Fig. 7. Shows the relative temporal and abundance patterns of a eurytopic taxon across a mass extinction boundary.

also noted as having broad adaptive ranges and high survival potential. Keller (1988a, 1989) noted that generalized, cosmopolitan planktonic foraminifera survived the K-T mass extinction interval, whereas the more specialized forms became extinct.

(iv) Persistent trophic resource exploiters. Mass extinction events are commonly associated with disruption and/or partial collapse of the global food chain (Kauffman 1977; Hsu et al. 1982). Mass extinction should most severely affect trophic specialists dependent upon food resources that are sensitive to changes in temperature, chemistry, light, moisture, etc. (e.g. upper-water-column plankton), and give higher survival potential not only to trophic generalists but also to those groups whose trophic requirements, no matter how specialized, are time-predictable and not affected by the causal mechanisms for mass extinction (e.g. benthic organic detritus). The most common trophic class among long-lived marine survivors are those based around non-specific detritus feeding (e.g. protobrachiate bivalves; Levinton 1974; Kauffman 1978; Sheehan & Hansen 1986).

Other common survivors are generalized algal-fungal grazers (many gastropods; Sohl 1987), selective detritus feeders (Teliinidae, many Nu-culanidae among bivalves; Kauffman 1977), and many deep-water microcarnivores (poromya-cean bivalves). Many of these groups, as well as trophic generalists, dominated survival faunas of the C-T (Elder 1987a, b; Harries & Kauffman 1990; Roy et al. 1990) and K-T boundary intervals (Hansen 1988; Hansen & Upshaw 1990; Roy et al. 1990).

Population dynamics

(i) Widespread (continuous or discontinuous) adult biogeographical dispersion. Extensive bio-geographic distribution by long-lived plankto-trophic larvae or adult mobility, without loss of gene flow, may ensure that at least some marginal populations will survive the large-scale perturbances of mass extinction intervals (Fig. 8). The cause of survivorship may be that a population is located beyond the range of mass extinction processes or due to the broader adaptive ranges of marginal populations. This may be especially applicable to taxa which have a biogeographical range extending from the Tropics towards the more polar regions. Jablonski (1986a) indicated that a broad geographic range was the best trait for survival, although at the clade level. Lenticulina rotulata (Lamarck), a C-T benthic foraminifera, was widely distributed prior to the mass extinction event and, in addition, had developed a spectrum of morphologies (polymorphism), possibly attributable to ecophenotypic variation, in response to the varying conditions inhabited by widely dispersed local populations. Expansion of the oxygen-minimum zone during the C-T Bonarelli Oceanic Anoxic Event (OAE II) reduced the once broad range of this species, but owing to its wide original dispersal, the species survived in certain regions which were unaffected and the species survived (Jarvis et al. 1988). Surviving populations were the more primitive forms (i.e. the stock from which subspecies proliferated). Palaeozoic brachiopods which survived mass extinction episodes were primarily generalized cosmopolitan forms (Bou-cot 1990). The final surviving ammonites and inoceramids in the C-T 'dead zone' and early recovery were among the most cosmopolitan lineages with widespread dispersion resulting from adult mobility (ammonites) and exceptional planktic larval drift (inoceramids; Kauffman 1975; Elder 1987a, b, 1989, Harries and Kauffman 1990).

Fig. 9. Shows the relative temporal and abundance patterns for a species that is able to remain in a dormant state during the ecologically stressed intervals of a mass extinction event. This allows the species to repopulate available niche(s) once conditions ameliorate.

Fig. 8. Shows the relative temporal and abundance patterns of a species with widespread dispersal. This model implies that several of the subpopulations may become extinct in certain regions, whereas others may survive. These surviving subpopulations can then expand into the available ecospace following the extinction boundary.


