260 HUGH S. TORRENS may represent bedding planes, unconformities, faults, or other significant lithological changes. Each unconformity-bounded 'package' of rock is called a 'seismic sequence'. Sequence stratigra- phy ultimately relies on the recognition of 'events', in this case supposedly generated by worldwide changes in sea-level, as revealed by such reflector horizons. Succinct introductions to the topic are provided by Prothero (1990, pp. 258-265) and Leeder (1999, pp. 258-266). Dott (1996, p. 244) has noted that this 'presently seems to be the dominant paradigm in sedi- mentary geology'. Larry Sloss was the pioneering figure here (Sloss 1963), which gives his current (dissenting) opinion - that such sequence boundaries have only local origins (Sloss 1991) - all the more credibility. If penetration of this technique into oil companies' research is taken to have occurred in 1975, we have the personal view of the chief protagonist (Peter Vail) as to how and when this revolution happened (Vail 1992). In Vail's opinion, the resulting 'renaissance of stratigraphy ranks in importance with the [other] plate tectonic revolution', which started at the same time, in the 1960s (Dott 1992, p. 13). Vail noted that the 1975 AAPG conference (Payton 1977) had been critical in advancing the speed of take-up of this new technique. There is now an enormous literature, involving both seismic, off-shore and non-seismic, land- or core-based, data, which it would be hard for one so ignorant as this author to review properly. However the real problem remains, as with impact as a cause of mass extinctions, that there is no consensus on the reliability and precision of sequence stratigraphy as a means of effecting time correlations. This much becomes clear from the writings of Sloss (1991), Miall (1997), and Wilson (1998). Miall has been particularly incisive in his dis- cussions of the limitations of sequence stratigra- phy and in a series of papers has questioned much of the methodology used, especially the relationship of these sequences to time (see Miall 1992, 1994, 1995). In particular Miall (1992, p. 789) demonstrated a minimum 77% successful correlation with the standard, Exxon chart using four columns of geological data. But these did not record actual geological data but pseudo-sections which had been randomly generated (see Fig. 2). Miall also pointed out that the claimed chronological precision of much of sequence stratigraphy is again greater than that of any available alternative and so is effectively untestable. While some sea-level changes clearly 'peaked simultaneously' across the (then Fig. 2. Miall's Correlation 'experiment' showing the 40 Cretaceous sequence boundaries (Fig. 2, centre column) of the 1988 Exxon global-cycle chart. These were compared with other event boundaries in four other 'sections' (Fig. 2, Nos 1-4). Table 1 (right) shows 'the high degree of correlation of all four sections with this Exxon chart', the lowest correlation success being with No. 3, at 77% fit. The catch is that all four of these test sections were constructed by random-number generation' (Miall 1992, p. 789)! smaller) Atlantic Ocean during the Cretaceous (Hancock 1993), it is notable that in the third edition of Miall's Principles of Sedimentary Basin Analysis (Miall 2000) the author plays down any supposedly worldwide eustatic control on such sequences, in favour of more local tec- tonic causes. Exactly this question - are such SOME PERSONAL THOUGHTS ON STRATIGRAPHIC PRECISION 261 sequence boundaries tectonic or eustatic in origin? - was being asked in 1991 (Aubry 1991). No consensus on the origins of sequence bound- aries, and thus the precision of their strati- graphic potential, has yet been reached. A fascinating discussion of the evolution of sequence-stratigraphic ideas has recently been published (Miall & Miall 2001), and should be required reading for all who study, or teach, stratigraphy. Impact: the ultimate event In May, 1979, the famous Alvarez extraterres- trial Cretaceous-Tertiary (K-T) impact theory was proposed (Alvarez 1979a; Alvarez et al. 1979). This was at first based only on a 20-25- fold increase in the abundance of iridium found in limestones in northern Italy. It was initially proposed with the expectation that this anomaly would prove to have been due to a supernova explosion, although the expected plutonium 244, osmium, and platinum increases had 'not yet been detected'. Soon afterwards, in September, 1979, the Alvarez team reported that this anomaly could not have been due to a super- nova, but that 'the 25 fold increase in iridium , which they found difficult to explain as an aspect of the sedimentary record at Gubbio, suggested that the Ir came from a solar system source, not a supernova' (Alvarez 1979b). Thus the evi- dence at first advanced in support of the K-T impact theory was entirely chemo-stratigraphic. In June, 1980, it was announced that the K-T iridium anomaly had now proved to be more widespread and was due to an asteroid impact (Alvarez et al. 1980, Alvarez 1983). 'Impactol- ogy' was born. Its influence throughout the whole of geology has since been incredible. One historian has written that impact carries 'gen- uinely revolutionary implications that are fatal to the uniformitarian principle itself (Marvin 1990, p. 147). The most impressive aspect, from a historical viewpoint, is the interdisciplinary nature of much, but not all, of the enormous amount of research which impact studies have inspired (Alvarez 1990). But it is notable that impactology was at first supported by chemical evidence, rather than the physical evidence that can best support it. Conway Morris urged more recently that eco- logical evidence must also be much more involved in such investigations, saying of the mass-extinctions of life at the K-T boundary, that 'at one level we can just as easily substitute the trigger for these extinctions being Martians waving laser-cannons rather than asteroids or a comet' (Conway Morris 1995, p. 292). In an inci- sive early review of the whole impact revolution, Van Valen rightly criticized the Alvarez's claims that their own evidence was experimental (i.e. 