, in early Permian, the first major rise in carbon dioxide followed a temperature rise. This would seem to indicate that a rise in temperature permitted cellulose digesting insects to move toward the poles, and thus increase release of carbon dioxide. When there was a drastic rise in carbon dioxide about 5 million years later, the temperature started its main rise about 2 or 3 million years after that, and it did not rise to its value it had at an equivalent carbon dioxide value [Montanez]. While the temperature rise was not caused by the carbon dioxide at first, I suspect that clearing the soil of mulch and vegetative cover, which enabled higher soil temperatures, was a large part of it. You can easily see this by feeling the dramatically lower temperatures under sun light by feeling the grass or tree leaves growing next to a stone walk under full summer sun. Subsequent changes in carbon dioxide showed no close parallel to temperature either.
The detritus removal also would have increased the areas of deserts. The affect was probably accentuated by a positive feed back implied in the release by warming of the huge stores of methane hydrate [Kvenvolden, 1988] in ocean sediments. In any case the southern glaciers around the Indian ocean subsided during the Permian and after they disappeared by its end, glaciers did not reappear extensively for more than another 200 million years.
It is difficult to speculate on the organic content of ancient soils. There is always the chance that carbonaceous material is added to or subtracted from the soil after burial, especially subtracted. Color is a poor indication even in modern soils [Vageler, 1947; p7]. However, coal and black soil shale had virtually disappeared from North America by the last of the Permian, and red shale became common. Red shale may have been caused by high alkalinity in the gut of humus eating roaches. Red shale appeared early in the Permian in South Africa and in late Pennsylvanian in North America and in the mid Permian in Europe and Argentina (Veevers p192). They formed below 40 degrees latitude and were associated with evaporates [Frakes p113], so they must have been in what we now call tropical savanna regions. If the red beds were caused by roaches they were not necessarily cellulose digesting species, although it is an indication that cellulose digestion arose in South Africa and then spread to North America on wood carried by ocean currents.
Toward the end of the Permian coal began to disappear from the tropics and became increasingly located at higher latitudes, unlike the Carboniferous [Erwin, 1993, p165]. This would be understandable if tropical wood roaches were moving toward the poles in a world which was warming up and they were already eating mulch (detritus) and possibly more importantly, reducing productivity by attacking trees at there most vulnerable place, the trunk interiors.
It is probable that some of those roaches, which were making use of mulch directly, we would have been tempted to designate termites or prototermites. This is because burrowing into mulch, soil, or logs imposes environments that require thin skin for respiration in a high carbon dioxide atmosphere, and favors loss of pigments and eyes. Also modern wood roaches are devoid of wings [Srinivas et al., 1996] and the most numerous species of humus eaters may have been so devoid then. Since they probably rarely or never flew or even emerged from their tunnels they would not have been likely to leave many fossils. However, their prototypes probably did leave fossils. Tilyard believes that Pycnoblattina genus of the Pyknoblattinae subfamily fossils from lower Permian Kansas are very similar to Mastotermes and are either an ancestor or sister group, as already mentioned. He bases this observation on the fact that the wing hinge of the rear wing is in the same location as Mastotermes and is found on no other modern insect (Tilyard 1937, p171 & 266). Most of the fossils are forewings. This may be because of the nature of Arachnid predation [[Duncan et al 2003]. There are no soft body parts known yet, but the pronotum head covering is similar in both genera also. There are a few fossil hard head parts which also bear some resemblance to Mastotermes.(Tilyard, p274), and other aspects of wing venation have many points of similarity.
Termites may have evolved from one of the forms similar to wood roaches and acquired their symbiosis from wood roaches. Gene sequences for 18s mitochondrial cytochrome oxidase subunit II and endogenous endo-beta-1, 4-glucanase indicates termites and wood roaches are in the same clade [Lo]. So it is fitting that we describe one of the two known survivors as reported by Cleveland [Cleveland et al., 1934]. The name of this roach is Cryptocercus punctulatus. It manages to exist in a narrow belt in the temperate regions where the winters are not too severe. It has no wings, is about 2.5 centimeters long, and manages to subsist in small colonies inside rotten logs. It still retains most of the roach features, but it bears some resemblance to termites. Its nymphs are not pigmented. The mouth parts are very similar to worker termites. Strangely enough it has a peculiar jerky motion at certain times, which some termites use. Its courtship is similar to termites and it mates before, during, and after it lays eggs [Nalepa, 1988]. Most fundamental of all is the close similarity of the digestive system to that of termites, including a symbiosis with cellulose digesting protozoa housed in the hind gut which are lost after each molt. It has a primitive social structure, required for transmittal of its protozoa after a molt.