(i) Dormancy. If a group, such as the algae (see papers in Fryxell 1983, Vermeij 1987) or the dinoflagellates, is able to encyst during the onset of harmful environmental stress, they may be able to retain encystment long enough to survive the mass extinction event. An example of this may be the proliferation of the algae Renalcis in the Frasnian-Fammenian reefal facies of the Canning Basin (Playford 1980). This group replaced the previously dominant stromatopor-oids during this Late Devonian mass extinction. Other taxa, such as fresh-water sponges (e.g. Spongilla), can also encyst and survive for considerable periods of time. Many plants are able to produce seeds and spores which can maintain their vitality for considerable periods of time. Plants, in general, have higher survivorship than many terrestrial and marine animals across mass extinction intervals.

(ii) Bacterial chemosymbioses. Diverse marine

Fig. 9. Shows the relative temporal and abundance patterns for a species that is able to remain in a dormant state during the ecologically stressed intervals of a mass extinction event. This allows the species to repopulate available niche(s) once conditions ameliorate.

invertebrate taxa associated with submarine thermal vents, H2S and methane seeps, petroleum seeps, and H2S-enriched anaerobic substrates are now known to have symbioses with sulphide- or methane-oxidizing bacteria which provide both a source of Corg and 02 for vital physiological processes in otherwise inhospitable environments (Desbruyeres & Laubier 1983; Grassle 1983; Hessler & Smithey 1983; Cava-naugh 1985; Felbeck et al 1985; Hecker 1985; Hessler et al 1985; McLean 1985; Suess et al. 1985; Turncliffe et al. 1985). Kauffman (19886) has suggested that this unique ecosystem may have been much more pervasive in the past under more typical Phanerozoic conditions characterized by warmer, more equable climates, slower deep-ocean circulation with pervasive oxygen restriction, and more technically and methanogenically produced seepage on the seafloor. Because the taxa associated with these environments were basically immune to the environmental perturbations which decimated more normal trophic and 02-based physiological groups during mass extinction, they had very high survivorship potential. Further, because many of the molluscan lineages associated with these springs and seeps have representatives (species) which are also common in normal marine environments (with or without symbiotic bacteria), it is probable that these survivor groups could have served as the rootstocks for radiation into available normal marine ecospace following mass extinction events. Some species could potentially have moved back and forth between the two life modes to escape environmental deterioration associated with global biotic crises. Most of the Bivalvia commonly associated with well-documented Cretaceous (Howe & Kauffman 1986; Howe 1987) and modern submarine springs (Desbruyeres & Laubier 1983; Grassle 1983; Hessler & Smithey 1983; Cavanaugh 1985; Felbeck et al. 1985; Hecker 1985; Hessler et al. 1985; McLean 1985; Suess et al. 1985; Turncliffe et al. 1985) have bacterial symbioses, and have a long history of survival across many mass extinction intervals (e.g. Lucinidae, Mytilidae, Calyptogena, Inocer-amidae, Solemyacea, possibly Nuculidae and Nuculanidae; see also Seilacher 1990).

(iii) Skeletonization requirements. Major thermal and chemical changes in the world's oceans associated with mass extinction intervals are evident from stable isotope fluctuations of a rate and magnitude that well exceeds background levels (Hsii et al. 1982; Kauffman 1986, 1988a; Hut et al. 1987). These changes and possible large-scale fluctuations in the Calcium Compensation Depth (Worsely 1974; Arthur et al. 1987) could strongly affect the availability and physiological mineralization processes of shellbuilding (especially carbonate) materials. This, in turn, might favour, as survivor taxa, those groups with siliceous skeletons (diatoms, radio-larians, hexacxtinellids), organic-walled skeletons (dinoflagellates, worms, sponges, many arthropods), phosphatic skeletons (inarticulate brachiopods), chemically inert, external organic layers that are capable of protecting the shell (bivalve periostracum in deep-water taxa), agglutinated test-builders (certain foraminifera, many worms), and naked soft-bodied taxa (worms, coelenterates). Diatoms and dinoflagellates are dominant survivors among the plankton across the K-T boundary interval (Tappan 1979, Kitchell et al. 1986); bioturbation patterns attributable to various worms and chitinous arthropods pass through the C-T and K-T boundary without significant change (Ekdale & Bromely 1984). Sponges show little change across many mass extinction boundaries. Phos phatic inarticulate brachiopods are the disaster species of the basal Turonian (beginning of the C-T repopulation) in the WINA (Elder 1987a; Harries & Kauffman 1990). Agglutinated foraminifera show little change across the C-T and K-T boundaries in Europe (Koutsoukos & Hart 1988).