'hard') as 'misleading propaganda' (Van Valen 1984, p. 122). We must be concerned here only with the 'fallout' of impactology on stratigraphy. After the claim that a K-T impact event had been recognized, the search began to find the impact site. Two such craters have special interest for the imprecision with which they were first dated. One was the Duolun impact crater in China, reported in New Scientist (Fifield 1987). This briefly then became a candidate for a dino-extin- guishing event at the K-T boundary, if only in a English newspaper. But this impact-object, when dated, proved to have struck eighty million years too early (Ager 1993, p. 179)! This was not precise stratigraphy. The other candidate proved a more serious one. This was the Manson crater in northwestern Iowa, the largest - 35 km - crater then recognized in the United States. This was proposed as the K-T boundary candidate on the basis of 40 Ar/ 39 Ar dating of shocked microcline from the resulting structure (Kunk et al. 1989). Physical evidence The clearest evidence by which to confirm, and date, impact comes when not only the impact crater is preserved, or can be revealed by seismic and then borehole evidence (as in the case of Chicxulub, Mexico), but can also be partially dated by examining what it struck and whether the physical fallout from the impact can be docu- mented in the surrounding rocks, as in the case of the Manson microcline. Such physical evi- dence has been the subject of a fine review by Koeberl (1996), but which significantly ignored the many, often subtle, biochronological and extinction questions raised by such impact studies. When such physical evidence was prop- erly investigated for the Manson crater, it emerged that it could not have been the K-T 'killer crater'. A sanidine clast from the melt- matrix breccia of this impact gave a new date of c. 73.8 Ma. This was consistent with the bio- stratigraphic level into which diagnostically shocked, metamorphosed mineral grains had been found ejected in the stratigraphic record nearby, at a lower level in the Pierre Shale, of South Dakota (Izett et al. 1993). The Manson crater, like the Duolun Crater, proved to pre- date the features it was hoped it would explain - but here by 'only' 9 Ma. This again was impre- cise stratigraphically, and only demonstrated how important 'wishful thinking' could become in impact stratigraphy. 262 HUGH S. TORRENS The best documented example of the precise dating of a crater by its physical ejecta seems to be provided by Australia's - 160 km - Neopro- terozoic Acraman crater in South Australia (Gostin et al 1986; Williams 1986). It has the best documented crater-cuw-ejecta impact on record, although one too old to have had much perceivable biological effect. However Frankel, an enthusiast for impact as the causal agent behind most of the major geological extinctions, and hence of most System-level stratigraphic boundaries, notes that the possibility that [this] major impact wiped clean the biological slate and allowed new life- forms (e.g. the Ediacara fossil assemblage) to evolve must be seriously considered (Frankel 1999, p. 146). When this ejecta-recognizing approach was taken to the now celebrated K-T candidate, Chicxulub crater in Mexico, using diagnostic physical evidence, good evidence for the date and potential scale of a terminal Cretaceous impact there was uncovered. A marker-bed of large microtektites and the thickest ejecta layer known from this impact were found in several places nearby, like southern Haiti (Maurrasse & Sen 1991) in support of a major 'event' nearby. The potential stratigraphic scale of such impact events is indicated by the title of the International Geological Correlation Project (IGCP) No. 384. The first results of this project were published in 1998 under the title Impact and Extraterrestrial Spherules: New Tools for Global Correlation (Detre & Tooth 1998). The same project also started a new international journal in 1997, called Sphaerula. Impacts, if proven to be global in effect, must have real stratigraphic potential. The separate stratigraphic problem of distin- guishing multiple impacts often closely coupled in time has also emerged in the late Eocene record. Here two impacts have been docu- mented which are variously calculated to have been separated by anything between only 2 Ka (Glass 2000) to between 10-20 Ka (Vonhof et al. 2000). But at most sites where records of these two should be expected, either 'one of the ejecta layers is missing, or the two ejecta layers are indistinguishable' (Vonhof et al. 2000). This demonstrates the problems that the available stratigraphic record produces, even when, as here, there is great expectation of what is likely to be present. Any consensus on the extent, and biological effects, of the K-T boundary event remains obstinately polarized amongst geologists. Some prefer to see the cause of the extinction at this boundary as partly or wholly due to volcanic events over a much longer period of time than the short-lived event implied by impact. This volcanic scenario has a prehistory as well as a history. The history can be said to have started in 1985, with the paper by Officer and Drake (1985). The prehistory need only be taken back as far as Vogt 1972 (Courtillot 1999, p. 58). Such volcanism is now being proposed as an expla- nation for other second-order mass extinctions, like the Karoo-Ferrar flood basalt volcanism to explain an early Jurassic extinction (Palfy & Smith 2000). Work using physical evidence of impact is in stark contrast to some of the earlier