All these facts constitute circumstantial evidence that this insect is closely related to the ancestor of the termites. However it does not leave any additional clues as to the importance of its progenitors in ancient ages. For this, a look at its protozoan population is instructive. Its hind intestine contains nine genera of Hypermastigina class, one of which is known from termites, and three genera from the more primitive Polymastigina, two of which are known in termites. A single species of termites is not found to have a random protozoa population in nature and does not vary [Steinhaus, 1947; p531]. If ancient roaches interchanged protozoa with equal difficulty, it implies evolution of a numerous population for a considerable time until displaced by termites. There has been an interchange affected in the laboratory, though. Nevertheless, at first glance it would seem that this roach might represent the survivor of a once much more numerous family of insects which remained so for a long time. The wood roach has a more complicated digestive system than termites, which would seem to be further indication of this possibility. Of course termites could have evolved from non symbiotic sister group of wood roaches long after the wood roaches we know, by transfer of the protozoa, and this could have diversified rapidly long after wood roaches had first appeared. Nalepa and Thorne disagree as to the likelyhood of the transfer of symbionts between termites and roaches in the past [Thorne][Nalepa, 1991]. However in the course of several hundred million years transfer without a doubt occurred times without number without necessarily taking, so not only is either hypothesis possible but both may be involved on several occasions. The difficulty in this aspect of evolution is not with termite habits or protozoa evolution per se but more likely the nature of the molecules termites must be synthesizing in order to control the protozoa and prevent disease and "weeds". I doubt if we have enough information to resolve the matter at present. The likelihood that termites themselves are derived from a single ancestral group is especially possible if you define a termite as a roach with a soldier caste. However most digestive features probably existed before the coal hiatus.
The Australian wood eating cockroach secretes an enzyme which digests cellulose and is almost devoid of protozoa [Scrivener, et al., 1989]. This must surely be a late development after termite progenitors developed a soldier caste because a symbiosis would not have been necessary if this trait had developed early on. It is conceivable that some of the termites developed from a cousin of this roach and then subsequently some of them lost the enzyme. However it is very unlikely because the most primitive termites lack this enzyme and the soldier caste only evolved once [Rosin, 1994]. Termites which have a cellulose digesting enzyme, such as reticulitermes, produces only small amounts of glucose at the salivary gland [Watanabe & Noda]. Coptotermes and Nasutitermes secrete cellulase only from the salivary gland [Hogan]. Even if such an enzyme existed then in that form (as opposed to being evolved later), it would not likely have been important or interfered with protozoan evolution.
TERMITES
Any social roach which bored into the soil or logs would probably come to resemble termites in appearance. However what really separates termites from roaches is the development of a soldier caste. This is a fundamental development which would require much more than merely modifying or losing existing structures or appearances. A soldier caste probably evolved before a worker caste because Mastotermitidae and Kalotermitidae have no worker caste and soldiers only arose once in termites [Rosin, 1994][Thorne 1997]. A soldier caste could only be acquired in a social insect. A soldier caste was with out a doubt extremely valuable for survival of a slow growing insect with nymph mouth parts almost useless for defense. Cryptocercus takes 6 or 7 years to mature. This is more than six times as long as many other insects. This speed of other insects is probably because other insects are usually privileged to eat more nutritious food containing less available calories per protein. I suspect that the wood roach spread around the world when sea levels dropped toward the close of the Permian [Algeo & Seslavinsky, 1995] and reached the Southern Hemisphere near its close. This delay is plausible if the preponderance of their species did not fly, as Cryptocercus does not fly. If Brachiopod distribution is any indication, the latitudes were similar to what they are today [Stehli, 1970]. It is therefore conceivable that they spread across the Bering Sea land bridge. The Canadian arctic was warm early in the Permian (Beauchamp). If it was warm because of a Pacific Ocean current, then the crossing would possibly have to delay until later in the Permian because that implies no land bridge. If that same current still existed when a bridge became established, the southern edge of a Bering Sea bridge would presumably have been warm. It is likely that they crossed this bridge or its water gaps eventually since it is unlikely that they could have spread through Antarctica even though underground and saprophytic insects should be able to tolerate lower winter temperatures than others. The Indian Ocean was very cold in early Permian and judging by glaciers on its northern perimeter as far north as India may even have been filled solid and piled high with ice at that time [Hambrey and Harland]. If the ice was piled as high as a mountain range on its northern perimeter as it approached the moisture of the Tethys Sea, it could have been self perpetuating as the moist northern air rose up on a mountain of ice and dropped huge deposits of snow. It would be unlikely to be destroyed by warm Pacific Ocean currents because there was probably largely land between Australia and Antarctica and between South America and Antarctica. Indeed, there is evidence that these land bridges were largely intact up until mid Eocene [Scher and Martin]. Once the ice reached southern Africa no warm currents could reach the Indian Ocean from the Atlantic either. The aridity of South Africa [Smith] may have been from winds from the southeast drifting down off those glaciers and turning to the west from Coriolis forces. Currently geologists explain this with drifting continents, but shallow earthquakes on ridge - ridge transform faults distant from ridges, mid ocean plateaus, and insufficient trenches to remove the ocean floor among other anomalies make this seem implausible to me. In any case there are early Permian glacial deposits associated with marine deposits around the whole periphery of the Indian Ocean [Hambrey & Harland] and ice moved from west to east, not south to north, in Tasmania [Darlington p182]. An ice filled Indian ocean could explain why the other oceans were so much saltier during the Permian [Knauth]. However the warming at the Permian's close made almost any kind of migration conceivably possible then.
I also suspect that a soldier caste appeared in late Permian and is responsible for the coal hiatus or gap which commenced at its close and lasted for ten million years into the Triassic. It has been suggested that the soldier caste appeared early in the evolution of termites (Noirot and Pasteels, 1987). Since all primitive termites have flying reproductives, it is obvious that at least some of these ancient prototermites had flying reproductives also and it is possible that none of the wood roaches did. This would help explain why the coal hiatus appeared simultaneously all over the world at the Permian -Triassic boundary [Retallack, 1996, p196].
