National Museum of Natural Sciences
National Museums of Canada

Musée national des Sciences naturelles
Musées nationaux du Canada

Ottawa

K-TEC I

Cretaceous-Tertiary Extinctions and Possible
Terrestrial and Extraterrestrial Causes

No. 39 1982

ES

SYLLOGEUS is a publication of the National Museum of Natural Sciences, National Museums of
Canada, designed to permit the rapid dissemination of information pertaining to those
disciplines and educational functions for which the National Museum of Natural Sciences is
responsible. In the interests of making information available quickly, normal publishing
procedures have been abbreviated.

Articles are published in English, in French, or in both languages, and the issues appear at
irregular intervals. A complete list of the titles issued since the beginning of the series
(1972) and individual copies of this number are available by mail from the National Museum of

Natural Sciences, Ottawa, Canada. KIA OMS

La collection SYLLOGEUS, publiée par le Musée national des sciences naturelles, Musée
nationaux du Canada, a pour but de diffuser rapidement le résultat des travaux dans les
domaines scientifique et éducatif qui sont sous la direction du Musée national des sciences
naturelles. Pour assurer la prompte distribution de cette publication, on a abregé les
étapes de la rédaction.

Les articles sont publiés en français, en anglais ou dans les deux langues, et ils
paraissent irréguliérement. On peut obtenir par commande postale la liste des titres

de tous les articles publiés depuis le début de la collection (1972) et des copies

individuelles de ce numéro, au Musée national des sciences naturelles, Ottawa, Canada.

K1A OM8

Syllogeus Series No. 39 Série Syllogeus No. 39

(c) National Museums of Canada 1982 (c) Musée nationaux du Canada 1982
Printed in Canada Imprimé au Canada

ISSN 0704-576X

TEA
CRETACEOUS-TERTIARY EXTINCTIONS AND POSSIBLE

TERRESTRIAL AND EXTRATERRESTRIAL CAUSES

by the K-TEC* group

Proceedings of the workshop held in Ottawa, Canada

19-20 May 1981

Edited by D.A. Russell and G. Rice

Sponsored by the Paleobiology Division, National Museum of
Natural Sciences (Canada) and by the Herzberg Institute of
Astrophysics, National Research Council of Canada.

Alvarez, W. Alvarez, F. Asaro, P. Béland, K. Brasch,

-V.M. Clube, D. Dilcher, P. Feldman, R.A.F. Grieve,

Halliday, K.J. Hsu, D.M. Jarzen, J.F. Jaworski, F.T. Kyte,
-E. Litherland, D.M. McLean, H. Michel, W.M. Napier,

lo PLROBYMERG, Yoo welalickS, WA. muSGseLl, we Sthe-

-R. Thierstein, W. Tucker (italicized names indicate
participants in the first K-TEC workshop, 16-17 November 1976).

TA > HU EH

Syllogeus No. 39

National Museums of Canada ; Les Musées nationaux du Canada

National Museum of Natural Sciences Musée national des Sciences naturelles

Participants in the K-TEC II workshop:

Standing, from left to right; V. Clube, I. Halliday, P. Feldman, W. Napier,

D. McLean, W. Tucker, K. Pirozynski, A. Litherland, G. Rice, J. Jaworski,

D. Jarzen, J. Rucklidge, L. Alvarez, D. Russell, W. Alvarez, J. Smit,

D. Dilcher, Stenonychosaurus inequalis. Seated, from left to right: F. Asaro,
K. Brasch, K. Hsu, F. Kyte, H. Michel, R. Grieve, P. Béland, H. Thierstein.

Contents

Preface
D A “RUSSELL

Proceedings of K-TEC II Workshop
References

Contributed Abstracts

Episodic Catastrophism
S.V.M. Clube

Looking Back on the Tunguska Event
Ian Halliday

Evolutionary and Environmental Consequences
of a Terminal Cretaceous Event
Kenneth J. Hsu

Deccan Volcanism and the Cretaceous-Tertiary
Transition Scenario: a Unifying Causal Mechanism
Dewey M. McLean

Effects of Solar System Passage through Giant
Molecular Clouds
W.M. Napter

Accelerator Mass Spectrometric (AMS) Measurement

of Pt and Ir at the Cretaceous-Tertiary Boundary
J.C. Ruckltdge

Participants

128

136

140

143

147

150

PREFACE

It is a pleasure to present herewith the proceedings of the second workshop on
Cretaceous-Tertiary extinctions and possible terrestrial and extraterrestrial causes
sponsored by the Paleobiology Division, National Museum of Natural Sciences (Canada) and by
the Herzberg Institute of Astrophysics, National Research Council of Canada. The first
workshop, held four and one-half years previously, received the acronym "K-TEC" after
Cretaceous-Tertiary Environmental Change (K being the geological symbol for Cretaceous) from
the pen of Pierre Béland. There were only eleven of us then, and all but two were supported
by Canadian institutions. The original workshop was both enjoyable and stimulating, — and
we were grateful to find that some of the concepts we considered figured in the subsequent
debate on the extinction of the dinosaurs. One of the outstanding needs perceived at the
time was for chemical studies of the boundary horizon. This need is now being filled so
vigourously that it appears that paleontology must yield pride of place to trace element
studies in resolving the nature of the extinctions.

Included within the volume resulting from the original workshop were 16 pages
summarizing discussions between participants. Reactions to this small section were so
favourable that its impact probably exceeded that of the remainder of the volume. For this
reason the proceedings of the second workshop consist almost entirely of a condensation of
the two-day dialogue between workers in different disciplines approaching a common problem
from different theoretical points of view. It represents an alternative to the general
practice of publishing formal presentations with minimal or no reactions from other
participants in a colloquium. A few contributed abstracts follow the discussions. The
prestige of our host institutions is not responsible for the academic talent which made
itself available for these symposia. Scholars have obviously attracted other, comparably
distinguished scholars. However, all of us were keenly aware of the absence of colleagues
who have made important contributions to understanding the problem of mass extinctions.

We acknowledge our gratitude to them, and wish them continued success in their researches.
I must also personally underscore my gratitude to Dr. Paul Feldman of the H.I.A. for his

wise counsel and generous consideration during the months when our workshop was taking form.

D.A. Russell

PREFACE

Je suis heureux de présenter le compte rendu du deuxième atelier consacré aux Extinetions
du crétacé-tertiaire et aux phénomènes divers, terrestres et extraterrestres, susceptibles de
les avoir provoquées. L'atelier était parrainé par la Division de paléobiologie du Musée
national des sciences naturelles (Canada) et par l'Institut Herzberg d'astrophysique du
Conseil national de recherches du Canada. Le premier atelier, tenu il y a quatre ans et demi,
avait été surnommé "K-TEC" pour "Cretaceous-Tertiary Environmental Change" (K est le symbole
du Crétacé en géologie) par M. Pierre Béland. Nous n'étions qu'onze a y participer, tous
envoyés par des établissements canadiens à deux exceptions près. Nous avions trouvé
l'expérience aussi agréable que stimulante — et nous avons constaté avec plaisir que certains
des concepts étudiés avaient été repris à l'occasion du débat subséquent sur la disparition
des dinosaures. Nous avions reconnus, entre autres besoins impérieux, la nécessité d'étudier
la composition chimique de l'horizon limite des extinctions. Ce besoin a été comblé avec
tant d'ardeur qu'il semble maintenant que la paléobiologie doive s'incliner devant les études
d'éléments-traces pour ce qui est d'expliquer la nature des disparitions.

Les travaux du premier atelier furent consignés dans un ouvrage qui comprenait seize
pages de résumé des discussions. Si l'on en juge par l'accueil qui lui fut réservé, cette
brève section a sans doute en plus d'impact que le reste de l'ouvrage. Forts de cette
réussite, nous publions surtout, dans le compte rendu du deuxième atelier, des résumés des
entretiens qu'ont eu pendant deux jours les spécialistes de diverses disciplines, qui
abordaient un problème commun sous divers angles théoriques. Nous avons préféré cette
solution à l'usage courant qui consiste à publier les communications sans mentionner les
réactions des autres participants. Le compte rendu des débats est suivi de quelques résumés
soumis par les auteurs. La compétence des chercheurs participants ne doit rien au prestige
des établissements grâce auxquels la rencontre a pu avoir lieu. De toute évidence, les
chercheurs distingués se sont attirés les uns les autres. Nous étions toutefois conscients de
l'absence de collègues qui ont grandement contribué à éclaircir le problème des extinctions
massives. Nous tenons à leur exprimer notre respect et à leur souhaiter de fructueux travaux.
Pour ma part, je désire remercier en particulier M. Paul Feldman, de l'Institut Herzberg, qui
m'a prodigué de précieux conseils et accordé généreusement son attention tout au long des mois

de préparation de l'atelier.

D.A. Russell

PROCEEDINGS OF K-TEC II WORKSHOP

Asaro: The basic framework upon which the physical science studies of the Cretaceous-
Tertiary (K-T) boundary are based has been provided by paleontological and geological
evidence which defines the location of the extinctions in the stratigraphic record. The
iridium anomaly was initially found associated with the Cretaceous-Tertiary boundary in
Italy, and was predicted to arise from sources which were extraterrestrial (Alvarez
et al. 1979A) but within the Solar System (Alvarez et al. 1979B). The anomaly was also
predicted to be world-wide in occurrence and hypothesized to be the result of a 10
kilometer type 1 carbonaceous chondrite asteroid impacting the Earth. The impact caused
an explosion equivalent to ~ 108 megatons of TNT and the resulting dust cloud enveloped
the Earth, spread the iridium, shut off the sunlight, stopped photosynthesis and was
responsible for the K-T extinctions (Alvarez et al. 1980).

The iridium anomaly associated with the K-T boundary has, up to the present time,
been found in 17 out of 18 localities in which it was carefully sought and the K-T
boundary was thought to be intact. This includes a site in the Pacific which Dr. Kyte
mentioned to me this morning. This work has been done by six different laboratories
(Smit and Hertogen (1980), Kyte et al. (1980), Ganapathy et al. (1981), Hsü et al. (1981,
Orth et al. (1981), Alvarez et al. (1980, 1982), and Michel et al. (1981)). The work by
the Berkeley group is presently directed to determine if the iridium is truly distributed
world-wide, if the iridium anomaly is of an extraterrestrial origin, if this extra-
terrestrial iridium has an extraordinary origin (i.e. it is not due to the normal
ablation of meteorites), and if iridium can be found associated with other extinctions in
the past.