evidence proposed to explain the first, merely chemical, discoveries of K-T iridium anomalies, with associated concentrations of phosphatic fossils, in the 'fish clays' of Denmark. These were immediately used to prove the impact must have occurred near Denmark. The most extraordi- narily subtle ocean currents had then to be invoked to explain the more fishy aspects of the evidence found here (Allaby & Lovelock 1983, pp. 95-99). The paper by Rocchia et al. (1990) was crucial in indicating that the iridium anomaly at the original, Gubbio, locality in Italy was much more extensive stratigraphically (and thus must have lasted 'longer') than had previ- ously been realized (see Fig. 3). The problem of anomalous iridium concen- trations must depend on how complete the stratigraphic record can be shown to be at the different localities that show such anomalies. This must now be our final consideration. How complete is the stratigraphic record? Nearly a century ago Buckman reminded us of the vital importance of separating sedimentary from chronological records in stratigraphy: 'the amount of deposit can be no indication of the amount of time, the deposits of one place correspond to the gaps of another' (Buckman 1910, p. 90). On the related question of the ade- quacy of the sedimentary rock record, Buckman noted earlier how fossil: species may occur [together] in the rocks, but such occurrence is no proof that they were contemporaneous . . . their joint occurrence in the same bed [may] only show that the deposit in which they are embedded accumulated very slowly (Buckman 1893, p. 518). The basic truth of these statements is still often ignored. The abundance of any particular material, element, mineral, chemical or fossil, in the stratigraphical record need not prove either Fig. 3. Whole-rock iridium concentrations across six metres of rock straddling the K-T boundary at Gubbio, Italy, with the K-T boundary (KTB) marked. Concentrations of Ir in limestones are much lower than Ir concentrations in shales which stand out as maxima. The existence of such Ir spikes in shales is not due to the occurrence of isolated 'Ir events', but to post-depositional enhancements related to dissolution of carbonates' (Rocchia et al 1990, pp. 214-215). SOME PERSONAL THOUGHTS ON STRATIGRAPHIC PRECISION 263 264 HUGH S. TORRENS Fig. 4. A cumulative diagram demonstrating 'pelagic sedimentation in the ocean', from Hay (1974, Fig. 2). the origin, or the contemporaneity, of that material. Attempts to assess the 'stratigraphic completeness of the stratigraphic record' by using timescales based on sedimentation rates as proposed by Schindel (1982) or Sadler & Strauss (1990) prove inappropriate because they take no account of the many gaps, erosion surfaces and all the other complexities of what has been called litho-chronology by Callomon (1995, p. 140). Similarly doomed are some of the attempts to assess the origins of some concentrations of fossils, whether of Palaeozoic nautiloid cephalopods (Holland et al. 1994) as the remains of fossils that lived together 'in schools' and then 'suffered mass mortality', or the geologically later 'belemnite battlefields' (Doyle & Macdon- ald 1993). These latter may be post-mortal accumulations of a nearly original ecological assemblage, as proposed, but they may as well be entirely condensed and accumulated over much longer periods of time, and concentrated together only because of the lack of any sedi- mentary dilutant, as in the fossil 'cemeteries' that Buckman worked on. The presence of a 'cemetery deposit' of fossils can never prove those fossils suffered a catastrophic death. Similar considerations apply to the Danish K-T boundary 'fish clay' or the 'fish mortality horizon' which was claimed 'may represent the first documented, direct evidence of a mass kill event associated with the bolide impact' at the K-T boundary on Seymour Island. These may equally have had secondary, condensed, and thus residual, origins, rather than a primary origin as a 'mass kill associated with an impact event' proposed for Seymour Island. The first, condensed origin, was rejected as an explanation here only because of the fish horizon's 'inescapable' relationship with an iridium anomaly below it (Zinsmeister 1998). 'Anomalous' abundances of iridium also need not have impactal origins. Some can have been derived through condensation, as Rampino orig- inally noted (1982), and as Hallam (1984) and Ager (1993) have more recently supported. One only has to follow the diagram showing the pro- cesses involved in getting such normal, but still cosmic, iridium deposited in pelagic sedimenta- tions on the ocean floor given by Hay (1974, p. 3) to realize how such an insoluble material as cos- mically derived iridium-rich dust might end up, condensed and isolated, on ocean floors. Most other potential dilutants would simply have been removed by chemical solution on their way down to the sea-floor (see Fig. 4). SOME PERSONAL THOUGHTS ON STRATIGRAPHIC PRECISION 265 The surprises in this field might be first, how different the past might prove from the present, in matters involving compensation depths and solubilities of organic materials; and second, how very condensed and incomplete pelagic deposits can prove to be. We need careful strati- graphic studies of abyssal clays with overall low accumulation rates, such as Kyte & Wasson's (1986) study of a thickness of only 24 metres ranging over more than 70 Ma from the central North Pacific. This gave confirmation of a major impact event having been recorded here, by showing that in this condensed abyssal sequence there was a significant, and surely here primary, increase in Ir concentration at the K-T boundary. At more distant sections in rocks of shallower water origin (such as Stevns Klint, Denmark), analysis showed how: a pulse of calcite dissolution in shallow water coincided