They may have been able to cross short water gaps for modern termites can cross short water gaps and the possibility that some were more enduring fliers then can not be ruled out. Termites' mating flights today can reach several thousand meters in the absence of a wind, although usually shorter. The warming at the Permian's close would have expedited such a migration, as will be discussed.
PROTOTERMITES
Ants did not appear until the Jurassic, probably descended from a parasitoid Ichneumon Scleroderma based on female polymorphism and similar family instincts combined with the complete lack of family instincts in wasps, as well as similar egg laying habits to ants [Malyshev, 1966, p198-228]. Unlike wasps there are no solitary ants [Bourke & Franks, 1995, p72]. So if roaches evolved a soldier caste at the start of the Triassic, below ground species must have been extraordinarily successful then with only solitary spiders, mantids, centipedes, and dragonflies to contend with. It is very likely that if we had been present in the Triassic we would have thought at first glance that some of them were indeed termites even though they probably retained the roach design of their wings, as does the Mastotermitidae termite today and Mastotermitidae also has similar gizzard, genitalia, and eggs in an ootheca packet [Gillott 1995 p170]. Living underground or inside a log involves pressure to form a thin skin which fossilizes poorly. So they would probably be poorly represented on the fossil record. If only their wings were preserved we would probably have called them roaches. I suspect that they spread through the soil and mulch as much as possible by colony fission as wood roaches do today and, if so, this would have considerably decreased further their fossilization. If winged forms were as successful after mating as I suspect, fossilization of wings would decrease, paradoxically, even further, since the very successful Amitermes meridionalis termite in Australia have very few winged adults or soldiers [Hill, p336] which is probably because of their success in saturating their niche. Eventually some or all of the wood roaches lost their wings since existing species have none [Spirivas 1996]. This, if it obtained, would have decreased fossilization of wood roaches further, since most insect fossils are wings.
When a soldier caste actually did appear in a winged prototype we would have had to call them termites (Isoptera) although we might have been reluctant to do so if we lived then. It is highly probable that termites gained their most potent instinct attributes during the Cretaceous. Fossils [Labandeira, 1993] and present day distribution support the last statement. They probably evolve so slowly [Wood & Sands, 1978, P248] that some fundamental social attributes must extend back into the Jurassic or further, at least primitive traits. Hasiotis, et al, have found burrows 0.15 to 0.5 cm in diameter and nests, which they interpret as termite nests up to one meter in diameter in late Triassic North America [Hasiotis, Peterson] [Hasiotis, Brown,Abston, Fig. 251-265], so the beginning of the Triassic is plausible for a termite soldier caste. Termites are extremely varied. Some eat wood. Others eat grass, cambium of trees, or animal dung. A very large number of species eat soil organic matter. The design of their digestive systems vary so drastically and fundamentally [Bignell (has diagrams)] that they must have been evolving for a long time before the Cretaceous by which time they had obviously become well diversified. The duration of the Jurassic would be by no means too long. Whether something resembling a soldier class appeared as far back as the early Triassic can not be determined from present fossils. Gay and Calaby believe that Stolotermes and Porotermes, primitive genera of the Hodotermitidae family, entered Australia through Antarctica in the Triassic [Gay & Calaby (1970), p395]. This is plausible because by the early Triassic the Indian Ocean had become considerably warmer as did the rest of the world. It is conceivable that the design of wood roach mouth parts changed sufficiently then that older nymphs could use them to some extent for defense even though they became less useful for eating hard materials, for defense can be very important to an insect trapped in the close confines of a log or soil cavity. If so, it would have made acquiring a soldier caste much easier. Perhaps when they are analyzed genetically their history will become clearer. The above two genera and the large primitive Zootermopsis and Mastotermes, which retain eyes, would be good genera to start with. There has been such a genetic analysis of the Cryptocercus wood roach and it has been found to be related to Blattidae roaches, which are a sister group to Polyphagidae roaches [Kambhampati 1996].
Removal of litter by ancient insects is quite plausible, for termites are extremely efficient at removing litter at present. Dung takes 25-30 years to be incorporated into desert soil in the absence of termites, while with termites, it disappears after the next rain [Whitford, 1986; p107]. Even large resistant items such as large logs have a fairly short life in tropical rain forests while logs in the north where freezing temperatures drastically inhibits the activities of termites, logs lay on the forest floor for many years. I suspect that it was these ancient roaches or more likely prototermites that brought the Carboniferous and Permian coal deposits to an end and produced the Triassic coal hiatus. Termites have then or since developed the ability to oxidize lignin [LaFage & Nutting]. The reduction of litter may have caused the extremes of moisture in the Permian and Triassic red beds because not only does litter and mulch considerably affect keeping soil moist and at an even temperature, but the litter's own extra moisture stores can cause a considerably enhanced transfer of moisture inland. The Triassic red beds spanned many climates [Retallack, 1996; 201] and started in late Pennsylvanian in North America [Veevers p192].