With respect to the extraterrestrial nature of the iridium anomaly, Ganapathy (1980)
measured nine elements, including several noble elements, and found a pattern which he
interpreted as that of type 1 carbonaceous chondrites. Later Kyte, Zhou and Wasson (1980)
and we in Berkeley (Asaro et al. 1980) also made measurements on the same section (Stevns
Klint, Denmark). All three investigations interpreted the data as evidence of the impact
of extraterrestrial material but drew slightly different conclusions on the exact form of
the object. A plot (Fig. 1) may demonstrate the significance of abundance ratios of
noble elements, especially Au/Ir and Pt/Ir for source attributions. Although platinum

and iridium are both noble elements, their ratio can be altered as a result of thermal

1.20

1.10

1.00

0.90

Noble element ratio / ratio in C1 chondrites

0.80

FIGURE 1:

+ Danish C-T

C1 Chondrites

Au/Ir Pt/Ir Ni/ Ir

XBL 8111-1590

Plot showing the significance of abundance ratios of
noble elements (Au/Ir, Pt/Ir, Nt/Ir) at the Cretaceous-
Tertiary boundary, Stevns Klint, Denmark, compared to
their ratios tn C1 chondrites.

mantle melts, platinum will go into the lower melting fractions, and iridium remains
behind in the higher melting fraction. The higher the partial melting of the mantle

the lower is the enrichment of the platinum. For example, in nickel sulphide flows
which occurred in considerable quantity during Archaean time, the Pt/Ir and Au/Ir ratios
are perhaps a factor of two or more higher than in chondrites, and the Ni/Ir ratios are
even higher. In Fig. 1 these ratios for the Stevns Klint K-T layer have been divided by
the average values for Cl chondrites. An object whose relative values of Ir, Pt, Au and
Ni were the same as the averages for Cl chondrites would have a value of one for each
ratio shown in Fig. 1. The boxes indicate the estimated standard deviations in the mean
for the Pt/Ir, Au/Ir and Ni/Ir ratios in chondrites. The Au/Ir ratio, with a standard
deviation of ~ 23%, is five percent smaller than the average chondrite value, but is
easily encompassed within the uncertainties of the latter. The Pt/Ir ratio, with a
standard deviation of about 34%, is also easily contained within the uncertainty in the
chondritic value. The ratio of Ni/Ir is 15% lower than an average for Cl chondrites.
Nickel, however, is not a noble element and would be subject to many more chemical
effects. If the Stevns Klint K-T Pt/Ir and Au/Ir ratios are compared with data for C2 or
H chondrites, the average disagreements are 11% (C2) and 5% (H), somewhat greater than
the Cl disagreement of 3%, comparisons with L, LL, E and C3 chondrite values exhibit more
significant differences. Thus the noble element ratios, as precisely as we can measure
(between 24 and 334%), agree very well with the expectations for Cl chondrites and some-
what less well for other chondrites. Terrestrial and iron meteoritic materials usually
have considerably different values.

Next we wished to determine if this extraterrestrial material was of an extra-
ordinary origin. Dr. Miriam Kastner from the Scripps Institution of Oceanography studied
the clay mineralogy of the Danish boundary layer which averages about one centimeter in
thickness (Alvarez et al. 1982). The clay component was found to be a nearly pure
smectite with very little indication of other types of clays, such as the mixed layer
clays. According to Dr. Kastner, such a clay would only occur as an alteration product
of an eruptive ash or glass. She could not distinguish between ash from an asteroidal
impact or a volcano. In contrast, the clay component in limestone above and below the
boundary are primarily smectite/illite mixed layer clays, which are detrital and derive

from the continents. The Danish boundary clay was therefore deposited in situ and is

extraordinary in the sense that it is not composed of detrital clays. As the
boundary clay is not detrital, its high iridium content is not due to a lag deposit
resulting from a cessation of calcium carbonate deposition.

With respect to the repeatability, our team sent a student to China to collect
samples from the Permo-Triassic boundary, with the very generous co-operation of
the Government of the Peoples' Republic of China. The samples came from Nanking and
supplement earlier ones from this boundary region which we received from Dr. John
Utting of Petro-Canada and in which we found no iridium. A boundary clay occurs
between Permian and Triassic limestones, which is composed of montmorillonite
according to information we received previously (J. Utting, personal communication
1980). Chemical studies of this boundary clay indicate that it differs greatly from
the material above and below. It is strongly depleted in strontium, and enriched in
thorium, yttrium and zirconium (Giauque 1981), but not in iridium. We initially
measured whole rock abundances with a sensitivity of about 0.2 or 0.3 PPB and
detected no iridium. We are now trying to improve the sensitivity of the iridium
measurement. So, this is where we stand at the moment; we are looking for the iridium

anomaly in a number of other extinctions and so far we haven't found any.

L. Alvarez: Would you compare your results with those of Ganapathy (1980) and Kyte, Zhou

and Wasson (1980) to show the spread in values?

Asaro: The noble element abundance measurements are difficult to do with high precision,
particularly that of platinum. Ehmann and Gillum (1972) stated that measurements
for gold and platinum could be carried out at a rate of 15 per working day. We
didn't do 15 per day. To do one measurement of the platinum required one week. Much
of the time was spent on determining how well the experiment was carried out and how
accurate the value was. Ganapathy (1980) and Kyte and his co-workers (1980) have
found values for gold, platinum and iridium abundances in the Stevns Klint K-T boundary

layer that are off scale, below the graph.

Kyte: I am surprised that you have plotted such large ranges for the Cl abundances. The

Cls do not vary at all except for sampling and analytical error (Kallemeyn and Wasson

1981).

Asaro: The standard deviation in the C1 Au/Ir ratios was 19%, with a standard deviation of
the mean of 6.6%. The individual errors in the platinum measurements appeared to be
about 10%. The standard deviation of the mean was taken as 6% for the Pt/Ir ratio.

The abundances in the meteorites are an order of magnitude higher than in the

Cretaceous-Tertiary boundary layers under study and the techniques that work with

meteorites are not necessarily the best for K-T studies.

Kyte: Your plot illustrates the fact that trace element analyses always present problems
in terms of interlaboratory comparison. There is no range of nickel concentration in
the Cl's. Also, in some Danish layers a significant fractionation exists particularly
with regard to nickel and chromium, relative to C1 abundances. Nickel is sometimes

depleted by as much as a factor of 3, not 20% as plotted in Fig. 1. The question

_———

remains open as to the compositon of any proposed extraterrestrial source material.

Asaro: I was very concerned about the discrepancy in the nickel abundance of Stevns Klint
boundary clay as measured in different laboratories and re-checked our secondary
standard (standard pottery) against standards from the United States National Bureau
of Standards. The value we obtained agreed closely with the previous calibration.

In addition, an x-ray flourescence calibration of our secondary standard (Giauque
et al. 1973) agreed within 1% with the first calibration, so our secondary standard
is probably accurately known. In each irradiation we use a secondary standard that
contains the element that is being measured so that nickel was compared to a

standard with nickel of known abundance.

Kyte: Using data from Deep Sea Drilling Project (D.S.D.P.) site 465A, the three highest
iridium samples indicate a nickel to iridium ratio of about 7,000 whereas the Cl
value is on the order of 23,000, or different by a factor of over three. Too much
could be made of this discrepancy, but I do think that not enough data are available
to assign the composition of the extraterrestrial source to a single type of meteoritic |
object. There is good evidence of chondritic material, although it may be difficult

to rule out an iron projectile. More analyses are needed.

Asaro: Nickel is not a noble element and the deviation is not unexpected in site 465A.

10

In the pyrite-bearing fraction the Ni/Ir ratio is higher, and in the calcareous phase,
which contains most of the iridium, it's lower than in chondrites (Michel et al. 1981).
We have measured the platinum to gold to iridium ratios in the calcareous phase and
find they are roughly consistent with chondritic values. The experiment is difficult
and the errors are large. However, at Stevns Klint, Caravaca and site 465A
measurements of platinum, gold and iridium are all consistent with type 1 carbonaceous

chondrite ratios. Nickel shows great variations from one deposit to another.

Kyte: Platinum is certainly more difficult to analyze radiochemically than gold and
iridium. However, in one of our 465A samples (Kyte, unpublished data) a gold to
iridium ratio was measured which was 23 times the Cl value. Fractionation has
occurred and your finding a precisely Cl ratio may well be fortuitous; a sample just

happened to fall within a very large possible range of sample concentrations.

L. Alvarez: Our original conclusion that an asteroid or a meteorite hit the Earth was
based on evidence from different sources. When Ganapathy (1980) first measured what
we call a fingerprint of relative elemental abundances, he drew attention to meteoritic
material in the Cretaceous-Tertiary boundary in the title of his paper. Although his
points are off the graph (Fig. 1), Ganapathy concluded that the material was

chondritic.

Kyte: The source was probably chondritic.

L. Alvarez: Everyone, as one of my friends used to say in a discussion like this, seems
to be "... in violent agreement". The analytical errors are everywhere, but different
kinds of evidence suggests that a chondritic meteorite hit the Earth 65 million years

ago.

Kyte: It is important in constraining any model to correctly determine the source
composition. One of the main factors in an impact is the amount of energy involved,
which is directly related to the mass of accreted material. Iridium concentrations in
chondrites vary by a factor of three to four (Mason 1971), and if carbonaceous
chondrites are cometary material, ice may also be present. A very large range of

masses becomes possible because of this uncertainty.

ait

L. Alvarez: The velocity of impact can also vary be a factor of seven, increasing the
energy by a factor of 50, adding cometary ice produces a factor of 100 in the amount

of material that could be thrown up.

Kyte: A possible iron meteroid source increases the range of possible concentrations to

perhaps a factor of 30.

Hsu: Comets have been described as dirty snowballs. What is the dirt or the solid phase

in the core of the comet?

Halliday: The spectrum produced by the dirt from cometary fragments as they come into the
atmosphere is amazingly similar to the shock tube spectrum of a chondritic meteorite.
The abundances in Solar System materials from which both comets and meteorites are
derived are fairly similar in spite of fractionating processes. Qualitatively,

cometary and meteor spectra show no suggestion that they are dramatically different.

Hsu: Are cometary cores similar to carbonaceous chondrites?

Halliday: Carbonaceous chondrites are the most primitive of the meteorites, or objects
which reach the ground. Cometary meteors are fragile dynamically, and break up at
very low pressures in the upper atmosphere. We would think of them as an extension
of the meteorite series beyond the carbonaceous chondrites into perhaps more primitive

materials.

Kyte: Extraterrestrial dust collected in the stratosphere provides the best evidence for
comet compositions (Brownlee et al. 1977). The particles range in size from 5 to 50
microns and largely consist of fine-grained aggregates of chondritic composition. In

some respects they appear to be more primitive than carbonaceous chondrites.

Hsu: What does more primitive mean?

Kyte: Carbonaceous chondritic material that has been least altered (Fraundorf 1981),
by, for example, the action of liquid water, is most primitive. Carbonaceous
chondrites have sulphate veins and clay minerals which indicate aqueous alteration.
The very fine grain size of these dust particles is such that if they were ever in the

presence of liquid water their high surface free energy would result in rapid alteration.

12

These are very primitive materials, having a high carbon content and perhaps originally
a high ice content. The main point is that a large fraction of the dust in the
stratosphere is apparently carbonaceous chondritic, or something very similar, in

compositon.