precisely with the era [K-T] bound- ary, and [that] this event played a major role in the formation of the Fish clay in eastern Denmark, which is a condensed series of smectitic clay-rich layers from which much calcite has dissolved. [Such evidence sug- gested that] no single catastrophe can account for the major biotic extinctions which occurred at the end of the Cretaceous period [here] (Ekdale & Bromley 1984). In other words there are anomalies and anomalies, which need to be carefully and separ- ately analyzed. It was at a Danish locality that the 160-fold increase of 'anomalous iridium', the highest recorded in the original research, sug- gested it had to have had a sudden, extra-terres- trial origin (Alvarez et al. 1980, p. 1100; Frankel 1999, pp. 19-21). Its extra-terrestrial origins need not be in dispute, but stratigraphers need to ask if all such extra-terrestrial material had to have arrived suddenly, through impact, or could have arrived by more slowly accumulated concentration. The same problem emerged at El Kef, in Tunisia, chosen in 1989 as the Global Stratotype Section and Point (GSSP) for the base of the Danian, and thus the Cenozoic (Cowie et al. 1989, p. 82). The question was again: how com- plete is the critical K-T section at this boundary here? Its great incompleteness has been con- firmed by MacLeod & Keller (1991), and in a more recent paper by Donze et al. (1996). None the less, this region is still regarded as 'unique in its documentation of one of the most critical intervals of Earth history. The most complete succession [here] is however that of El Kef [GSSP for the Danian]' (Remane 2000b). The same situation re-emerged at the first K-T iridium anomaly locality, Gubbio in Italy, when a more extended vertical extent of 'the iridium anomaly' was investigated. Here Ir associations with clay minerals were thought due 'to post- depositional enhancements related to dis- solution of carbonates in a sequence characterized by a low sedimentation rate' (Rocchia et al. 1990; see Fig. 3). The same problem faces the claim that the iridium anomaly detected in the English Ludlow Bone Bed, Upper Silurian, had a single primary, impactal origin. This occurrence again demon- strates a secondary, condensed, origin (Schmitz 1992; Smith & Robinson 1993), like some of the 'anomalous' sequences known at the K-T boundary. The real problem, as with sequence stratigra- phy, is the difficulty of achieving accurate cali- brations of rates and durations of many of these geological processes and, or, events, as Dingis (1984) has pointed out. Indeed, the initial idea of using iridium concentrations, to the single- minded extent that was first proposed by the Alvarezes, as a sedimentary rate-metre (Frankel 1999, p. 19), has now been re-invented as a means of measuring rates of sedimentation, and to prove the completeness of sequences contain- ing iridium 'anomalies' (Bruns et al. 1996, 1997). This marks a return to the original, pre-impactal, intentions of the Alvarez team before their work revealed 'over anomalous' amounts of iridium. One man's anomaly has become another's nor- mality. Wallace (1991) and Sawlowicz (1993) have discussed different ways in which iridium can become 'anomalously' abundant in sedi- ments. Another real problem when discussing strati- graphic precision is again conceptual. A recent paper on dinosaur abundances near their critical terminations in Montana and North Dakota was highlighted on the cover of Science. It sup- posedly proved, of dinosaur remains found here close to the terminal Cretaceous boundary, that 'Dinosaurs were going strong till the last minute [of the Cretaceous]' (Sheehan et al. 2000). But space and time are not the same, even in a science as unscientific as geology! Buckman had noted in 1893 how fossil 'species may occur together in the rocks [e.g. in space], yet such occurrence is no proof that they were contem- poraneous [e.g. in time]' (Buckman 1893, p. 518). Others have added to this confusion. Gould (1992), when discussing extinctions at the K-T boundary at Zumaya, Spain used an ammonite found spatially 'within inches' of that boundary to prove these ammonites had become extinct at the time of that boundary. Hudson 266 HUGH S. TORRENS (1998, p. 414), noting two occurrences a short distance (less than 1 metre), whether below the boundary clay in Montana or the Raton For- mation, asked 'can either distance be regarded as "well below" the boundary?'. The answer to this rhetorical question depends on precise separation of those quite different entities; space and time. This problem has now also reached the museum. A recent acquisition on display at the Manchester Museum, England, excavated from underground caves at Guelhemmerberg, near Maastricht in November 1999, claims to 'record the exact point in time of the end Cretaceous extinction when many animals, including the dinosaurs, became extinct' (Anon. 2000, pp. 15-16). If only the stratigraphic record could so precisely record such matters! Conclusion The Quo Vadis conference of 1982 urged on par- ticipants the need for: a better understanding of the degree of accu- racy and precision that can be reached in regional and global correlations, and more insight into the nature and interrelation of physical, chemical and biological processes in space and time (Seibold & Meulenkamp 1984, pp. 65-66). While discussing the problems of using eusta- tic events in stratigraphy Dott pointed out in 1992 that: one of the consequences of the renaissance of stratigraphy during the past two decades [using such a wide range of techniques] has been the rekindling of enthusiasm for eustasy and for cycles of several kinds. This has even resulted in a fervent new orthodoxy, which Sloss (1991) has appropriately dubbed 'neo- neptunism' (Dott 1992, p. 13). The general incompleteness of the strati- graphic record in the Eocene was