BOLIDE IMPACT
At the close of the Permian at least three extremely drastic extinction events took place on land and at sea. [Erwin, 1994] [Wignall & Hallam, 1993]. They were so sudden and so widespread and diversity of plants and animals was so much reduced on land and especially in the ocean, that they must have been caused by meteorites or comets containing very little iridium [Kajiwara et al., 1994, p375, 377] which overturned an anoxic (lacking oxygen) ocean [Knoll, et al, 1960]. There would have been better than an 80% chance of hitting the ocean. There was brief erosion which may have been from a tsunami [McLaren & Goodfellow, 1990; p139]. It is not likely that such erosion could have been from terrestrial causes because the tsunami that devastated Indonesia recently was probably as big as they get from earthquakes and it had very little affect on erosion. Such a huge tsunami would have brought carbon dioxide, methane, and poisonous hydrogen sulfide rich water to the surface at tens of thousands of miles of coastline when the wave broke. It would have spread the water over millions of square kilometers of land where the coast was flat. The explosion from a very large bolide (object from space) could have created a tsunami initially kilometers high or more. Such a tsunami would almost certainly be overturned in the immediate vicinity of the impact as well. The resulting tsunami could conceivably have been 50 to 100 meters high in the open ocean [Toon, et al, 1997, p51]. It would have reached tens of thousands of kilometers of coast line. In addition there may have been auxiliary small tsunamis from earthquakes and ejected material [Toon, et al, 1997, p52]. To gain some perspective of possible effects, consider that the tsunami which devastated Hawaii was 20 centimeters high in the open ocean [Toon, et al, p53]. It is thought that a tsunami 10 meters high can flood 20 kilometers inland against a flat plain [Toon, et al, 1997, p53]. There may have been as many as 5 or more surges in carbon dioxide across the years [Holser, 1987, p161] [Kajiwara, et al, 1994, p324]. If so, the worst and middle splashdown which ended the Permian must have taken place in the Atlantic Ocean because marine Permian organisms lasted into the Triassic in eastern China [Wignall & Hallam, 1993]. It is also thought that the Siberian flood basalts were triggered by a very large bolide [Alt, et al]. It is more likely that disruption of the crust at the antipode of a huge meteorite impact cause flood basalts. It is also possible that those flood basalts were caused by disruption resulting from massive earthquakes similar to the antipode to an impact violent earthquakes that have been proposed to have caused the Deccan traps at the Cretaceous close (although those traps were more likely caused by an impact in the ocean west of Peru). There is also an impact crater west of Australia that has the correct age.
TREES
Immediately above the boundary the glossopteris flora was suddenly [Hosher p173-174] largely displaced by an Australia wide coniferous flora containing few species and containing a lycopod herbaceous understory. Conifers had became common in Eurasia also, starting suddenly 302.5 million years ago with wild surges up and down before a steady rise starting 295 million years ago [Montanez]. Calipterids also rose then similar to conifers until starting a drastic decline starting around 278 million years ago [Montanez]. Maybe sucking insects were responsible for the last. Each of those groups of conifers arose from endemic species because conifers are very poor at crossing ocean barriers and they remained separated for hundreds of millions of years, largely to the present. Podocarpis was south and Pines, Junipers, and Sequoias were north, for instance. The dividing line ran through the Amazon Valley, across the Sahara, and north of Arabia, India, Thailand, and Australia [Florin, 1963][Melville, 1966].It has been suggested that there was a climate barrier for the conifers [Darlington, 1965, p168], although water barriers seem more plausible to me.
COAL HIATUS
If so, something which can cross at least short water barriers must have been involved in the coal hiatus. Climate could have been an important auxiliary factor, however. There was a spike of fern and lycopod spores immediately after the close [Retallack, 1995]. In addition there was also a spike of fungal spores immediately after the Permian-Triassic boundary [Eshet & Rampino, 1995 p969].This spike may have lasted 50,000 years in Italy and 200,000 years in China. If so, something besides an instant catastrophe must have been involved to cause the coal hiatus because funguses would surely have removed all dead vegetation in less than a few decades in most tropical places. Besides, the fungal spores rose gradually and declined similarly. There was also much woody debris. Each phenomenon would hint at widespread vegetative death. If the wood roaches and/or prototermites were responsible for a considerable part of this plant death in the years after an impact, it is obvious that they were not nearly as efficient at removing the remaining dead material before the funguses got to it at first as the termites probably were at the Cretaceous extinction and certainly are today. For one thing, they may have been largely confined to removing wood, mulch, and possibly roots because everything else was too poisonous or too out in the open where almost any predator, even dragonflies, could be effective. They, or something, must have become very effective by early Triassic because coal disappeared worldwide. Retallack, et al believe that extinction of peat plants caused it. This does not seem possible because coal formed extensively under rain forests with root traces under the coal. Whatever caused it must have operated in North America 25 million years sooner [Retallack 1996, p196].
Insects which attack the trunk are good candidates for lowering productivity because this is the single most damaging part of the plant to destroy. Such insects would have a considerable indirect effect on the speed of litter removal also, because removal of shade would heat the soil up greatly. The carbon dioxide in the atmosphere which had been rising erratically during the last of the Permian rose to a peak. This widespread plant death , fungal decay, and perhaps wood roach and/or prototermite digestion may have been a measurable part of the cause of that carbon dioxide rise, for there can be considerable carbon contained in the litter and vegetative cover if the thickness of those ancient coal deposits is an indication. During the succeeding Triassic coal hiatus the soil of south Australia was extremely low in organic matter, had no litter at all, and, while stump and root impressions were evident, there were no stumps or roots [Retallack 1997 p193,194] which looks as if they were removed by termites or roaches. Even so, it is doubtful if reduction of detritus could have been a major part of the rise in carbon dioxide. A considerable part of it was probably oxidation of those Pennsylvanian coal deposits which were eroded and oxidized off as the top layers of the Appalachian and Ural mountains rose up. There may have been a fair amount of carbon dioxide released from settling of calcium carbonate as shells [Ridgwell] before the extensive limestone and dolomite under those mountains became exposed and initially even after exposure, because calcium existed in the ocean associated with two bicarbonate ions, while only one carbonate ion precipitated. The reason why initial exposure may not have raised carbon dioxide much or even at all from erosion of limestone, may have been because the calcium was probably largely associated with organic acids while in the soil and rivers and preempted carbon dioxide when it reached the ocean. Scientists are uncertain about carbon net balance even in regard to present day mountains [Gaillardet].