Rucklidge: It was mentioned that the accelerator measurements were off figure 1 in the
upper direction. Our method of making these measurements is in the early stages yet
and the problem is one of calibration. This method was developed by a group from the
universities of Toronto and Rochester, and General Ionex Corporation, for direct
measurement of carbon 14 using a tandem accelerator as a mass spectrometer, and can
be used for the ultrasensitive detection of other elements. Platinum and iridium
have the fortunate property of forming negative ions very readily, a necessary condition
for these elements to be analyzed by this method. Our results for iridum from the
boundary clay at Stevns Klint, Denmark, confirm the measurements published by
Alvarez et al. (1980). Platinum can bemore easily measured than iridium, contrary
to the case with neutron activation, and we have been able to make platinum to iridium
comparisons. The problem of covering a larger mass range, to embrace the lighter
platinum group elements, such as rhodium and palladium, is more difficult. The sample
is in a pressed powder form, and a beam of cesium ions is directed at it. Ions
sputter off the surface and as time proceeds the cesium beam erodes a hole. If the
material is not homogeneous then the signal varies. We discovered that the platinum
signal varied with time quite dramatically and this could be correlated with the
presence of pyrite in the sample. Assuming that the high values corresponded with the
presence of pyrite, we prepared a heavy mineral separate using heavy liquids to extract
the pyrite from the samples. The heavy and light fractions were compared, and it was
found that platinum and iridium were enriched in heavy fractions relative to light
fractions. There appears to be heterogeneous distribution in these materials in the
Danish boundary clay. A sample was taken of a pyritiferous Ordovician shale from
Oshawa, Ontario. The pyrite from this shale yielded about three parts per billion of
iridium. In a second series of experiments an improved ion source was used, in which
the cesium, which had formerly flooded the sample, was changed to focus onto a spot
of about 200 microns in diameter and became, in effect, a microprobe. In the second

series of experiments it transpired that a heavy metal bearing phase might or might not

13

be hit, and extremely low or considerably higher values could be encountered. Three
samples were taken from D.S.D.P. site 465, two of which were pyritiferous and one of
which was a white ooze just below the pyritiferous layer. There was a factor of
approximately 20 between the pyritiferous and ooze samples, which is approximately
what one would expect. The pyritiferous samples, however, yielded concentrations of
only 0.04 PPB of platinum and 0.06 PPB of iridium in one,and the elements were not
detected in the other. We estimate the detection limit for iridium and platinum to
be approximately 1 PPT, on the basis of a 15 minute measurement. The ratios vary
considerably I must admit, and lie quite a bit above points shown on Fig. 1. One other
experiment was made by a student of Geoffrey Norris, Silvana de Gasparis. She was
interested in the possibility that the noble metals might in fact be associated with
the pollens and other organic materials, and prepared an organic extract from some of
the Danish clays. The platinum and iridium contents of this extract were well over a
part per million. The extraction had been performed with nitric acid, and the noble
metals may have been concentrated in the organic residue. Samples were subsequently
prepared in which extreme care was taken to eliminate by physical means all the
metallic phases, and only minute amounts of organic material were extracted. These
also showed significantly high platinum and iridium content. Now this is interesting
in the light of an abstract by a group at Los Alamos (Orth et al. 1981) where they
have reported an iridium anomaly in New Mexico in association with coal seams at the
Cretaceous-Tertiary boundary. Geologists are familiar with the fact that coal seams
provide a reducing environment in which heavy metals are often concentrated. Uranium
in fact has been mined from coal seams,and a possible geochemical affinity exists

which should not be overlooked in examining this problem.

Asaro: They have located the boundary using palynofloral evidence to within one meter,
and the iridium anomaly was in that meter. The maximum concentration of iridium is
two to four parts per billion and the enhancement over the background is about a
factor of 100. The iridium was concentrated chemically and sensitivities are comparable

to your work.

Kyte: Our recent work indicates that a reducing environment may not be required to

produce an iridium horizon. We found it in an abyssal clay in the mid-Pacific, which

14

is typically a rather oxidized environment. Preliminary data suggest an iridium

2

content of about 80 nannograms per cm“ or more in that horizon (Corliss and Hollister

1979)

Hsu: Dr. Kastner noted that unusual clay minerals are present in the boundary layers in
Denmark and at D.S.D.P. site 465. In our boundary layer in the Southern Hemisphere
core, the iridium concentration is not as high as in the Northern Hemisphere, reaching
a maximum of only 3.3 PPB (Hsu et al. ms.). The anomaly may be obscured by a detrital
dilution effect, because most of the clay minerals are terrestrial detritus. Smectite
is present but it is not unusually abundant. I would like to pose the problem on how
this iridium anomaly should be expressed. You show it in nannograms per square

centimeter on your map (Fig. 2).

Asaro: ves.

Hsu: This would solve the problem of vertical disturbance like bioturbation. However
problems would remain in the case of lateral transport through slumping, turbidity
currents or bottom currents. The boundary layer is thicker in some areas because of
resedimentation. It might be worthwhile to prepare a map showing the maximum

concentration as well as the total excess iridium.

W. Alvarez: Our map (Fig. 2) shows excess iridium in nannograms per square centimeter
in all of the localities for which we have information. It contains work by
ourselves, and data published by Smit and Hertogen (1980), Kyte (et al. 1980), and
Ganapathy (1980). Hsu and Kräyenbühl's data (personal communication) are also
included, as well as new information from New Mexico (Orth et al. 1981) and from the
Brazos River section in Texas (Ganapathy and Gardner, in press). Another new point
is based on samples from Haiti given to us by Florentine Maurrasse at Florida

International University. There are two northern Spanish samples, and Smit's section.

L. Alvarez: The largest value of 520 is from the thickest layer, and may be taken as
evidence of lateral transport of material. The concentration is similar to that in

other sites.

Hsu: What is the concentration in the North Pacific core?

15

SPAIN

SAN
CARAVACA SEBASTIAN ITALY (5)

NEW MEXICO

(Continental) SITE 465A

NEW ZEALAND

Excess Ir at C-T
10 °g-cm?
May 1981

SITE 517F

SITE 356 SITE 524 UC Berkeley /LBL

XBL 816-10174

FIGURE 2: Excess Ir in nannograms per square centimeter in Cretaceous-Tertiary boundary
samples from United States, Haïti, New Zealand, Spain, Italy, Denmark and
several oceanic sites.

16

Asaro: In the calcareous part it's about 9 PPB, and in the pyritic part it's considerably

Kyte:

higher.

We measured 16 ng/g, but in addition to pyrite the clay content was also very high,

with the total insoluble residue approaching 35% (Kyte et al. 1980).

Asaro: We found that the material containing the pyrite in D.S.D.P. site 465A isa

Hsu:

magnesium aluminum silicate. Maximum pyrite concentrations are about 25%, but the
rare earth content is extremely low (Michel et al. 1981). A significant cerium
depletion indicates the material was deposited in place. It is not as depleted as are
the normal calcareous materials so another component may be present. An analogous
situation exists in the low abundances of slightly different rare earths in the acid
insoluble fraction in the Danish boundary layer. These contrast with montmorillonite
deposits in general where rare earths are considerably higher. We have tried to
interpret the low rare earth content in terms of an impact of an asteroid in the
ocean, with a large component of mantle material in the resulting ejecta. So far we

are not getting good quantitive agreement with this model.

What would be the distribution pattern of ejecta if it is deposited from the
atmosphere as fallout? Would the pattern resemble that of fallout from nuclear

explosions?

L. Alvarez: The atmosphere would be very different from conditions following an atomic

bomb test. Many cubic kilometers of vapourized water would be injected into the
atmosphere. Turbulence and currents would be like nothing I could possibly guess.
Someone who is good at atmospheric modelling should examine this carefully. It should
not be explored by analogy. Even Mt. St. Helens did not perturb the atmosphere
appreciably. Only a very small amount of material was added to the atmosphere, and the

atmosphere remained very well behaved.

Grieve: We cannot make analogies with volcanic explosions. On penetration, a bolide

would create a hole in the atmosphere, which would disturb atmospheric circulation

patterns.

L. Alvarez: As the atmosphere rushes back into the vacuum one can't imagine the

consequences to the weather.

17

Smit: Among the rare earth elements, was only this cerium depletion found in the Danish

section?

Asaro: Cerium is slightly depleted with respect to the other rare earths in the acid
insoluble residue of the Stevns Klint boundary clay, but it is in no way comparable
to the depletion in the pyrite-containing material of D.S.D.P. site 465A. The rare
earths as a group are in rather low abundance in the Danish acid-insoluble boundary

residue, as they also are in the site 465A pyrite-bearing silicate.

Smit: Using neutron analyses in whole rock, we (cf. Smit and Hertogen 1980, unpublished
data) found a series of significant depletions of five rare earth elements only in
the basal millimeters of the Danish boundary clay section. We found a similar
depletion in the Caravaca section (Spain) and in the Le Kef section (Tunisia), in the
basal millimeters of the boundary clay. At the level of the highest nickel, chromium
and iridium abundances, we also found significant rare earth element depletions. We
devised a model where an asteroid plummets into the ocean, and is mixed with oceanic
basalts to produce the correlation between the highest iridium enhancements and the
rare earth element depletions. In four sections, the highest iridium concentrations
are associated with a significant rare earth element depletion. About 600 PPM of

sulphur have been detected.

Kyte: An oceanic impact should not be modelled with only basalts in the ejecta. There
would also be a significant mantle component which would probably be enriched in
elements such as nickel and chromium. Yet there is barely enough nickel and chromium

present to support a chondritic projectile.

Smit: In the Caravaca section we detected levels of 3,000 PPM nickel, which is an order
of magnitude higher than in the Danish sections. The nickel-iridium correlation is
probably better in that section than in Denmark, and conditions were not anoxic. An
abundant benthic fauna is present, unlike in Denmark. The only correlation we see
is between a rare earth depletion and an iridium enhancement. The latter amounts to

45 PPB at Caravaca, or about as much as in the Pacific cores.

W. Alvarez: In the Gubbio section, the highest iridium levels occur in the upper half

of the boundary clay, which is bright red.

18

Hsu: The iridium anomaly at D.S.D.P. site 524 in the South Atlantic also occurs in a well-
oxidized, hematite-rich clay without pyrite concentrations. By the way, Katharina
Perch-Nielsen, a nannofossil specialist, believes that the Cretaceous-Tertiary sequence
is incomplete at D.S.D.P. site 356, and that the absence of iridium at this site could

be explained by removal of sediment at the boundary, on the Rio Grande Rise.

Kyte: The possibility of using concentrations, as opposed to net iridium per square
centimeter, could be useful in estimating the distribution of the horizon around the
world and calculating the total mass of material accreted. However, these
concentrations should be reported free of indigenous sediment. For example, if 44 PPB
iridium occurs in the Caravaca section, which is is turn 50% carbonate, a value of

about 88 PPB or more should be reported.
L. Alvarez: That's why we used nannograms per square centimeter.

Hsu: It can also easily be expressed, for comparative purposes, in terms of concentration in

the insoluble residue, as Krayenbuhl (Hsü et al., ms.) did.

McLean: The possible volcanic origin of the smectite has been raised. At the time of the
marine extinctions one of the greatest episodes of volcanism in all Earth history
began. The Deccan Traps volcanism in India was initiated 65 million years ago and

lasted up to 60 million years ago. Is it possible that the smectite originated as a

result of Deccan Traps volcanism?

Grieve: The Deccan Traps are basalts, and basalts are not usually associated with large

scale pyroclastic eruptions.