specifically commented upon by Aubry (1995) who, in a later important abstract, also reminded us of the vital consequences for both sequence stratigra- phy and geochronology of the stratigraphic record being, as it is so often shown to be throughout the geological record, incomplete. She noted that the challenge for the next decade was to docu- ment further the architecture of the strati- graphic record using the temporal component as an essential component, a fact that sequence stratigraphy has somehow failed to recognize (Aubry 1996). Van Andel (1981) and Bailey (1998) have equally urged a reappraisal of those features of the rock record such as 'perceived cycles and sequences', because of the sheer complexity of that record which often embraces gaps and in which record there may often be 'more gap than record'. Zeller (1964) in a fascinating paper has equally shown how easy it is, through human nature, to discern cycles in stratigraphy. The critical point is that, amid all the wars of words about 'hard' and 'soft' science, or whether 'all science is either physics or stamp collecting' (as Ernest Rutherford memorably said (Birks 1962, p. 108)), no consensus on either the cause, the extent, or precise timing of the extinctions, even at the K-T boundary, has yet emerged (Glen 1994, Courtillot 1999, Frankel 1999). There is a near consensus that there was a large impact at or near the K-T Boundary in Mexico. But its effect on terminal Cretaceous life around the world is much less clear and perhaps must remain so. The lack of consensus becomes clear by comparing the detailed biostratigraphic data assembled by MacLeod et al (1997), with the response from Hudson (1998). The authors of a recent paper (Albertao et al. 2000) were duly forced to draw the K-T bound- ary at two quite different stratigraphic horizons in NE Brazil when trying to define this boundary there, depending on whether biological data or physical evidence were invoked. This was another site which provided 'no direct evidence for an impact origin'. One gets a clear view of the lack of consensus by comparing the American view of the debate given by two of its main American protagonists (Alvarez 1997; Frankel 1999) with that of a French competitor (Cour- tillot 1999). The need to return to more careful assessment of all temporal components in stratigraphy is the most important lesson from all the new strati- graphies, in which the last fifty years have been so prolific. One cause for some future optimism is the way in which graphic correlation (Shaw 1995), which uses statistical analysis of first and last appearances in ranges of fossil taxa, has been demonstrated as a means of investigating the degree of completeness of incomplete sequences (Macleod 1995a, b). Another is the potential of the methods used by Mc Arthur et al (2000a) in integrating strontium isotope profiles to document durations of geological events, with the ammonite biozones used in biochronology. The future lies, not in complaining about 'the current imprecisions of biostratigraphical corre- lation' (Jeppsson & Aldridge 2000, p. 1,137) but, in integrating stratigraphical studies in the way McArthur et al. (2000a) have demonstrated. SOME PERSONAL THOUGHTS ON STRATIGRAPHIC PRECISION 267 Localities (Fig. 5) Resolution: Stages scope* number % completeness Resolution: Zones scope number % completeness Resolution: faunal horizons scope number % completeness 1 BB 3 3 100 14 11 78 56 20 36 2 Ch 3 3 100 14 8 57 56 18 32 3 WH 3 3 100 14 11 78 54 21 39 4 HP 3 3 100 14 9 64 56 23 41 5 Be-CF 3 3 100 t 11 6 43 45 14 31 6 Se 3 3 100 t 9 8 89 37 10 27 7 LH/Hh 3 3 100 14 9 64 56 21 38 g BA 3 3 100 14 9 64 56 22 39 9 SL 3 3 100 t 10 8 80 42 20 48 10 Cl t 1 1 100 8 8 100 32 22 69 11 Ob t 2 2 100 7 7 100 29 20 69 12 Br-L 3 3 100 14 9 64 56 22 39 13 Du 3 3 100 14 11 78 56 29 52 Average 100 74 43 * Only the Lower Bathonian is represented in the Inferior Oolite. But even at Substage level (Lower and Upper Aalenian, Lower and Upper Bajocian, Lower Bathonian), at which the maximum scope would be 5, the representation would be everywhere 100% complete. t These sections have exposed only parts of the Inferior Oolite, either cut off at the tops by erosion or covered at the base. Fig. 5. The three differing 'completenesses' of the geological record, in percentages, as revealed using three different levels of resolution, based on ammonite biochronologies, in the Inferior Oolite of southern England. At Stage level (e.g. Aalenian, Bajocian, Bathonian, etc.) all thirteen sections show complete records where rocks of these ages are exposed (average 100%). At the next lowest, Zonal, level of resolution, completeness varies from 100% to 43% (average 74%); while at the lowest available, Faunal Horizon, level, completeness varies from 69% to 27% (average 43%) (Callomon 1995, p. 147). Only when such integrated studies are prop- erly attempted may we be able to start to investi- gate the biological consequences of some of the more extraordinary events to which the Earth has been subjected over its long history. Until then stratigraphy will indeed remain a 'science in a crisis' (Glen 1994). For as Buckman (1921, p. 2) so presciently recorded long ago: 'additions to fauna decrease the imperfection of the zoo- logical, but increase that of any local geological record: the gaps caused by destruction stand revealed more plainly'. Buckmans's claim has been entirely confirmed by Callomon (1995; see Fig. 5). It does indeed seem that the harder you look at rocks the less complete their record of the passage of time becomes. Van Andel has said the same. To him, it: appears that the geological record is exceed- ingly incomplete and that the incompleteness is greater the shorter the time-span at which we look. [He too urges] 'the need for a vastly increased care in stratigraphy and chronology' (Van Andel 1981, p. 397). I