For possible contributions by prototermites to the end Permian extinctions and discussion of the primary source of atmospheric carbon dioxide from ocean sediments see this site.
REFERENCES for this article are at the end.
Continue to Permian marine phosphorus as caused by amphibians, especially dragonflies or to Cretaceous marine phosphorus as caused by sheet erosion from runway building termites, or to the effect of runway builders and incompetent ants on the phosphorus of Cretaceous soils and vertebrate bones and teeth especially dinosaurs.
Cretaceous and Eocene marine phosphates are explained as arising from sheet erosion of termite runways in the article about termites affect on soil.. For more details of the termite effect in different areas see the termites' affect on soil around the Paleocene.
For a hypothesis which explains loss of silica from tropical soils by the alkaline gut of termites see “Did the alkaline gut of termites cause laterization of soils?”
For links to information about the Permian age see this site.
For those interested in dragonflies try IORI
----For a hypothesis that explains the large volcanoes of Mars and the bulges associated with them as the disruption from the antipode (opposite side of a sphere) of a huge meteorite or comet impact, see this site.
----It has been proposed that the Decca traps and other large lava flows were caused by violent movement of the crust in the antipode (opposite side of a sphere) opposite to a huge meteorite or comet impact..
----For a hypothesis that explains the gullies and canyons of Mars as erosion by rivers of silicone dust, click here.
----For a site that proposes a thin plate hypothesis to explain the plates in the crust of the earth, see this site. It has a link that explains the formation of ocean trenches.
----All you need to know about physical constants.
HEALTH ARTICLES
---- There is information here about how to obtain a very comprehensive book called “POTASSIUM NUTRITION” and thus cure or prevent rheumatoid arthritis, heart disease, gout, and high blood pressure and ameliorate diabetes and high blood potassium. It discusses requirements, amounts in foods, cooking losses, supplements, and physiology of potassium.
----For an article which proposes copper to prevent hemorrhoids, slipped discs, emphysema, anemia, and maybe gray hair access this URL ---- For an article which describes how to cure a tooth abscess with raw cashew nuts, see this URL, and for some observations on diabetes see this URL.
There is evidence that cell phones can produce tumors. Using remote ear phones would seem to be a good idea.
Fluoride in city water will cause fluorosis discoloration of teeth, weakened bones, damage to the kidneys and immune system, bone cancer, and, worst of all, damage to the nerves resembling Alzheimer’s disease.
For a hypothesis about human female evolution see this URL.
There is a free browser called Firefox, which is said to be less susceptible to viruses or crashes, has many interesting features, imports information from Iexplore while leaving Iexplore intact. You can also install their emailer. A feature that lists all the URLs on a viewed site can be useful when working on your own site. .
REFERENCES for ROACH
Algeo TJ Seslavinsky KB (1995) The Paleozoic world: continental flooding, hypsometry, and sea level. Am. J. Sci. 295, 787-822.
Alper J 1994 Earth's near-death experience. Earth Magazine, Jan94, Vol. 3 Issue 1, p42,
Alt D Sears JW Hyndman DW 1988 Terrestrial maria; the origins of large basalt plateaus, hotspot tracks, and spreading ridges. Journal of Geology 96; 647-662.
Alvarez W Claeys P Kieffer SW 1995 Emplacement of Cretaceous - Tertiary boundary shocked quartz from Chicxulub crater. Science 269; 930-935.
Axelrod (1969) Evolution of the Psilophyte paleoflora. p375-378 in; Papers on Evolution. (Erlich PR et al, eds) Little Brown & Co.Boston.
Anderson J Anderson H Fatti P Sichel H 1996 The Triassic explosion: astatistical model for extrapolating biodiversity based on the terrestrial Molteno formation. Paleobiology 22; 318-328.
Anonymous (1982) Scientific American 246; 78.
Anonymous (1993) Petrified nests reveal termite's true age New Sci. 138, 14.
Beauchamp, B. (1994) Permian climatic cooling in the Canadian Arctic. Geolog. Soc. Am Special Paper 288; 229-246.
Becker L Poreda RJ, Basu AR, Pope KO T. Harrison M, Nicholson C, Iasky R 2004 Bedout: A Possible End-Permian Impact Crater Offshore of Northwestern Australia. Science 10; 1126.
Berner RA Canfield OE (1991) A model for atmospheric CO2 over Phanerozoic time. Am. J. Sci. 291, 339-376.
Besly BM Fielding CR 1989 Paleosols in Westphalia coal – bearing and red bed sequences, Central and Northern England. Paleogaeography, Palaeoclimatology, Palaeoecology 70; 303-330.
Bignell DE Anderson JH 1988 Determination of the pH an oxygen status in the guts of lower and higher termites. Journal of Insect Physiology. 26; 183-188.