(i) Reproductive mechanisms. One of the key elements to survival is the ability to continue production of viable offspring during and immediately following mass extinction intervals. Listed below are some of the taxa to have developed different adaptations which can become especially beneficial in times of severe ecological and environmental stress (Jablonski & Lutz 1980).

(a) Taxa which are capable of producing an exceptionally high yield of larvae or offspring during periods of pronounced stress. The sheer number of offspring may help to insure that at least a fraction of them will survive. The Ostreidae are a group which utilize this reproductive strategy, and it is interesting to note that at a number of C-T (Elder 1987a; Harries & Kauffman, 1990; Harries, 1993) and K-T (Kauffman & Hansen in prep.) boundary sections, oysters are found in much greater abundances just prior to these extinction events and have high survivorship. Many ostreid lineage survived Mesozoic through Cenozoic mass extinction intervals.

(b) Young which are brooded, born at a more advanced juvenile stage, or develop from lecithotrophic larvae have adaptive advantages in that the young are more advanced when they settle than those requiring a long, exposed stage of metamorphosis. This is important because among ontogenetic stages the larval state is generally the least tolerant to stress in most groups, and mass extinction processes seem to affect the photic zone, where most plankto-trophic larvae reside, more than deeper waters where brooding, lecithotrophism and short-lived nektonic larvae dominate. Hence those taxa which produce even slightly more developed offspring have a survival advantage. An interesting example of this strategy has been suggested for the pattern of ostracod extinction at the C-T boundary. The podocopids suffered drastic extinction during the mass extinction, whereas the platycopids were much less adversely affected. It has been hypothesized that the platycopids brooded their young for the first few instaars within their tests, releasing them into the environment when they were consider ably less susceptible to the stressed conditions present during the mass extinction interval (Jarvis et al. 1988). Freshwater bivalves (Union-idae) and deeper marine Astartidae (Kauffman & Buddenhagen 1969) are known to brood their larvae, and they have very high rates of survivorship through Mesozoic-Cenozoic mass extinction intervals. Landman (1984) hypothesized that the reason for the extinction of the ammonites at the K-T boundary may be tied to their pelagic larvae, whereas the nautiloids, which survived this and many prior extinction events, employ an egg sack in which the young are protected during the early stages of development (Arnold 1987).

(c) Some taxa (especially deep benthic forms) reproduce only at long time intervals (tens of years), and the adult members are very long-lived and slow-growing, allowing them to pass through at least the most intense periods of the crisis in a semi-suspended physiological state. In this case, the individuals devote most of their resources and physiological energy to their own survival rather than to reproduction. But, once conditions ameliorate, the reproductive cycle is resumed. Deep-water molluscs (especially bivalves) employ this strategy (Turekian et al. 1975).

(d) Those taxa capable of sexual reproduction may have advantages over asexual reproduction in being able to accelerate the rate of potential evolution and adaptive change during stressed mass extinction intervals. This has been proposed to explain the greater survivorship of deeper water, large foraminifers (Braiser & Buxton 1988).

(e) Taxa which produce an egg resistant to desiccation and freezing, such as some of the ostracods (Whately 1990) may also be able to increase the number of offspring which survive.

(f) Another potential reproductive mechanism to promote survivorship is parthenogenesis. This requires that only one individual survive the extinction event, because each individual is furnished with the male and female reproductive organs and can reproduce alone. Certain groups of ostracods (the Cypridacea and the Darwinu-lacea) are capable of parthenogenesis (Whately, 1990).