McLean: India during the period of the Deccan Traps volcanism was located near the
equator where oceanic currents could carry volcanic materials into each hemisphere. A

lot of submarine volcanism occurred at the same time.

Hsu: But Deccan volcanism spanned several million years, and the smectite rich layer is

only a few millimeters thick.

McLean: A key issue here is, how much time does that boundary clay really represent? A

very short amount of time? Or is it a lag deposit representing quite a long interval.

19

W. Alvarez: The Deccan Traps cover a very large area in the northwestern part of India
and they are about 1,000 m thick at their maximum on the west coast. One of the
problems in studying Deccan Traps stratigraphy is that outcrops are poor and usually
covered with a lot of vegetation. However, sections along the coast near Bombay cover
only two magnetic polarity zones. The lower half is reversely polarized and the upper
half is normally polarized. Volcanism thus began in a reversed polarity zone, and in
the Italian section the extinction occurs in a reverse polarity zone. The best
available work on correlating polarity zones to the eastern side suggests that below
the reversed zone volcanism occurred in a normal polarity zone, and below that
another reversed polarity zone occurred. These correlations are not well established,
but appear valid. There are thus at least two and probably four polarity zones
involved in the volcanism of the Deccan Traps. This many polarity zones cannot be
missing in the Italian section. The reversal sequence at Gubbio matches the sea floor
spreading magnetic anomalies well, and there are no missing reversal zones. The Deccan
Traps clearly represent much more time than can possibly be missing in the Italian

sections.

McLean: The extinctions may not have spanned five million years. Perhaps they occurred
over a much shorter interval, during the earlier phases of volcanism. Could some of

the smectite be of volcanic origin?

Asaro: Kastner stated that she could not distinguish between a volcanic and an asteroid
impact origin for the material. If it were derived from the mantle, I would expect
it to be enriched in platinum with respect to the iridium. The material from the
Danish boundary layer does not appear to be enriched in platinum with respect to

iridium.

McLean: Morgan (1972A, 1972B) has suggested that Deccan volcanism is a result of mantle
plume activity, with the plumes originating near the base of the mantle. Relatively
large concentrations of the noble elements would be expected to occur in the Earth's
core. And if in fact the Deccan Volcanism does represent mantle plume activity, the
plumes would provide a conduit to bring noble elements from Earth's core to Earth's

surface, where they would be distributed geographically.

20

Kyte: One problem in deriving the noble metals from the core is that, during crystalization,
iridium strongly fractionates into the first solid, so that the inner core probably

contains most of the iridium below the mantle.

McLean: Elsasser and his colleagues (1979) suggest that circulation may even involve the

outer core itself. They visualize whole mantle circulation.

Kyte: The partition coefficient between solid and liquid for iridium is nevertheless so
high that the solid inner core probably contains 90% or more of the whole core's
iridium. Also, because of this fractional crystalization, it seems highly unlikely

that chondritic abundances would occur in the outer liquid core.

Hsu: Many have suggested that the boundary clay layers are lag deposits representing a long
interval of geologic time. The Gubbio paleomagnetic work is excellent, and an
amplified section at D.S.D.P. site 524, essentially confirms its validity. Another
objective of deep sea drilling in our leg was to investigate the chronology of sea-
floor-spreading anomalies and determine if spreading rates were linear. Using
precision stratigraphy, the linear rates were confirmed (Hsu et al., in press). Within
this framework the negative interval of sea-floor Anomaly 29, during which the

extinctions occurred, could not represent more than a half million years.
McLean: A half million years is the figure I use.

Hsu: The extinction horizon represents only one centimeter out of the five or 10 meters
of the clay contained in sediments deposited during Anomaly 29 time. The time
recvired for its deposition is on the order of a few hundred to a few thousand years.

Magnetic stratigraphy restricts the time available to a geological instant.

McLean: In Denmark, Bromley (1979) shows that this boundary clay is sitting on top of an
unconformity. We don't know how much upper Cretaceous is missing before the early
Tertiary boundary clay was deposited. In some cases this boundary clay was apparently

deposited upon a surface that was even lithified before clay deposition began.

Thierstein: We are discussing high resolution stratigraphy in sequences which may or may
not be complete. We can search for sections where we think, according to partly

subjective evolutionary and phylogenetic models, the succession is complete. The

27

22

phlogenetically youngest planktic foraminifera and calcareous nannofossils that have
ever been found in any late Cretaceous section should be present as well as the most
primitive foraminifera and lowest diversity coccolith assemblages that have been found
of earliest Tertiary age. In addition, extinctions and first appearances below and
above the bounday can be used to define zonal subdivisions. In Gubbio, Italy, for
example, only foraminiferal zones can be recognized (Premoli Silva 1977). In most of
the deep-sea drilling sites, it is possible to recognize coccolith zones as well. In
several deep-sea sections, equivalent foram zones near the boundary are two to three
times thicker than the same zones in the Gubbio section. Over 90% of the planktic
foraminifera and the coccolithophorids became extinct at the boundary itself. In
most of the sections not disturbed by preservational changes in carbonate fossils or
by slumps and turbidites, the carbonate content of the earliest Tertiary sediments is
very much reduced. Large sedimentation rate changes across the boundary must,
therefore, be expected. Some microfossil assemblages in different sections do not
resemble each other and micropaleontologists cannot agree whether differential
preservation or evolutionary patterns are responsible for the dissimilarities. For
instance, the G. eugubina Zone, which is the earliest foram zone of the Tertiary,
contains a very primitive Globigerina type foram. At Gubbio the zone is about 40
centimeters thick. At D.S.D.P. site 356, where the iridium anomaly has not been found,
the zone is about 50 centimeters thick. In Tunisia, in the El Kef section, Dr.
Katharina Perch-Nielsen (in press) notes that the zone is nearly three meters thick,
and at site 465A it is more than two meters thick as well. Very high resolution
stratigraphic studies may not elucidate evolutionary patterns very close to the
extinction horizon. It may be more useful to compare biostratigraphic events further
removed from the Cretaceous-Tertiary boundary such as the base of the latest Cretaceous
foram and coccolith zones or the first appearance of Cruciplacolithus tenuts and
Chiasmolithus danicus which occur above the extinction horizon. However, the section
may appear to be complete over a time span of one or maybe one-half million years, but
allowing for differential sedimentation rates within that interval, several tens of
thousands or several hundreds of thousands of years could be accommodated in a clay
layer. Only comparisons between sections can shed light on whether one section is more

complete than another or whether redeposition or volcanic inflow has occurred in

individual sections.

Hsu:

In Denmark, the boundary layer or Fish Clay is a hard ground, which is a little
harder and more slowly sedimented than adjacent layers. However, according to
Perch-Nielsen (1979), all of the nannofossil zones are present there. Correlations
suggest that this hard ground was of very short duration. Late Quaternary hard
grounds in the Mediterranean Sea have been radiometrically dated to have been formed
within a few thousand years (N. Shackleton, personal communication). Thus the
presence of hard ground at the Cretaceous-Tertiary contact cannot account for the
high iridium concentration in Denmark. The time represented by the hard ground in
the Danish section must be within a minor portion of the Anomaly 29 negative-polarity

interval.

Thierstein: Unfortunately, the thickness of the latest Cretaceous foram and coccolith

zones in Denmark are not accurately known, but they must exceed 60 meters (Stenestad
1979). This is much thicker than in any deep-sea section. Sedimentation-rates were
very high during the latest Maastrichtian and were also high during Danian time. The
boundary clay layer could still represent a significant amount of time, although the

mineralogy argues against it.

Rucklidge: A new section has been exposed in Jutland at Nye Klgv and I was able to obtain

a few samples of marl from Dr. Hans Jgrgen Hansen in Copenhagen. Our measurements

indicate they are extremely low in both platinum and iridium. No anomaly is indicated.

Smit: Were your samples taken from the correct marl? I have a difference of opinion

concerning the identification of the boundary clay as published in the Copenhagen
symposium. The true boundary in my opinion was one or two meters below the horizon

indicated.

Kyte: Were the samples from the lower marl?

Rucklidge: I believe so.

Kyte: Hansen sent us a few of those samples too, but they did not resemble the Fish Clay

at Stevns Klint.

Hsu: In Denmark there are hard grounds suggesting bottom current activities, as well as

a variably thick boundary clay. Denmark was evidently an area of strong bottom

23

currents and materials were removed from some areas and concentrated in others.

Thierstein: Paleontological investigations based on forams and coccoliths indicate that

the Jutland sections are more complete than the section at Stevns Klint.

Asaro: Ganapathy et al. (1981) have reported an iridium anomaly in Jutland. We have
separated out the dense mineral fraction in the Danish boundary clay and in the
pyrite-containing material in D.S.D.P. site 465A. The iridium is not concentrated in
the heaviest fraction containing the pyrite. We also did a density-fractionation
study on the Danish boundary material and many of the elements, such as iron, selenium
and antimony, follow the dense fraction, but not iridium. In site 465A the iridium
was not depleted in the pyrite-containing phase, but since that phase is so small it

contains only a small amount of the total iridium present.
Rucklidge: Do you have any comparable figures for the platinum?

Asaro: We have made only one measurement of platinum in site 465A, from the whole rock.

We have no fractionation or heavy mineral information.

Dilcher: We have discussed the composition of the Cretaceous-Tertiary bounday clay layer
in some detail. Have there been any comparable analyses of volcanic materials, or
deep mantle, shallow mantle or even deeper core materials which span the Cretaceous-

Tertiary boundary?

Asaro: The lavas that are most similar to the mantle in composition are the komatiites.
These were large nickel sulphide flows that occurred in Archaean times (Arndt et al.
1979). The platinum to iridium ratio (Naldrett and Duke 1980) is a factor of 1.7 to
1.8 higher than the C1 chondrite value for the classic komatiites. The Au/Ir ratios
were at least a factor 1.9 higher than the Cl values. There is an ultrabasic nodule
(Kaminskiy et al. 1974) which has a roughly chondritic Pt/Ir ratio (1.8). Other Eos.

which we have not considered here (e.g. Pd/Ir), do not occur in chondritic proportions.
Kyte: Were these xenoliths?

Asaro: Yes.

24

Kyte: Probably the best approximations of mantle composition can be determined from
xenoliths in basalts. The source of the noble metals in the upper mantle is believed
to be from late stage bombardment following segregation of the core (Chou 1978, Arculus
and Delano 1981). Lunar evidence indicates that this occurred up to four billion years
ago. Most estimates for the composition of the upper mantle include about 1% chondritic
abundances for noble metals. Estimates of platinum to iridium ratios are probably within
a factor of four of the chondritic abundances. However, these concentrations are
significantly lower than those in the Cretaceous-Tertiary boundary clay. A mantle

source for the latter would seem unlikely.

Dilcher: As a paleobiologist, could I conclude that the noble metals in the boundary clay

would likely not have a volcanic source?

Kyte: The mantle is an unlikely place to find noble metals in such a high concentration.
Even a plume from the core seems a rather unlikely possibility because one would
expect the outer liquid core of the Earth to be highly fractionated relative to its

initial composition.