Bowrke AFG Franks NR 1995 Social Evolution in Ants. Princeton University Press.
Carlquist S (1965) Island Life. Natural History Press, Garden City,NY.
Chaloner WG Scott AC Stephenson J 199i Fossil evidence for plant – arthropod interactions in he aleozoic and Mesozoic. Philosophical Transactions: Biological Sciences, Royal Society 333; 177-184.
Cleveland LR Hall SR Sanders EP Collier J (1934) The wood-feeding roach Cryptocercus, its protozoa and the symbiosis between protozoa and roach. Mem. Acad. Arts & Sci. 17, 1-342.
Cox CB 1974 Vertebrate paleodistribution patterns and continental drift. Journal of Biogeography 1; 75-94.
Creber GT Chaloner WG (1985) Tree growth in the Mesozoic and early Tertiary and the reconstruction of paleoclimates. Paleogeography 52, 35-60.
Darlington PJ 1965 Biogeography of the southern end of the world. Harvard University Press, Cambridge Mass.
Dott, R.H. and Batten, R.L. (1971) Evolution of the Earth, 4th ed. McGraw Hill, NY.
Duncan IJ Titchener F Briggs DEG 2003 Decay and disarticulation of the cockroach: implications for preservetion of the Blattoids of Writhlington, UK. Palaios 18; 256-265.
Emerson AE 1935 Symbiosis between roaches and protozoa. Review of LR Cleveland monograph. Ecology 16; 116-117.
Emerson AE 1955 Geographical origin and dispersions of termite genera. Fieldiana: Zool. 37; 465-521.
Erwin DH (1993) The great Paleozoic crisis; Life and death in the Permian. Columbia University Press.
Erwin DH (1994) The Permo-Triassic extinction. Nature 367, 231-236.
Eshet Y Rampino MR (1995) Fungal event and palynological record of ecological crises and recovery across Permian-Triassic boundary. Geology 23, 967-970.
Faure K de Wit MJ Willis JP 1995 Late Permian global coal hiatus linked to 13C depleted CO2 flux into the atmosphere during final consolidation of Pangea. Geology 23; 507-510.
Florin R (1963) The distribution of Conifer and Taxad genera in time and space. Acta Horti Bergiani. 20, 121-312.
Frakes LA 1979 Climates Throughout Geologic Time. Elsevier, Amsterdam.
Gastaldo RA DiMichele WA Pfefferkorn HW 1996 Out of the ice house into the green house:A late paleozoic anaog for modern global vegetation change. GSA Today 6; (10) 1-7.
Gay FJ Calaby JH 1970 Termites of the Australian region. in; Krishna K Weesner FM eds. Biology of Termites, Vol. II Academic Press NY.
Gillott 1995 Entomology. Plenum Press.
Godfrey HCJ 1994 Parasitoid's Behavioral and Evolutionary Ecology. Princeton University Press, Princeton.
Grandcolas p 1996 The phylogeny of cockroach families; A cladistic appraisal of morpholanatomical data. Canadian Journal of Zoology 74; 508-527.
Gregor CB Garrels RM Mackenzie FT Magnard JB 1988 Chemical Cycles in the Evolution of the Earth. John Wiley.
Hambrey MJ Harland WB 1981 Earth's PrePeistocene Glacial Record Cambridge University Press, Cambridge.
Hasiotis ST Bown TM Abston CC 1994 Photoglossary of marine and continental ichnofossils. CD Rom computer disc Doc USC I 19m121.23 US Dept of the Interior, US Geological survey (Fig. 251-265 for termites, and Fig 180 for ants, and Fig. 163-167 for wasps). Note; because of MSDos format, disc is a little difficult to use.
Hasiotis, S.T. and Dubiel RF (1995) Termite (Insecta: Isoptera) nest ichnofossils from the upper Triassic Chinle formation, Petrified Forest National Park, Arizona. Ichnos 4: 12?-130.
Hasiotis ST Peterson CJ 1996 Termite (Insecta: Isoptera). Nests from the upper Jurassic Morrison Formation: evolutionary, paleoecolgical, paleoclimatic implications. Geological Society of America, Rocky Mountain Section, 48th Annual Meeting. Abstracts with programs. Geological Society of America Vol. 38, page 10, abstract #3843.
Hasiotis ST 2002 Continental Trace Fossils. Maybe purchased from www.sepm.org.
Haughton SH 1963 The Stratographic History of Africa South of the Sahara. Oliver & Boyd, Edinburgh.
Heydari E Hassanzadeh J Wade WJ Ghazi AM 2003 Permian-Triassic boundary interval in the Abadeh section of Iran with implications for mass extinction: part I-sedimentology. Palaeo 193;405-423.
Higashi MN Yamamura N Abe T Burns TP 1991 Why don't all termite species have a sterile worker caste? Proceedings of the Royal Society of London, Series B. 246; 25-29.
Hill GF 1942 Termites (Isoptera) from the Australian Region H.E. Daw, Govt. Printer, Melbourne, Australia.
Hogan ME Shulz MW Slaytor M Cozolii RT Obrien RW 1988b Components of termite and protozoal cellulases from the lower termite, Coptotemes lacteus Froggatt. Insect Biochem 18; 45-51.
Holldobler B Wilson EO 1990 The Ants. Belknap Press of Harvard University Press, Cambridge.