thank the editor, D. Oldroyd (Sydney), for his attempts to guide this difficult paper through the edi- torial process. As Dietz (1994, p. 8) has noted: 'scien- tists now know more and more about less and less'. This is particularly true in stratigraphy. The same move (also confirmed by Dietz), which has taken geology from the field into the laboratory, has had a similarly negative effect on academic library provision, which has caused new difficulties. In the face of so much 'information', more and more literature gets locked or thrown away. The Senate's reaffirmation of library dis- posal policy at my former university in November, 1999, makes chilling reading to all of us who care about even recent history. It read: 'old and superseded texts can be misleading or worthless and unsought material can obstruct the search for relevant items'. My attempt to combat such attitudes in this paper has also had to be biased towards those parts of the stratigraphic column with which I have experience. It has been equally influenced by a lifetime spent attempting to teach the central importance of stratig- raphy to declining numbers of students of geology. I thus hope this paper will provoke as much as it informs. I have also tried to repay a long-held debt to J. Cal- lomon (London), who first showed me how subtle and complex the stratigraphic record so often is. For specific help I thank W. Cawthorne (London), C. Lewis (Macclesfield) and G. Papp (Budapest). I am most grateful to A. Rushton (Keyworth), R. Dott (Wiscon- sin) and B. Webby (North Ryde, New South Wales) who were all sufficiently provoked to make many com- ments on earlier versions, which has improved it. References AgER, D. V. 1963. Principles of Paleoecology. McGraw-Hill, New York. AgER, D. V. 1973. The Nature of the Stratigraphic Record. Macmillan, London (3rd edn, John Wiley, Chichester). AgER, D. V. 1986. A reinterpretation of the basal 'Lit- toral Lias' of the Vale of Glamorgan, Proceedings of the Geologists' Association, 97, 29-35. AGER, D. V. 1987. The excitement of traditional stratigraphy. Geology Today, July-August, 116-117. 268 HUGH S. TORRENS AGER, D. V. 1993. The New Catastrophism. Cambridge University Press, Cambridge. AGER, D. V. & CALLOMON, J. H. 1971. On the Liassic age of the "Bathonian" of Villany (Baranya). Annales Universitatis Scientiarum Budapestinenis. . . . Sectio Geologica, 14, 5-16. ALBERTAO, G. A., MARINI, K, OLIVEIRA, A. D., DELICIO, M. P. & MARTINS JR, P. P. 2000. Peculiarities con- cerning the KIT Boundary in N E Brazil. Paper/poster presented to the IGC at Rio de Janeiro, Session 25-6. ALGEO, T. J. & WILKINSON, B. H. 1988. Periodicity of Mesoscale Phanerozoic sedimentary cycles and the role of Milankovich orbital modulation. Journal of Geology, 96, 313-322. ALLABY, M. & LOVELOCK, J. 1983. The Great Extinc- tion. Seeker & Warburg, London. ALVAREZ, L. W. 1983. Experimental evidence that an asteroid impact led to the extinction of many species 65 million years ago. Proceedings of the National Academy of Sciences, U. S. A., 80, 627-642. ALVAREZ, L. W, ALVAREZ, W, AsARO, F. & MICHEL, H. V. 1980. Extraterrestrial cause for the Cretaceous-Tertiary extinction. Science, 208, 1095-1108. ALVAREZ, W. 1979. Dinosaur extinction possibly linked to extra-terrestrial cause, Episodes, 2, July 1979, 30. ALVAREZ, W. I919b. Anomalous iridium levels at the Cretaceous/Tertiary boundary at Gubbio, Italy. In: CHRISTENSEN, W. K. & BIRKELUND, T. (eds) Proceedings of the Cretaceous-Tertiary Boundary Events Symposium, 2. Copenhagen University, Copenhagen, 69. ALVAREZ, W. 1990. Interdisciplinary aspects of research on impacts and mass extinctions: a per- sonal view. Geological Society of America Special Paper 247, 93-97. ALVAREZ, W. 1997. T. rex and the Crater of Doom. Princeton University Press, Princeton. ALVAREZ, W, ALVAREZ, L. W, As ARO, E & MICHEL, H. V. 1979. Experimental evidence in support of an extra-terrestrial trigger for the Cretaceous- Tertiary extinctions, EOS, 60, 734. ANON. 2000. Manchester Museum, Annual Report, August 1999-31 July 2000. Manchester. ARKELL, W. J. 1957. Treatise on Invertebrate Paleon- tology. Part L, Mollusca 4, Cephalopoda Ammonoidea, University of Kansas and Geo- logical Society of America, Kansas & New York. AUBRY, M P. 1991. Sequence stratigraphy: eustasy or tectonic imprint? Journal of Geophysical Research, 96, 6641-6679. AUBRY, M P. 1995. From chronology to stratigraphy. In: BERGGREN W. A., KENT, D. V., AUBRY, M P. & HARDENBOL, J. (eds) Geochronology, Time Scales and Global Stratigraphic Correlation, Society for Sedimentary Geology, Special Publications 54, 213-274. AUBRY, M P. 1996. On the incompleteness of the Stratigraphic record: implications for sequence stratigraphy and geochronology, Abstracts of the 30th International Geological Congress, Beijing, 2, 10. AUBRY, M P., BERGGREN, W. A., VAN COUVERING, J. A. & STEININGER, F. 1999. Problems in chronos- tratigraphy, Earth-Science Reviews, 46, 99-148. AUBRY, M P., VAN COUVERING, J. A., BERGGREN, W. A. & STEININGER, F. 2000. Should the golden spike glitter? With comments and a response. Episodes. 23, 203-214. AUBRY, M P., HAILWOOD, E. A. & TOWNSEND, H. A. 1986. Magnetic and calcareous-nannofossil stratigraphy of the lower Palaeogene formations of the Hampshire and London basins. Journal of the Geological Society, London, 143, 729-735. BAILEY, R. J. 1998. Stratigraphy, meta-stratigraphy and chaos. Terra Nova, 10, 222-230. BAKER, N. 1997. The Size of Thoughts. Vintage, London. BERRY, W. B. N. 1987. Growth of a Prehistoric Time Scale. Blackwell, Palo Alto. BIRKS, J. B. (ed.) 1962. Rutherford at Manchester. Haywood, London. BRASIER, M., COWIE, J. & TAYLOR, M. 1994. Decision on the Precambrian-Cambrian boundary strato- type. Episodes, 17, 3-8. BRETT, R. 2000. Frontiers of life, Brazil 2000 IGC News, 1-3. BRICE, W. R., 1989, Cornell Geology Though the Years. Cornell University Press, Ithaca. BRUNS, P., RAKOCZY, H., PERMCKA, E. & DULLO, W C. 1997. Slow sedimentation and Ir anomalies at the Cretaceous/Tertiary boundary, Geologische Rundschau, 86, 168-177. BRUNS, P., DULLO, W C, HAY, W. W, WOLD, C. N. & PERMCKA, E. 1996. Iridium concentration as an estimator of instantaneous sediment accumu- lation rates. Journal of Sedimentarv Research, 66, 608-612. BUCKMAN, S. S. 1889. On the Cotteswold, Midford and Yeovil Sands. Quarterly Journal of the Geological Society, London, 45. 