McLean: Platinum group elements in the boundary clay have been acted upon by the
chemistry of seawater. Some of them are far more mobile than they were previously
thought to have been. Sodium-iridium-chlorine compounds have been made in the
laboratory that are soluble in water and it is reasonable to assume that they might
exist in nature. Platinum and iridium also have different calcophilic characteristics,
and the ratios in the boundary clay may not be comprehensible unless one can address
the chemistry of the oceans at that time. Some platinum group elements seem to go
into solution with organic acid compounds and the problem of what was distributed here
and there is difficult to come to grips with. Platinum-group concentrations are often

cited from basic rocks. Are these derived from sulphide phases?

Kyte: Most measurements of basic rocks were carried out on bulk samples. With respect to
Cretaceous-Tertiary sediments, we noted the ubiquitous presence of pyrite.
Subsequently, however, the iridium enhancement has been identified in oxidizing

environments.

25

Smit: Apart from that, one would expect the ratios to deviate from the chondritic ratio
due to the effects of oceanic environments. A pattern should emerge which deviates

from the chondritic pattern Ganapathy (1980) described.

McLean: Considering that the Earth is made up of accreted Solar System material, why
would one not expect that some areas within the Earth would not approximate

extraterrestrial ratios.

Asaro: It is not likely that one would obtain volcanic material from the upper mantle
which would have these ratios. It is difficult to obtain material which would produce
the proper absolute amount of iridium. Then marine processes and other effects would
be required to change the platinum and the gold to chondritic ratios within a few
percent. This is a very low probability situation. We have obtained consistent
measurements from two other places besides Denmark. Uniform, world-wide conditions of

marine chemistry and weathering would therefore be necessary to produce similar ratios.

McLean: Hakansson and Hanson (1979) have suggested that a world-wide clay event or
calcium carbonate dissolution event may have occurred. Chamley and Robert (1979)
suggest that the clays indicate a warmer, more aggressive climate. If in fact this

boundary clay represents a lag deposit over a relatively long period of time, of 50

to 100 thousand years, then extraterrestrial meteoritic dust would accumulate
as a concentrate in the clay. Barker and Anders (1968) discuss this point. The time of

deposition of the boundary clay must be known with better precision.

Kyte: Barker and Anders (1968) studied iridium and osmium concentrations in abyssal clays
which were accumulating at rates on the order of one millimeter per thousand years.
The concentrations were on the order of 100 times lower than in some Cretaceous-

Tertiary boundary sediments.
McLean: If the boundary clay is a lag deposit, could part of it be extraterrestrial dust?

Kyte: If the modern flux rate is assumed to be normal, and if this rate was valid 65
million years ago, the boundary clay would represent on the order of a million years

of accumulation.

26

Thierstein: It would take only 18,000 years to deposit all the iridium present in the
Gubbio bounday clay at present day iridium accumulation rates in the deep sea, if

one assumes:

- today's average accumulation rate of iridium in deep-sea sediments

Rp = 3.5 x 10 10 g cm ? ka ! (Crocket and Kuo 1979, p. 840)

- the insoluble fraction of the 1 cm thick boundary clay at Gubbio is 44% (CF = 0.44)
and contains an average of 6.5 ppb of iridium (ppbIR) (Alvarez et al. 1980, p. 1097

and 1100).

- the sedimentation rate (Sc) in the Gubbio G-magnetic zone is 1.28 cm

ka ! (Kent 1977, p. 771)

- the specific gravity of clay pe = 2.7 q cm 3 (Keller and Douglas 1980), and

the porosity of clay @ = 0.2 (Pettijohn 1975),

3

then the total iridium contained in 1 cm” of boundary layer:

EIR = 1 x CF xp, x (1-#) x ppbIR = 6.2 x 102g

and the time required to accumulate that amount of Ir at present day rates

ee
Ro

Hsu: Then why are comparable iridium concentrations not found in the red clays in the
north Pacific cores? Sedimentation rates there are about 0.2 millimeters per thousand
years. At this rate, economically exploitable quantities of iridium should occur

within 100 thousand years.

Thierstein: The problem is to keep away all other terrestrial, detrital material from

the Gubbio clay.
Kyte: If deposition stopped at Gubbio, would it not have continued elsewhere?

Thierstein: Not necessarily. These calculations assume steady state mechanisms. Most of
the oceanic biomass is today recirculated in the photic zone. A very small amount of

it reaches the sea floor an even smaller amount is incorporated into sediments. If

27

one assumes, and I have not done these calculations, that iridium and some of the
other metals are fractionated and enriched even to a small extent in the biomass, and
if it were not transrorted to the ocean floor by fecal pellets of grazing organisms as
it is today, iridium could become enriched in the ocean. When the first grazers
reappeared and began wrapping the biomass up and transporting it to the ocean floor,
perhaps the organics might be recirculated, but the heavy metals might not be. There
was no steady-state mechanism at the Cretaceous-Tertiary boundary and the biochemical
cycling must have been interrupted because much of the biomass in the photic zone out

in the oceans was exterminated.

Hsu: This brings us back to the definition of a rare or catastrophic event, e.g. that a
major perturbance of the steady state produced an iridium anomaly. Why did this
particular disturbance of the steady state occur at that particular time? This

cannot be understood by involving gradualistic processes.

Thierstein: The point is, could the iridium enrichment in deep-sea sediments at the
boundary be a consequence of the biotic extinctions rather than a cause, as has been

proposed by K.M. Towe (personal communication).

Feldman: Thierstein said earlier that some 18,000 years might have been all that was
required to obtain the iridium enhancement from accretionary accumulation of
extraterrestrial dust. This would require the Solar System to have passed through the
darkest of the dark interstellar gas and dust clouds, with molecular hydrogen
number densities n(H>) ~ 10° cm ? and a standard dust-to-gas ratio

n(grains)/n(H)) * 8 x 10 13 (cf. Lang 1980). In this extreme case 10 ngm cm 2

of iridium could be accreted on time scales between 8,000 and 100,000 years. Thus, a

very unusual kind of accretionary scenario is required to produce the kinds of

abundances found in the boundary clay on a time scale of ~ 20,000 years. In the
original paper by Alvarez et al., (1980) the ratio of the iridium 191 to 193 isotopes

came out very close to the Solar-System value. Has further work been done on other

sections that would shed more light on this?

Asaro: We have only made that measurement on the iridium from the Gubbio boundary layer.

28

L. Alvarez: I might explain why. When Frank Asaro and Helen Michel first measured the
iridium ratios, they found values that differed from Solar System values by 5% and we
suspected we had evidence of a supernova. They measured it again and found it was 5%
off in the opposite direction, alerting them to the fact that there was a serious
systematic error in the measurements. The error was traced to two quite separate
effects. One was geometric and the samples had to be positioned at the same distance
from the counter within a few thousandths of a centimeter. The other was that the
samples had to be bombarded in the reactor in the same place. Because two samples
cannot occupy the same position, two standard samples of normal iridium were placed
three or four millimeters apart with the unknown halfway between. If an average of the
two standards was taken and compared to the sample, then the ratio could be determined
to within 0.2 to 0.3 of a percent. I have rarely seen an experiment that was more
difficult or required more insight to explain the systematic errors. Soon after, Smit
and Hertogen reported that the osmium ratios were the same to within a tenth of a
percent. We were uncertain how to accept that, simply because one can be off by 5%
and not even know anything was wrong. It is an exceedingly difficult experiment. One
of the isotopes has an 18 hour half-life, requiring work around the clock for a

couple of days.

Smit: Most of the work was done by Hertogen, who notes that the osmium isotopes 184 and

190 have relatively long half-lives.

L. Alvarez: The difficulty arises not from the half-lives but from the fact that the
gamma rays come out in different proportions and in coincidence. The counter may
pick up two simultaneous gamma rays and will then add the energies together as if they
were one. This effect leads to the loss of a gamma ray count, and it is very
sensitive to the geometry. So the samples were finally located on little sapphire
balls to within about a thousandth of a centimeter. I have asked a number of very
good physicists what could make the apparent isotopic ratios change so rapidly with
source-counter distance, and not one of them could explain it, until given some broad
hints. The answer showed up accidentally in this technique, which had never been used
before. Because Hertogen didn't mention all the problems he went through to obtain his

number, we think he was probably lucky.

29

30

Smit: You repeated the experiment twice with more or less the same results?

L. Alvarez: It was repeated, and had it come out 5% off in the same direction we would

have published it as evidence of a supernova.

Asaro: Perhaps the osmium ratios are not as difficult as the iridium.

L. Alvarez: We know they are somewhat easier, as we have examined the gamma ray spectra,

but not an order of magnitude easier.

Smit: Hertogen decided not to use the iridium isotopes, preferring radiochemically

purified samples of osmium.

L. Alvarez: This brand new effect was discovered by Asaro and Michel as they performed
the experiments, and has not been described in the literature where Hertogen and
others could have been aware of its existence. It would be interesting to remeasure

the osmium ratios.

Feldman: Were the osmium and iridium measurements on the same sample, from the same

section?

L. Alvarez: No.

Feldman: This is an important point to pursue, in order to be certain that the material

is of Solar-System origin.

L. Alvarez: There seems to be no information on how big a variation one could expect from

different supernovas.

Tucker: The idea that all these heavy elements originated from supernovas is suspect.
Most attempts to reproduce the patterns have failed, and the only attempt that has
succeeded, reproduces the pattern of abundances in a red giant (Cowan et al., in
press). A red giant of about one solar mass is a very common star, and one would
expect that the interstellar material was similar to the Solar System material.
Fluctuations of two would not be expected because mixing scales are now on the order of
two parsecs rather than hundreds of parsecs. It may not be useful to pursue these

experiments in view of their difficulty and the uncertain value of the results.

Asaro: With respect to a possible lag in the sedimentation of the boundary clay, there is
a way of looking at this which might be helpful. In Denmark, we have detected 31.5 PPB
of iridium in an approximately 1-cm-thick K-T layer. The clay is 44.5% non-calcareous,
yielding between 65 and 70 PPB of iridium per unit of clay. Dr. Kastner has studied
the clay mineralogy and states that 90% or more is authigenic, that is, not due to the
normal deposition of detrital clay. This implies that the abundance of iridium as a
function of normal detrital clay deposition is at least 700 PPB, or about the
chondritic abundance. The amount of time needed to deposit 5 mm of meteoritic material
is something like 10 million years. All detrital clay deposition must cease and the
result would be fairly pure meteoritic material mixed in with the authigenic

montmorillonite that Dr. Kastner found.

McLean: How does this observation translate to the boundary clays in other sections; are

they similar or does a different scenario apply to each one?

Asaro: To a good approximation, a different scenario applies to each one, in terms of the
major element abundances. In New Zealand, the boundary sediments are not clay; they
are limonite which is a hydrous iron oxide. We do not have this comparable information
for other sections. In Gubbio for example, there is a very considerable detrital
component in the boundary material as indicated by Dr. Kastner's work, and high rare

earth abundances.

Smit: In the Caravaca section, there is no difference between clay content below and
above the boundary, and the boundary clay. Only the basal millimeter of the 10 centi-
meter-thick boundary clay contains some smectite and is different in clay composition
from the remainder of the clay. The latter also contains a significant amount of

iridium.