Holser WT Schonlaub H-P,Moses AJr Boekelmann K Klein P Magaritz M Orth CJ Fenninger A Jenny C Kralik M Mauritsch EP Schramm J-M Sattagger K Schmoller R 1989 A unique geochemical record at the Permian/Triassic boundary. Nature 337; 39.
Hosher WT Magaritz M Clark D 1987 Events near the Permian-Triassic boundary. Mod. Geol. 11; 155-180.
Isozaki I 1997 Permo - Triassic Boundary superanoxia and strattified superocean: Records from lost deep sea. Science 276; 235-276.
Keller L Genoud M. 1997 Extraordinary life spans in ants: a test of evolutionary theories of aging. Nature 389; 958-960.
Kenrick and Crane (1997) The origin and early evolution of plants on land. Nature 389: 33-39.
Kajiwara Y Yamahita S Ishida K Ishiga H Imai A (1994) Development of a largely anoxic stratified ocean and its temporary massive mixing at the Permian Triassic boundary supported by the sulfur isotopic record. Paleogeogr. Paleoclimatol. Paleocol. 111, 367-379.
Khambhampati S 1995 A phylogeny of cockroaches and related insects based on DNA sequences of mitochondrial ribosomal RNA genes. Proceedings of National Academy of Sciences 92; 2017-2020.
Khambhampati S (1996) Phylogenetic relationship among cockroach families inferred from mitochondrial 12s ribosomal-RNA genesequence. Syst. Entomology 21; 89-98.
Keller L Genoud M. 1997 Extraordinary life spans in ants: a test of evolutionary theories of aging. Nature 389; 958-960.
Knauth LP 1998 Salinity history of the earth's early ocean, Nature 395; 554-555.
Kvenvolden KA 1988 Methane hydrate - a major reservoir of carbon in the shallow geosphere? Chemical Geology 71; 41-51.
Knoll AH Bambach RK Canfield DE & Grotzinger JP (1960) Comparative earth history and late Permian mass extinction. Science 273, 452-458.
Labandeira CC Sepkoski CC 1993 Insect diversity in the fossil record. Science 261, 310.
Lo N Tokuda G Watanabe H Rose H Slaytor M Maekawa K Bandi C Noda H 2000 Evidence from multiple gene sequences indicates that termites evolved from wood-feeding cockroaches. Current Biology 10; 801-804.
Malyshev SI 1966 Genesis of the Hymenoptera. Methuen & Co. London
Martin MM 1991 The evolution of cellulose digestion in insects. Philosophical Transactions: Biological Sciences, Royal Society 333; 281-288.
McLaren DJ Goodfellow WD (1990) Geological and biological consequences of giant impacts. Review Earth Planet. Sci.. 18, 123-171.
Melville R (1966) Continental drift, Mesozoic continents and the migrations of the angiosperms. Nature 211, 116.
Miall LC Denny A 1886 Lovell Reeve & Co., London.
Montanez IP Tabor NJ Niemeier D DiMichelle WA Frank TD Fielding CR Isbell JL Birgenheier LP Rygel MC 2007 CO2-forced climate and vegetation instability during late Paleozoic deglaciation. Science 315; 87-91.
Mora CI Driese SG Colarrusso LA (1996) Middle to late Paleo atmosphere CO2 levels from soil carbonate and organic matter. Science 271, 1105-1107.
Nalepa CA (1988) Reproduction in the wood roach Cryptocercus punctulatus Scudder (Dictyoptera cryptocercidae). Entomol. Soc. Am. 81, 637-641.
Nalepa CA (1990) Early development of nymphs and establishment of hindgut symbiosis in Cryptocercus punctulatus (Dictyoptera: cryptocercidae). Entomol. Soc. Am. 83, 786-789.
Nalepa CA 1991 Ancestral transfer of symbionts between cockroaches and termites: an unlikely scenario. Proc. Royal Society of London B Biol. Sci. 246; 185-189.
Noirot CH 1969 Glands and secretions of termites p89-119 in; Biology of Termites. Krishnar K Weesner M, ed. Vol I, Academic Press NY.
Noirot C Pasteels JM 1987 Ontogenetic development and evolution of the worker caste in termite. Experientia 43; 851-860.
Otto-Bleisner BL Upchurch Jr.GR 1997 Vegetation induced warming of high latitude regions during the late Cretaceous period. Nature 385; 804807.
Prosser CL et al (1950) Comparative Animal Physiology. W.B. Saunders Co. Philadelphia.
Rampino MR Stothers RB 1988 Flood basalt volcanism during the past 250 million years. Science (ISSN 0036-8075), vol. 241, Aug. 5, 1988, p. 663-668.
Reichow MK Saunders AD White RV Pringle MS Al'Mukamedov AI Medvedev AI Kirda NP 2002 40Ar/39Ar dates from the west Siberian basin: Siberian flood basalt province doubled. Science 296; 1846-1849.
Retallack GJ (1995) Permian -Triassic life crises on land. Science 267, 77-79.
Retallack GJ Veevers JJ Morante R (1996) Global coal gap between Permian-Triassic extinctions and middle Triassic recovery of peat forming plants (review). Geological Society Am. Bull. 108, 195-207.
Retallack G 1997 Paleosols in the upper Narrabeen group of New South Wales as evidence of early Triassic paleoenvironments without exact modern analogs (review) Australian Journal of Earth Sciences 44; 185-281.
Ridgwell AJ Kennedy MJ Caldeira K 2003 Carbonate deposition, climate stability and neoproterozoic ice ages. Science 302; 859.