(ii) Larval mechanisms. There are also variations in planktotrophic larval behaviour which may play a critical role in determining whether a species will be able to survive through a mass extinction crisis. Although Jablonski (1986c) saw no relationship between larval strategy and survivorship for Gulf Coast bivalves and gastropods, the parameters he used (solely plankto-

trophy v. non-planktotrophy) may not have been sensitive enough to discern the types of survival patterns listed below.

(a) Those larvae whose planktic habit involves a deep, rather than a shallow or surface position in the water column. The larvae's depth in the water column could help to buffer the effects of stress in cases where this was most intense at or near the air/water interface.

(b) Those larvae which are more eurytopic would also have a greater opportunity for survival.

(c) Larvae that have the ability to regulate their rate of metamorphosis could potentially increase their chances for survival. This implies that certain larvae may be able to speed up, delay or completely suspend their ontogenetic development until conditions are ameliorated or they reach a region that is less affected by stress. The bivalve Mytilus edulis employs this larval strategy and is able to drift for week in a physiologically suspended state under difficult environmental conditions before resuming metamorphosis when favourable planktic environs are reached (Kauffman 1975). Mytilidae, in general, have high survival rates at the species and lineage level during Mesozoic and Cenozoic mass extinction intervals.


By the luck of the draw, representatives of certain species are in the right place at the right moment for a few individuals to survive mass extinction intervals and to repopulate vacant ecospace rapidly following these crises (Fig. 10).


Given the various adaptive mechanisms (traits) for survival, combined with preliminary data on surviving taxa from a number of different mass extinction boundaries, the overall fabric of the survival and recovery intervals following a global biotic crisis is potentially much more complex than a scenario calling on explosive radiation from a few eurytopic and/or opportunistic groups. In a general sense, the combination of these numerous different survival mechanisms implies that repopulation can be extremely rapid and can involve genetically and ecologically more complex forms than previously hypothesized (Fig. 11).

The nature of the mass extinction event itself may have played a substantial role in determining the effectiveness of the various survival mechanisms. Data collected from the K-T boundary suggest that at least the main portion of the extinction events may have been essentially catastrophic (within a few Ka), whereas the data for the C-T and E-O boundaries suggest that the mass extinctions were step-wise or nearly so, and drawn out over 2-3 Ma. Thus, not only is the timing of the events different, but there is also considerable variation between the magnitude of the extinction and the overall survival and recovery patterns (Fig. 12). It should be stressed that these two models represent virtual end members of a spectrum, and that the majority of mass extinction events probably represent 'hybrid' extinctions.

Catastrophic extinctions are short-term, high intensity events that are much less selective and more pervasive than step-wise or gradual mass extinctions of similar magnitudes. Due to the shock effect of catastrophic forcing mechanisms, there are several predictions about the nature of survival that can be made. Of the mechanisms outlined above, the ones that may play major roles in survival through a short-term catastrophic mass extinction event are basically those that operate on ecological time scales. These include: chance, dormancy (which includes encystment as well as durable, protected germ cells), some opportunism, trophic strategies favouring taxa with predictable food resources (i.e. organic detritus), disaster species, and wide geographical dispersal. In addition, ecological generalism, certain larval strategies and chemosymbioses may also play important roles in determining the survivability of various taxa. Because these events are catastrophic and occur over a geologically (and perhaps even historically) short time-span, evolutionary mechanisms and abilities to withstand major population reduction should not be prominent. The resulting pattern of repopulation may be characterized by the increased span of survival and recovery intervals as well as their divisions (see Fig. 12).

However, step-wise or graded extinctions result in the breaching of a number of ecologically specific thresholds due to the lesser magnitude and greater selectivity of each extinction step in the progression. Step-wise extinctions contain partial normalization of environments and may even allow significant origination between extinction events, although few of these new taxa (crisis progenitors; see Kauffman & Harries this volume) range out of the extinction interval (Elder 1989). Extinction levels are not extensive enough to decimate the entire biota, and 'dead zones' following extinction steps are missing or very restricted in time and space (because there are more potential gene pools to provide new inhabitants of a devastated region).

0 0

Post a comment