Asaro: A similar situation occurs in D.S.D.P. site 465A. A considerable clay component
is spread out over some 50 centimeters and the iridium is also spread out. It isa
detrital clay and at this point we do not have an adequate explanation for the

phenomena.

Hsu: It is interesting that rare earth elements are depleted in the boundary clay in

contrast to conditions in plume basalts which are enriched in rare earths. Oceanic

31

basalts are usually depleted in rare earth elements.

Asaro: The rare earth abundances are low in some K-T boundary materials we have measured,

but they are not as low as in chondritic materials.

Hsu: Are they depleted with respect to the terrestrial average?

Asaro: Yes. The depletion I spoke about concerns cerium, which is chemically different
from the other rare earths. It is deposited in manganese nodules, leaving a
deficiency of cerium in seawater. A large cerium depletion in the acid soluble
components in clay in D.S.D.P. site 465A indicates it was laid down out of the sea-

water.

Hsu: The question of whether the elemental abundances could be volcanic could be
resolved by comparing the boundary clay rare earth data with a tremendous amount of

data from the Deep Sea Drilling Project on the mantle plume basalts.

Brasch: Is it possible that more than one catastrophic event occurred, such as
successive impacts of different types of bolides or perhaps a chain of volcanic events

following an initial impact?

McLean: A hot spot under the Earth's surface is where one of the deep origin mantle
plumes comes to the surface, and Morgan (1981) suggests that such an area will be
inundated by flood basalts. A hot spot is present for tens of millions of years
before the volcanism occurs, which might mitigate against a meteoritic impact causing

the plume. As a lithosphere plate migrates over a hot spot, it is domed up and

fissured, and lavas emerge from the fissures. Flood basalts are not characteristically

explosive in nature. I know of no explosive cratering associated with the Deccan

volcanism. Where is the crater of a large Cretaceous-Tertiary bolide today? Ejecta
from large craters are not associated with large quantities of iridium and osmium. It
would appear that some of the large earth-crossing asteroids which have impacted did

not produce large amounts of siderophile ejecta.

Grieve: Not all, but some have. The impact melts in about 10 craters have identifiable

siderophile enrichments, often at about one to two percent in C1 abundances.

32

Kyte: Some impact melt rocks in Clearwater Fast contain up to 7% chondritic abundances

(Palme et al. 1979).

Hsu: Are impact melts composed largely of terrestrial materials?

Grieve: The impact melt is made up of melted target rocks, which are contaminated by
material from the projectile. The nature of the impacting body can be identified

through the contaminants.

Hsu: Therefore, the reason why iridium anomalies do not occur more frequently in
sediments is because the ejecta from the impact of solid objects does not have a high

iridium content.

Kyte: Impact events, in general, should produce ejecta that is only slightly enriched in

iridium.

Hsu: The bolide, then, disintegrated in the air, and this is why an iridium-enriched

layer occurred.

Kyte: We suggested the possibility that a cometary object was disrupted tidally within
the Roche limit, and broke into several fragments which then disintegrated in the
atmosphere. The material was not diluted by large amounts of crater ejecta. This
could lead to an accretion event that would span several months or perhaps several
years. Another possibility would be an enriched source, such as an iron meteoroid.
However, the chromium enrichment in the boundary horizon might be significant enough

to rule out an iron meteoroid.

L. Alvarez: A large asteroid would probably not be broken by tidal forces as it passes

by.

Kyte: It is a complicated problem. Opik (1972) suggested that if the proto-Moon had
passed within a few Earth radii of the Earth it would have been disrupted by tidal

forces.

Halliday: The broken cometary fragments will continue in essentially the same orbit. If
the nodes of the orbits are near the Earth's orbit, then the probability is high that

they will collide with the Earth within a few tens or hundreds of thousands of years.

33

W. Alvarez: There was a news article in Science (Kerr 1981) recently about new evidence
that asteroids may possess satellites of their own, or that two similar-sized
asteroids may tumble about each other. Would this allow a binary set to impact

simultaneously, an otherwise extremely improbable event?

Halliday: This was considered many years ago for the Clearwater Lakes. An astronomer
from the former Dominion Observatory investigated this and found that the configuration
would be stable against planetary perturbations, but the objects would diverge in the

last few hours before impact.

W. Alvarez: Apparently there is tenuous observational evidence, based on star

occultations, of binary asteroids.

Halliday: Several have been reported and are still debated.

L. Alvarez: Several have been seen in the same neighbourhood by independent observers.
Some (Kyte et al. 1980) have suggested that the biggest problem with the asteroid
theory was that there was too much iridium in the boundary layer. We used the number
Richard Grieve supplied to us that an asteroid would be diluted by 60 or 70
times its own weight of crustal material. In order to further reduce the amount of
crustal material ejected, it could be postulated that the asteroid impacted in the
ocean and some reasonable fraction of its energy was dissipated by vapourizing water.

What is your best estimate to date?

Grieve: Massive impact events are poorly understood. Our knowledge of crater mechanics
is largely derived from small-scale experiments, such as TNT or nuclear explosions.
These are very small in comparison to major impacts. Various computer codes, which are
very expensive, can model the flow field, excavation process, and amount of material
excavated by craters up to a few kilometers in size. For the 3.8 km Brent crater,
the dilution factor of the volume of material excavated relative to the volume of the
projectile is modelled to be of the order of 3-400 (Grieve and Cintala, in press).
The amount of material excavated depends on initial conditions, i.e. for a given
projectile size, the nature of the target, the density of the projectile, and its
velocity. We have been attempting to model large impact events, such as the 40 million-

year-old, 100-km-sized Popigai crater in the Soviet Union. In that case it seems that

34

the flow field changed with time. In small impacts gravity does not act upon the flow
field, and the calculations are simpler. In larger structures it appears that dilution
factors may approach the thousands. With the exception of E. Clearwater, impact melts
concentrated in the centre of the crater, rarely contain more than or one or two percent
of Cl abundances (Palme et al. in press and references cited therein). In the few
instances where we have examined ejecta thrown from the crater, siderophile enrichments

were not detected. Perhaps a real problem exists in the case of dilution effects.

L. Alvarez: If, using the asteroid model, it is difficult to account for the iridium
anomaly in Denmark, then the difficulties are exacerbated by a cometary model for two
separate reasons. Firstly, a comet is half ice and secondly, it impacts at higher
velocities. Iridium concentrations could thus be reduced by a factor of 100. An

asteroid would be preferable.

Kyte: There can be different sources of cometary material. There are long-period comets
which go beyond Pluto, and there are short-period comets. Also, some Apollo asteroids,
may be extinct comets. In order for the tidal disruption and accretion model to work,
the object has to occupy a nearly circular orbit approximating that of the Earth. A

low encounter velocity is required.
L. Alvarez: Such an object could be called an asteroid.

Kyte: The asteroid belt is generally believed to be the source of ordinary chondrites,

while the volatile-rich comets may be a source of carbonaceous chondritic material.

L. Alvarez: What appeared to be a disagreement between our two groups is really not all

that severe, now that we have had a chance to discuss these matters with Dr. Grieve.

Feldman: Aggarwal and Oberbeck (1974) discussed the effects of the Roche limit on a solid
body. Break-up occurs for a body approaching either the Earth or the Sun only if the
tensile strength of the body is less than of the order of 106 dynes cm ?, which is
comparable to that of glacial ice, assuming the incoming body is about 10 kilometers in
diameter. Allowing for equal amounts of icy volatiles and iridium-bearing material in
the cometary nucleus, its size would be somewhat larger than 10 kilometers. Break-up

might then occur just outside the Earth's atmosphere or the Sun's photosphere. In the

more attractive case of a comet breaking up as it passed the Sun the dispersed material

35

might be accreted by the Earth over a number of its yearly orbits. This could yield
a series of iridium anomalies. An asteroid approaching the Earth would not break up.
The lowest estimate that I've seen for the tensile strength of asteroidal rock, cited

by Aggarwal and Oberbeck, is of the order of 107 dynes cm 2

Kyte: Break up can occur at virtually any velocity. A very low encounter velocity would
allow the Earth to capture a significant amount of material at one point in time,
rather than sweeping it up on another pass 10,000 years later. In a low velocity
encounter, break-up would be followed by an accretionary event, in which case objects
in elliptical orbits, involved in a 3-body problem which includes the Moon, would

probably be accreted over a period of several months.

K. Hsu: Two craters which are about 65 million years old occur in the southern part of
the Soviet Union. The craters contain impact melts and breccia with Cretaceous
fossils, over which were deposited marine sediments of earliest Tertiary or Danian age.
The Kamensk crater is about 25 kilometers in diameter, and the Gusev crater to the
northeast is 3 kilometers in diameter (Masailis 1976). Is anyone aware of additional

information pertaining to these craters?

Grieve: Craters are being located at a rate of one or two per year in the Soviet Union.

Another crater 60 kilometers in diameter, Kara, has been dated at 60 + 5 million years

(Masailis et al. 1980).
Hsu: Many such craters could form after a cometary breakup.

Grieve: A comet is a low density body and may not penetrate deeply into the target. The

amount of material excavated would be proportionally less.

Feldman: The Tunguska event in 1908 is now identified as a fragment of the periodic
comet Encke (Kresak 1978). Spherules found by Glass (1969) in the Tunguska dust show
an anomalously high potassium-to-sodium ratio, but not as high as in those found by

Dr. Smit. Was the ratio you found about a hundred to one?

Smit: Yes. In the level of the highest iridium concentration, at the very base of the
clay layer at Caravaca, we found thousands of tiny spherules consisting of sanidine.

They did not form authigenically on the sea floor, but must have cooled from

temperatures of about 600 to 700 degrees centigrade. Only the occurrence in Caravaca
is known, where they occur at a density of about 500 per cubic centimeter. We
suggested they were derived from an iron meteorite, one specimen of which is known to
contain large single crystals of potassium feldspar with a similar potassium-sodium
ratio. I was not aware of the Tunguska comet fragment containing a high amount of
potassium. If comets are not fractionated how can high potassium concentrations occur

in them?

Halliday: Is it reasonable to think of this as a differentiation in the atmosphere, when

the material became molten?

Kyte: Potassium is one of the more volatile elements in meteoritic material and would
be likely to vapourize. It also might be likely to condense rather rapidly from a

vapour.

Smit: We were surprised by the enormous amount of spherules in the clay layer at Caravaca,
where they form a nearly continuous surface. Perhaps this is because the site was

situated in a strewn field.

W. Alvarez: Were the Tunguska spherules found precisely at the site?

Halliday: They were distributed by the wind, and carried from the southeast to northwest

of the site.

Kyte: Did the Tunguska spherules contain chondritic material?

Halliday: Analyses are only available for the major elements, not for the trace elements.

Kyte: We have found a horizon in a core from the Antarctic basin a little over two
million years old. It has an iridium anomaly (Kyte et al. 1981) comparable to that at
the Cretaceous-Tertiary boundary, and is also full of chondritic debris, apparently
from an atmospheric explosion or oceanic impact of an object. The mineral chemistry
and texture is similar to chondritic ablation spherules normally found in deep-sea

sediments.