Robinson JM 1990 Lignin, land plants, and fungi: Biological evolution affecting Phanerozoic oxygen balance. Geology 10; 607-610.
Rosin Y (1994) Intragroup conflicts and the evolution of sterile castes in termites. Am. Nat. 143, 751-65.
Scrivener AM Slaytor M Rose HA (1989) Symbiont-independent digestion of cellulose and starch Panesthia cribrata Saussure, an Australian wood eating cockroach. J. Insect Physiol. 35, 935-41.
Shear WA 1991 The early development of terrestrial ecosystems. Nature 351; 285-289.
Smith RMH 1995 Changing fluvial environments across the Permian - Triassic boundary at Karoo Basin, South Africa and possible extinctions. Palaeogeography, Palaeoclimatology, Paleoecology 117; 81-104.
Smith J 2003 Geophys. Res. Lett. 10; 1029.
Snyder TE (1925) Notes on fossils. Proc. Biol. Soc. Wash. 38, 149-156.
Srinivas K Luykx P Nalepa CA (1996) Evidence for sibling species in Cryptocercus punctulatus, the wood roach, from variation in mitochondrial DNA and kariotype. Heredity. 76, 485496.
Stehli (1970) A test of the earth's magnetic field during Permian time. J. Geophys. Res. 75, 3325.
Steinhaus EA (1947) Insect Microbiology. Comstock Pub. Co., Ithaca NY.
Stothers RBRampino MR 1990 Periodicity in flood basalt, mass extinctions, and impacts; a statistical view and a model. in; Global Catastrophes in Earth History` - An interdisciplinary conference on impacts, volcanism, and mass mortality. Geological Society of America, Boulder Colorado.
Suess E, Bohrmann G, Greinert J, Lausch E 1999 Flammable ice. Scientific American 281; 77-83
Thorne BL 1990 A case for the ancestral transfer of symbionts between cockroaches and termites. Proc. Royal Society of London. B. Biol. Sci. 241;37-41.
Thorne BL 1991 Ancestral transfer of symbionts between cockroaches and termites: an alternative hypothesis. Proc. Royal Society of London B 246; 185-189.
Thorne BL Carpenter JM 1992 Phylogeny of the diictyoptera. Systematic Entomology 17; 253-268.
Thorne BL 1997 Evolution of eusociality in termites (review). Annual Review of Ecology and Systematics 28; 27-54.
Tilyard RJ (1917) Permian and Triassic insects from New South Wales. Proc. Linn. Soc. NSW 42, 721.
Tilyard RJ 1937 Kansas Permian insects.. Part XX the cockroaches, or order BlattariaI, II Am. Journal of Science 34; 169-202, 249-276.
Toon OB Turco RP Covey C 1997 Environmental perturbation caused by the impacts of asteroids and comets. Reviews of Geophysics (Am. Geophysical Union) 35; 41-78.
Vageler P (1947) An Introduction to Tropical Soils (Translation, Greene H) MacMillan & Co.
Veevers JJ Conaghan PJ Shaw SE 1994 Turning point in Pangean environmental history at the Permian / Triassic (P/Tr) boundary. Geological Society of America Special Paper, Klein GD ed. Boulder, Colorado.
Visscher H Brinkhuis H Kueracher W Looy C Van Konijnenburg-Van Cittert J 2001 Raised UV-B stress at the time of the end-Permian biosphere crisis. Am. Geophysical Union, fall meeting. Abstract PP21B-0482.
Walker, J.C.G. and Drever, J.I. (1988) Geochemical cycles of atmospheric gases. p55-76 in Chemical Cycles in the Evolution of the Earth. Gregor CB Garrels RM Mackezie FT Maynard JB, eds. John Wiley, NY.
Wang, Z.Q. (1996) Recovery of vegetation from the terminal Permian mass extinction in North China. Review of Paleobotany and Palynology 91: 121-142.
Wang ZQ 1996 Past global floristic changes – The Permian great Eurasian floral interchange. Palaeontology 39; 189-127.
Watanabe H Noda H Tokuda G1998 A cellulase gene of termite origin. Nature 394; 330-331.
Weesner FM 1960 Evolution biology of termites. Annual Review of Entomology. 5; 153-170.
Whitford WG (1986) Decomposition and nutrient cycling in deserts. in Pattern and Process in Desert Ecosystems. (Whitford WG ed) Univ. of New Mexico Press.
Wignnall PB Hallam AI (1993) Griesbachian (earliest Triassic) paleoenvironmental changes in the salt range, Pakistan and southeast China and their bearing on the Permo-Triassic mass extinction, Paleogeogr. Paleoclimatol. Paleoecol. 102, 215-237.
Wignall PB Twitchett RJ (1996) Oceanic anoxia and the end Permian mass extinction. Science 272, 1155-1158.
Wilde, P.E. and Berry B.N. (1984) Destabilization of the oceanic density structure and its significance to marine extinction events. Paleogeography, Paleoclimatology, Paleoecology 48: 143.
Wood TG Sands WA 1978 The role of termites in ecosystems. in; Production Ecology of Ants and Termites. Brian MV ed. Cambridge University Press.
Wright S 1931 Evolution in Mendelian Populations. Genetics 16; 97-159.
Zimmerman EC (1948) Insects of Hawaii, Vol. II. Univ. Hawaii Press.
updated January 2007