440-474. BUCKMAN, S. S. 1893. The Bajocian of the Sherborne district. Quarterly Journal of the Geological Society, London, 49, 479-522. BUCKMAN, S. S. 1901. Jurassic brachiopoda. Geological Magazine, Decade 4, 8, 478. BUCKMAN, S. S. 1910. Certain Jurassic ('Inferior Oolite') species of ammonites and brachiopoda. Quarterly Journal of the Geological Societv, London, 66, 90-110. BUCKMAN, S. S. 1921 Type Ammonites, 3 (Part 30). The Author, Thame. BURCHFIELD, J. D. 1975. Lord Kelvin and the Age of the Earth. Macmillan, London. CALLOMON, J. H. 1984. The measurement of geological time. Proceedings of the Royal Institution of Great Britain, 56, 65-99. CALLOMON, J. H. 1995. Time from fossils: S. S. Buckman and Jurassic high-resolution geochron- ology. In'. LE BAS, M. J. (ed.) Milestones in Geology, Geological Society, Memoir No. 16, 127-150. CHALLINOR, J. 1978. A Dictionary of Geology. Uni- versity of Wales Press, Cardiff. CONWAY MORRIS, S. 1995. Ecology in deep time. Trends in Ecology and Evolution, 10, 290-294. COPE, J. C. W., GATTY. T. A., HOWARTH, M. K SOME PERSONAL THOUGHTS ON STRATIGRAPHIC PRECISION 269 MORTON, N. & TORRENS, H. S. 1980. A correlation of Jurassic rocks in the British Isles, Part One. Geological Society of London Special Report, 14, 1-73. COURTILLOT, V. 1999. Evolutionary Catastrophes: The Science of Mass Extinction. Cambridge University Press, Cambridge. COWIE, J. W., ZlEGLER, W. & REMANE, J. 1989. Stratigraphic Commission accelerates progress 1984 to 1989. Episodes, 12, 79-83. Cox, B. M., 1990. A review of Jurassic chronostratig- raphy and age indicators for the UK. In: HARDMAN, R. F. P. & BROOKS, J. (eds) Tectonic Events Responsible for Britain's Oil and Gas Reserves. Special Publications 55, The Geological Society, London, 169-190. DANIEL, G. E. 1950. A Hundred Years of Archaeology, Duckworth, London. DETRE, C. H. & TOOTH, I. (eds). 1998. Impact and Extraterrestrial Spherules: New Tools for Global Correlation, Papers Presented to the 1998 Annual Meeting of IGCP 384, Hungarian Geological Institute, Budapest. DEWEY, J. F. 1999. Reply when awarded the Wollaston Medal. Geological Society Awards 1999. London. DIETZ, R. 1994. Earth, sea and sky: life and times of a journeyman geologist. Annual Reviews of Earth and Planetary Science, 22, 1-32. DIETZE, V. & CHANDLER, R. B. 1997. S. S. Buckman und der Inferior Oolite, Fossilien, 4, 207-213. DINGIS, L. 1984. Effects of Stratigraphic completeness on interpretations of extinction rates across the Cretaceous-Tertiary boundary. Paleobiology, 10, 420-438. DONOVAN, D. T. 1966. Stratigraphy: An Introduction to Principles. Thomas Murby, London. DONZE, P., BEN ABDELKADER, O., BEN SALEM, H., MAAMOURI, A L., MEON, H. et al. 1996. At K-T boundary, the stratotypical section (El Kef, NW Tunisia) shows a concomitance of three different events. Abstracts of the 30th International Geo- logical Congress, Beijing, 2, 111. DOTT, R. H. Jr 1983. Episodic sedimentation. How normal is average? How rare is rare? Does it matter? Journal of Sedimentary Petrology, 53, 5-23. DOTT, R. H. Jr 1992 An introduction to the ups and downs of eustasy. Geological Society of America Memoir, 180, 1-16. DOTT, R. H. Jr 1996. Episodic event deposits versus Stratigraphic sequences - shall the twain never meet? Sedimentary Geology, 104, 243-247. DOTT, R. H. Jr 1998. What is unique about geological reasoning?, GSA Today, October, 15-18. DOYLE, P. & MACDONALD, D. I. M. 1993. Belemnite battlefields. Lethaia, 26, 65-80. DUNAY, R. E. & HAILWOOD, E. A. (eds). 1995. Non- biostratigraphical Methods of Dating and Corre- lation. Special Publications, 89, The Geological Society, London. EKDALE, A. A. & BROMLEY, R. G. 1984. Sedimen- tology and ichnology of the Cretaceous-Tertiary boundary in Denmark: implications for the causes of the terminal Cretaceous extinction. Journal of Sedimentary Petrology, 54, 681-703. EINSELE, G., RICKEN, W. & SEILACHER, A. (eds) 1991. Cycles and Events in Stratigraphy. Springer, Berlin. FIFIELD, R. 1987. Chinese find giant crater. New Scien- tist, 113, No. 1543, 19. FISCHER, A. G. 1991. Orbital cyclicity in Mesozoic strata. In: EINSELE, G, RICKEN, W. & SEILACHER, A. (eds), Cycles and Events in Stratigraphy. Springer, Berlin. 48-62. FLETCHER, C. J. N., DAVIES, J. R., WILSON, D. & SMITH, M. 1988. Tidal erosion, solution cavities and exhalative mineralization associated with the Jurassic unconformity at Ogmore, South Glamor- gan. Proceedings of the Geologists' Association, 99, 1-14. FLETCHER, C. J. N. et al. 1986. The depositional environment of the basal 'Littoral Lias' in the Vale of Glamorgan-a discussion of the reinter- pretation by Ager (1986), Proceedings of the Geologists' Association, 97, 383-384. FRANKEL, C. 1999. The End of the Dinosaurs: Chicxu- lub Crater and Mass Extinctions. Cambridge Uni- versity Press, Cambridge. GLASS, B. P. 2000. Upper Eocene impact/spherule layers: a status report. Paper presented to the I. G. C. at Rio de Janeiro, Session 25-6. GLEN, W. 1982. The Road to Jaramillo. Stanford Uni- versity Press, Stanford. GLEN, W. 1994. The Mass-Extinction Debates: How Science Works in a Crisis. Stanford University Press, Stanford. GOLDMAN, D., MITCHELL, C E., BERGSTROEM, S. M., DELANO, J. W. & TICE, S. 1994. K-bentonites and Graptolite Biostratigraphy in the Middle Ordovi- cian of New York State and Quebec. Palaios, 9, 