3ÿ2

Halliday: Different kinds of spherules were found at Tunguska, including magnetite ones,

silicate ones, and occasionally even a silicate one with a smaller magnetite one

inside, all with an average diameter of 80 to 100 microns.

L. Alvarez: I am interested in hearing suggestions on what should be done to demonstrate

more fully the validity of the asteroid impact hypothesis. Our resources are
limited, and should be concentrated on the task of obtaining that kind of data which

would be most useful in making a convincing case.

Kyte: I think there was a large accretionary event at the end of Cretaceous time.

Several models on how this took place have been suggested, including an asteroid
impact, tidal disruption and accretion of numerous small bodies, and accretion of
cosmic dust clouds. Methods should be found to test these hypotheses. Concerning

the asteroid impact model, one test would be that both meteoritic components and
ejecta components should have been distributed on a world-wide scale. Does the ejecta
component exist in all the sites that have been investigated and is it of relatively
uniform composition? For example, if there was an oceanic impact, a large fraction of
mantle material should occur in the ejecta and be correlated from location to
location. If the meteoritic signal is diluted by only normal terrigenous sediments,
then perhaps something like the interstellar dust cloud or the tidal-disruption model

should be favoured.

McLean: We must entertain alternatives and not fix upon one model too quickly. The

38

boundary clay, if it is a lag deposit resting on top of a Cretaceous unconformity,
might account for the range truncations below it and the illusion of a catastrophe. A
world-wide regression of the seas occurred at the end of the Cretaceous, which would
have created unconformities in continental sections, and removed the terminal
Cretaceous record there. Hiatuses are known to occur at the Cretaceous-Tertiary
boundary in continental sequences. If the boundary occurred 65 million years ago it
coincided with the onset of the Deccan Traps. The extinctions were most severe in this
region of the world. It could also be postulated that acid volcanic gases were injected
into the atmosphere and passed into shallow oceanic waters through gaseous exchange

where calcite-producing organisms became extinct. In the marine realm a carbonate

dissolution event is known to have coincided with the Cretaceous-Tertiary boundary.
Loss of record could again create the illusion of simultaneous extinctions. The
paleontological and stratigraphical record in the vicinity of the range truncations,
and the length of time represented by the boundary clay are two questions which should

be more carefully considered.

Smit: Dinoflagellates have a turnover rate comparable to that of planktic foraminifera.
If range terminations of the planktic foraminifera occur, similar range terminations

should be evident for dinoflagellates, which is not the case.

McLean: Dinoflagellates have organic walls and would continue to settle down. In fact

they are quite concentrated in this boundary clay.

Thierstein: In general terms, the geochemical cycles of rare and noble elements should
also be examined more carefully. How much of that material is dissolved in oceanic
waters or stored in marine and terrestrial biomass? Another problem is identifying
the mechanism that produced the significant extinctions of the calcareous marine
phytoplankton. These organisms have high turnover rates and high regeneration rates.
The only survivors with an adequate record prior to the Cretaceous-Tertiary bounday
among the calcareous phytoplankton are Braarudosphaera and Thoracosphaera, and they
are considered to be cysts. Dinoflagellates can encyst themselves, and survive for
many months on the ocean floor at a depth of over 200 meters and then resurface and
reproduce (Williams 1978). They can survive adverse environments in the surface.
Could the asteroid impact model provide an effective mechanism to eliminate the
phytoplankton, upon which many organisms higher in the food chain depend? Such a
mechanism would be the screening out of light, although living calcareous phytoplankton
is known to survive for months at light intensities of less than 1% of that available
at the ocean surface (Blankley 1971). It is physically difficult to imagine a
stratospheric dust cloud that would screen out 99% of the currently incident light
for extended periods of time (Thierstein, in press). At the densities required,
preliminary calculations suggest that it is impossible to keep enough material in the
stratosphere for more than a few weeks. Atmospheric modelling would be useful.
Another possibility which I have favoured, and which I still think is conceivable,

would be a significant salinity change in oceanic surface waters. The model is

39

unfortunately virtually untestable because of the very rapid exchange rates required

between a freshened ocean basin and open ocean environments.

Volcanic sources have recognizable signatures in the rare earth elements. Would it
be possible to compare the rare earth elements of the boundary layer more

systematically with the various materials from basalts?

Asaro: We have analyzed all of our samples for their rare earth abundances and have in

addition many analyses of akalic, oceanic and tholiitic basalts made as part of a
geochemical programme. The rare earth patterns in the acid-insoluble residues and

whole-rock samples from the K-T boundary regions are different from the basalts.

W. Alvarez: Are the extinctions synchronous all over the world, which one would expect

40

if they were the result of a sudden event like an impact? This would seem to be

the case in the marine environment. There are two sections in Italy in which the
extinction occurs in a particular magnetic polarity zone. In site D.S.D.P. 524 in

the South Atlantic, the extinctions occur in the same polarity interval. A subsequent
leg in the South Atlantic reveals the same relationship. The extinctions also occur

in the same polarity interval at Caravaca. Problems remain in the two terrestrial
sequences, containing dinosaurian remains, which have so far been studied. In the

San Juan Basin section in northwestern New Mexico, which was studied by Butler, a

well substantiated gap in the section occurs right at the boundary. A channelled
unconformity with conglomerates seven meters thick is located at this level (Baltz

et al. 1966). One cannot, therefore, be certain of the correlation of the polarity
zones. The other section is in the Red Deer Valley, in Alberta and was studied by
Lerbekmo, Evans and Baadsgaard (1979). In this section it would appear that the
dinosaurs became extinct at the same time as did the marine microfauna, but to a
paleomagnetist the reversals do not seem very clean. I have had many discussions with
Butler, who carried out the work in the San Juan Basin, and we have come to the
conclusion that with the data available from the land sequences the correlations remain
uncertain. Terrestrial sequences should be sought in which good paleomegnetic
stratigraphic studies can be done. A thick enough sequence is required so that a
characteristic fingerprint of long and short, normal and reversed polarity zones can be

obtained which will permit confident correlations.

Rucklidge: Background abundances of trace elements should also convincingly be
established, so that similar lithological samples from different levels in the
stratigraphic column can be compared. Pyrite from an Ordovician shale in Ontario
contained three PPB of iridium in it, which is greater than the anomaly quoted in some

of the boundary layers.

Russell: I have compiled data pertaining to genera of organisms that have been identified
in strata which were deposited during the last 5 to 10 million years of the
Cretaceous and the first 5 to 10 million years of the Tertiary (Table 1). The
numbers represent a state of the art and certain serious sampling inhomogeneities.
They should not be interpreted in the same sense as the numbers which were being
considered earlier. They constitute the paleontological basis for recognizing an

event in the fossil record which is quite interesting.

Hsu: The timing of the extinction should be defined as closely as possible. One of the
best sections for paleomagnetic stratigraphy, which is not affected by facies
changes or other complications, is the D.S.D.P. site 524 in the South Atlantic
(Fig. 3). One of our major objectives was to measure the interval of time represented
by the polarity intervals. The core passed through 200 meters of Paleocene strata and
then 100 meters of Cretaceous sediments, all of which were undisturbed. Coring was
nearly continuous. The extinction took place in the Anomaly 29 reversed interval, as
in the Gubbio section. The correspondence between site 524 and the Gubbio section in
other polarity intervals is good. The thickness of the polarity intervals in the
sediments is correlated with the width of the sea-floor-spreading magnetic anomaly
patterns in the South Atlantic so that age limits can be obtained for various polarity
intervals. Our data confirm the estimates of Kent, and Anomaly 29 reversed interval
lasted about 480 thousand years (Hsu et al., in press).

Sedimentation rates for the four intervals identified were of order of 30 to

25 meters per million years, or about 3 centimeters per thousand years. Time
resolutions of the equivalent of one centimeter's deposition could be approximated.
Suggestions of an extraterrestrial event were reflected in the vertical carbon and
oxygen isotope excursions. The stable isotopes of carbon are 13c and l?c and are

usually expressed in delta 13C in PPT. A shift of minus 3 parts per thousand occurred

41

TABLE 1. WUMBER OF GENERA OF FOSSIL ORGANISMS CURRENTLY RECOGNIZED AS HAVING LIVED PRIOR TO
AND FOLLOWING THE TERMINAL CRETACEOUS EXTINCTIONS. *

Fresh-water organisms

charophytes (algae) 19 ial
cartilagenous fishes 4 2
bony fishes LL 74
amphibians 10 12
reptiles 3) mi
57 49 86%

Terrestrial organisms
(including fresh-water organisms)

charophytes (algae) 19 ab
higher plants 100 90
snails 16 18
bivalves 10 7
cartilagenous fishes 4 2
bony fishes al 7/
amphibians 10 12
reptiles 58 25
mammals _24 (11) 30
252 202 80%
Floating marine microorganisms
acritarchs 28 10
coccoliths 43 4
dinoflagellates 95 90
diatoms 10 10
silicoflagellates 4 3
radiolarians 63 63
foraminifers 18 3
ostracods 79) _40
340 223 66%
Bottom-dwelling marine organisms
calcareous algae 47 47
sponges 261 81
foraminifers 95 93
corals 87 31
bryozoans 387 204
brachiopods 28 22
snails 300 135
bivalves (with 52 gen. rudists) 381 129
barnacles 32 24
malacostracans 69 52
sea lilies 16 16
echinoids 150 69
asteroids 36 35
1839 938 51%
Swimming marine organisms
ammonites 34 0
nautiloids 10 7
belemnites 4 O
cartilagenous fishes 14 10
bony fishes 38 10
reptiles 0) 2
130 29 22%
Overall totals: 2561 1392 54%

* Modified from Russell (1977) after data in Bujak and Williams (1979), Carpenter (1979),
Danilchenko (1978), Glezer (1977), Korde (1977A, 1977B), Kues et al. (1980), Lupton et al.
(1980), Molnar (1978), Naidin (1979), Naylor (1978), Rasmussen (1979), Sohl (1960),

Van Valen (1978), Wall and Galton (1979), Whetstone (1978) and Weishampel and Jensen (1979).

42

20 4 60 8 109% CaCO3

%o

—
°
oO

nm
>
°
a
~ <
o j
« 1.0
pe
c
[e]
O
0 :1
£ 0.01 =
°
= O
= -0.01 7
o he
-0.1
c
o æ | ©
Oo =
= =
© 0
~ -1.0 20 40 60 80 100% = oat =
un K Nannofossils s Ir ppb.
— ”
Li
Q 2 ©
ra =
œ
o
-10.0
| -20.0
-30.0
-40.0
-50.0 CY
m % 00
20 40 60 80 100° =. -1 +1 + +3

FIGURE 3. The tridiwn-spike detected by Urs Kriahenbiihl indicated the deposition of extraterrestrial
fallout several hundred years before the first appearance of the Tertiary "taxa" of

nannofosstls.

Work done by Steve Percival and Katharina Perch-Ntelsen showed the raptdly declining
percentages of Cretaceous "taxa" of nannofossils in the earltest Terttary sediments
depostted during the first 30,000 years after the extraterrestrial event.