124-143. GOSTIN, V. A., HAINES, P. W., JENKINS, R. J. F, COMP- STON, W. & WILLIAMS, I. S. 1986. Impact ejecta horizon within late Precambrian shales, Science, 233,198-200. GOULD, S. J. 1989. Wonderful Life. Penguin Books, London. GOULD, S. J. 1992. Dinosaurs in the haystack, Natural History, March, 2-13. GRETENER, P. E. 1967. Significance of the rare event in geology. Bulletin of the American Association of Petroleum Geologists, 51, 2197-2206. HALLAM, A. 1984. Asteroids and extinction - no cause for concern. New Scientist, 104, No. 1429, 30-32. HALLAM, A. 1986. Origin of minor limestone-shale cycles: climatically induced or diagenetic? Geology, 14, 609-612. HANCOCK, J. M. 1993. Transatlantic correlations in the Campanian-Maastrichtian stages by eustatic changes of sea-level. In: HAILWOOD, E. A. & KIDD, R. B. (eds) High Resolution Stratigraphy, Special Publications 70, The Geological Society, London, 241-256. HAY, W. W. (ed.) 1974. Studies in Paleo-Oceanogra- phy. Society of Economic Paleontologists and Mineralogists, Special Publications 20, Tulsa. HEDBERG, H. D (ed.) 1976, International Stratigraphic Guide. John Wiley, New York. HEDLEY, D. 1987. Barcodes - selling by numbers. Esso Magazine, 144, 18-21. [...]... recorded them from the Jurassic Solenhofen Stone of Bavaria, Germany, demonstrating their affinity and proposing the generic names Eunicites and Lumbriconereites ( 186 8a,b) Extensive studies by George J Hinde of material from England, Wales, Canada and Sweden ( 187 9, 188 0, 188 2, 189 6) established a basis for the nomenclature of what he regarded as being isolated components of annelid jaws; but study of them... stratotype Journal of the Geological Society, London, 154, 709-7 18 VAIL, P R 1992 The evolution of seismic stratigraphy and the global sea-level curve Geological Society of America Memoir, 180 , 83 -91 VALLANCE, T G 19 68 The beginning of geological system Scan, November 19 68, 28- 34 VAN ANDEL, T H 1 981 Consider the incompleteness of the geological record Nature, 294, 397-3 98 VAN LOON, A J 1999 The meaning of... Erdtman ( 189 7-1973) on left, with William S Hoffmeister (1901-1 980 ); uncredited, reproduced from Traverse (1 988 ) 2 78 WILLIAM A S SARJEANT Fig 8 John Rowley and Eszther Nagy at the International Palynological Congress in Brisbane, Queensland (photograph by the author 1 September 1 988 ) and archaeologists, wherever environmental conditions permit They have contributed immensely to our understanding of... pp 18- 23) With the construction of the first microscopes by Robert Hooke, Antoni van Leeuwenhoek and others, the study of pollen and spores was greatly facilitated; however, the history of the development of microscopes is told by Bradbury (1967) and does not require repetition here A major contributor to the understanding of flower and pollen morphology, and of the processes of pollen dispersal and. .. Jansonius & McGregor, 1996, p 1) made the first detailed description of the structure of flowers, noting that the anthers served as 'the Theca or Case of a great many From: OLDROYD, D R (ed.) 2002 The Earth Inside and Out: Some Major Contributions to Geology in the Twentieth Century Geological Society, London, Special Publications, 192, 273-327 0305 -87 19/02/$15.00 © The Geological Society of London 2002... matter In 184 8, Goppert likewise observed macrospores but again misinterpreted them, designating them as Carpolithes coniformis; this name was to be used as late as 188 1 by Otto Feistmantel, even though he recognized the bodies to be macrospores The fact that the macrospores commonly occurred within a mass of microspores was first noted by the eminent botanist Joseph D Hooker ( 184 8), who observed them in... All these fields form a part of a larger sub- The first observation of fossil Quaternary pollen was by the great German microscopist Christian Gottfried Ehrenberg (1795- 187 6, Fig 5) who reported Pinus pollen in sediments from northern Sweden ( 183 7a) A Swiss naturalist, J Fruh ( 188 5), succeeded in enumerating most of the common tree pollen The Swedish geologist Filip Trybom ( 188 8), having noted the. .. (1949, 1964); and on Permian to Triassic assemblages Fig 21 D Colin McGregor on field work in southern Ireland (photograph by the author, 14 September 1 982 ) Fig 22 The Danish palynologist Kaj Raunsgaard Petersen on left, and Australian palynologist Basil E Balme at the International Polynological Congress in Brisbane, Queensland (photograph by the author, 1 September 1 988 ) PALYNOLOGY 289 Fig 23 Elena... those lands Notable work was done on Carboniferous megaspores by M Brzozowska (with Zoldani, 19 58) and J Karczewska (1967) in Poland and by M Kalibova-Kaiserova (1951 and later papers) and Blanca Pacltova (1966) in Czechoslovakia Eszther Nagy (1968a, b, and later papers; Fig 8) worked on Hungarian Tertiary microfloras M Rogalska (1954, 1956) studied the spores and pollen of Polish Jurassic lignites and. .. ( 184 5) and later proposed the name Spiniferites for them ( 185 0) Unfortunately, this was done so obscurely that the name he had proposed did not come to the attention of other scientists for more than a century (see Sarjeant 1967a,19700,19920) In 184 3, Ehrenberg reported the first Jurassic 'xanthidia' from Poland Both dinoflagellates and 'xanthidia' were illustrated in his massive Mikrogeologie ( 185 4), . struck and whether the physical fallout from the impact can be docu- mented in the surrounding rocks, as in the case of the Manson microcline. Such physical evi- dence has been the . Research, 66, 6 08- 612. BUCKMAN, S. S. 188 9. On the Cotteswold, Midford and Yeovil Sands. Quarterly Journal of the Geological Society, London, 45. 440-474. BUCKMAN, S. S. 189 3. The Bajocian . dating of the geological record. EOS, 78, 285 - 289 . HOLLAND, C. H. 1 986 . Does the golden spike still glitter? Journal of the Geological Society, London. 143, 3-21. HOLLAND, C. H.,