Work done by Q.X. He, Judy McKenzte, Hedy Oberhänsli, and Helmut Weissert indicated: 1)
the rapid decrease of CaC03 in latest Cretaceous and earliest Tertiary sediments,
stgntfying perhaps a drastte reduction of biomass and a rapid rise of CCD; 2) oxygen-
tsotope shift, signifying perhaps an initial cooling of about 5°C, followed by a 10°C
rise in temperature; and 3) a systematic carbon-shift, signifying perhaps an increase of
dissolved C02 in the oceans with an anomalously high proportion of ‘?C atoms. Two data-
points near the contact were based upon isotope analyses of samples containing less than
0% CaC03, which are not reliable.

The text of the work produced by this diagram is a manusertpt submitted to Setence,
entitled Environmental and evolutionary consequences of terminal Cretaceous event, by
K.J. Hsu and 19 co-authors.

43

dd

across the boundary. This is a very large shift. If all of the biomass on land

were placed in the ocean enough ‘2c would be available to cause a shift of only minus
2 parts per thousand. It took place within a very short time, and certainly less
than a few hundred thousand years. We are not certain if this shift took place in
less than 300 years or less than 30 thousand years. Simultaneously, a shift occurred
in the oxygen 18/16 ratio. Boersma et al. (1979) have found a shift in both benthic
and planktic organisms, amounting to one to two parts per mille or a thermal warming
of four to eight degrees centigrade. The thermal excursion might have had devastating
effects on dinosaurs (Hsu 1980). Did it occur immediately or did it build up over a
few thousands, or tens of thousands of years? The work by Dr. Alvarez and others
indicates, to my satisfaction, that an extraterrestrial event took place and we need
to know more about it. It is no longer useful, in my opinion, to question whether or
not such an event happened. What stresses caused the biologic catastrophe? Perhaps
this can be approached by examining the rate at which various environmental changes
took place. A paleontologic boundary must first be recognized on the basis of
frequently occurring nannofossil species. About 30 nannofossil species occur in
Cretaceous sediments. The first appearance of four Tertiary species defines the base
of the Tertiary. Many Cretaceous species are found up to a level which is equivalent
to about 40 thousand years after the beginning of the Tertiary. They were either
descendents of survivors of the holocaust, or dead organisms which were reworked by
sedimentary processes or mixed up to this level through bioturbation. We do not know
which alternative is correct. Below the boundary 100% of the species are Cretaceous;
by definition there are no Tertiary species. During the first 40 thousand years of the
Tertiary many of the microfossils belong to Cretaceous species. In fact the first
Tertiary sample contains only 3% Tertiary forms.

The base of the iridium spike, with abundances of 2.4 to 3.3 PPB against a
background of less than 0.1 PPB, occurs one centimeter below the paleontologic
boundary. Note that the paleontologic boundary is arbitrary, and that the cometary
impact may actually have occurred during the last few hundred years, or few thousand
years of what we call Cretaceous time. Calcium carbonate concentrations reflect

biologic productivity; during the Cretaceous time the CaCO3 content of sediments was

about 30%, it then dropped to 2 or 3% following a mass mortality after the cometary
impact, and returned to about 30% approximately 100 thousand years later. Again,

the decline in the CaCO3 content of the sediments occurred below the earlier
paleontological indicators of Tertiary time. A carbon isotope excursion of minus 3
parts per thousand is found and it is about as large as any documented in the
extinction interval. The delta !3c declined logarithmically, to a minimum value after
30,000 years and slowly returned to the previous level after about 100

thousand years. There was also an excursion of paleotemperatures as indicated by

the oxygen isotope data. However, the trend is more erratic. A decrease of 3-4°C
coincided in timing with the iridium spike, followed by an increase of 6-8°C during
the next 30,000 years. Conditions then gradually returned to normal as carbon-isotope
values returned to normal (Hsu et al., ms.).

We are now considering the possibility that there are two phases in this
catastrophe. The first phase was a mass mortality, the biomass in the oceans was
destroyed to a great extent. The second phase of mass extinction resulted from
environmental consequences of the initial mass mortality. If the oceanic biomass were
drastically reduced, carbon dioxide utilization in photosynthesis would also decline
sharply. The oceans would become more acid, and a carbonate dissolution event would
occur. What brought about the carbon isotope excursion? Rivers deliver carbon atoms
to the ocean in the form of dissolved carbonate and dissolved organic carbon,
representing an influx of isotopically light carbon. How much time would be required
for the observed carbon isotope excursion to occur if the isotopically light carbon
were not being deposited in oceanic sediments through interactions with oceanic
biota? Various models suggest that this change can be produced in 40 thousand years
if an average of one-third of the oceanic biomass were destroyed. From the point of
view of step functions, initially on the order of 90% of the biomass could have been
eliminated, and near the end the reduction may have been only by 10%, producing the
logarithmic change in 1306. The modelling also suggests an increase in CO» content of
the atmosphere by about 5 to 10 times. The cooling trend, if real, could have
resulted from the insulation of sunlight by ejecta from an impact event. The
subsequent warming could not be associated with atmospheric heating during the fall of

the meteor, because the warming required a process operative for a longer term. We are

45

thus favouring the CO» greenhouse model for the warming. Are the data derived from
this section exceptional or are they in accordance with a general rule? I suspect

they are typical.

Thierstein: The replacement of the Cretaceous by the Tertiary flora and fauna usually

Hsu:

Clube: Do you exclude extraterrestrial input of

Hsu:

46

occurs over an interval of 75 centimeters to 3 meters. Could that indeed be a
transition? One way to approach this question is to examine the evolutionary patterns
within the surviving Cretaceous assemblage in basal Tertiary sediments. If there was
a gradual deterioration of Cretaceous taxa, abundance flunctuations should be apparent.
In the three sections I have examined there is no drastic change in the relative
abundances of the Cretaceous taxa above the boundary. This leads me to believe that

the Cretaceous fossils are indeed reworked by lateral transport and benthic mixing.

In either case, the transition represents about 40 thousand years, at most. We are
not discussing normal evolutionary changes. Even if the nannofossils are not reworked,
we are "violently agreeing" in the sense that the time scales involve between a few
hundred and a few tens of thousands of years. Dr. Perch-Nielsen, one of our co-workers,
has studied nannofossils in 10 or 12 European sections and has identified no large
paleontologic hiatus. I do think that a terminal Cretaceous catastrophic change in
the pelagic environment was real, and the change occurred synchronously over a wide
part of the ocean areas. Perhaps the extinction of the dinosaurs occurred a few tens
of thousands of years later, because of other environmental effects. A maximum of

temperature was attained near Anomaly 29 positive in our core.

Le in your modelling, because this

could be a very powerful test of meteoritic versus cometary impact?

After we defined the trend that I have described, we analyzed additional samples

deposited at a time thousands of years after the catastrophic event. We found a 36

anomaly of minus 2.5 to 3 parts per mille. The three points for this anomaly are based
on five or six replicate analyses of surface-dwelling organisms, after benthonic
‘organisms were removed from the samples. This could be interpreted as evidence of

extraterrestrial input into surface oceanic waters. A very small amount could change

the composition of surface waters (Hsu 1980). Unfortunately this is the only site

where this has been identified so far. Perhaps it will be found in the Tunisian

section.*

31072

Feldman: The interstellar value of the ratio of !?C to l°c is 67 + 10 (Penzias 1980).

What is it on Earth?

Hsu: On Earth 1.11% of the carbon atoms are l3C and 98.89% are l2c. It is possible that
carbon is fractionated in comets because 12c may preferentially enter the ice phase.
The delta l3C of a comet with all of its water and dry ice might be of the order of
minus 15 parts per mille. A body of 1018 gm could produce a sigral in the surface

layers of the ocean, but not within the ocean as a whole.

McLean: The 13C value of volcanic CO» and reduced carbon in igneous rocks is quite low,
being of the order of minus 7 to minus 30 (Hoefs 1973, Javoy and Pineau 1978).
Volcanic CO» flooding into the oceans over a long period of time could account for a

portion of this drop in 13C values.

Thierstein: An additional biologic complication exists. Individual species of
nannoplankton fractionate carbon and oxygen in different proportions. The range is
about three per mille in living coccolithophorids (Dudley and Goodney 1979), and the
shift across the Cretaceous-Tertiary boundary could in part be due to vital effects
caused by the change in the assemblages. A hint of this is provided by the isotopic
shift in the benthic foraminifera at D.S.D.P. site 356, which actually become heavier
rather than lighter in carbon across the boundary. Several mechanisms could be

involved in producing the shifts.

Russell: The swing in carbon isotope balance could have been more profound in the
atmosphere, although no stable carbon isotope measurements have been taken in
terrestrial boundary sequences. Perhaps this is an area worth looking into. In
terrestrial environments of deposition in Montana, Clemens and Archibald (1980) have

noted, as I have too, a separation of several meters between the highest articulated

* A similar °C excursion was subsequently found in Tunisia and Germany across the
Cretaceous-Tertiary boundary (K. Hsu).

dinosaur bones, and the horizon of the palynofloral change usually associated with the
Cretaceous-Tertiary boundary. In terms of terrestrial sedimentation rates, this

might represent tens of thousands of years.

Jarzen: I have studied two sections in western Canada, where sampling at close intervals

has been carried out from the last Triceratops remains to the number 1 coal, or Z

coal, or "boundary" coal, and then to several meters above the coal. There does not
appear to be a drastic reduction or change in the palynofloral content. My efforts
have been primarily concerned with the angiosperm component, but the spore and
gymnosperm components have been examined as well. The changes that have been observed
involve two major groups of pollen forms; one typical of the western interior of

North America and Siberia called Aquilapollenites and the other typical of eastern
North America and Europe called Normapolles. These two groups do in fact suffer

major extinctions. The remaining angiosperms seem to pass through the transition
relatively unaffected. From 10 to 40% of the palynomorph taxa become extinct,
depending on the palynologist, the section sampled and perhaps the sampling interval.
The change observed in sampling at close intervals may be less than the change observed
in sampling at widely-spaced intervals. In both of the sections from Saskatchewan,

and in a few other sections from this province, in the last two or three meters of the
Cretaceous the preservational quality of the palynomorphs decreases so drastically as
to render the identification of some of the palynomorphs impossible. Twenty
centimeters below the boundary coal the samples become for all practical purposes
barren. The coal itself is often barren, but above that coal an abundant palynoflora is
again preserved with many of the same taxa present that were observed below. Why does

this drop in preservational quality occur?

McLean: One thing that seems to be causing a lot of confusion is the definition of the

48

Cretaceous-Tertiary boundary in the western interior. Brown (1962) chose the first.
persistent coal above the highest occurrence of dinosaur bones. That reflects a
tectonic change. Leffingwell (1971) identified the Cretaceous-Tertiary contact with
a pollen change. This is nearly 20 meters below Brown's boundary coal, and is the
same pollen break that Lerbekmo et al. (1979) believe is synchronous over Montana,

Wyoming and Colorado. The pollen break in Colorado is about 35 meters below volcanic

strata that have been dated at about 6€ million years. A hiatus is generally present

in sequences around the western intermontane basins, and the most continuous sedimentation
occurred in the inner parts of the basins that are usually only accessible by the drill
bit. Here, there seems to be a sort of gradation of palynofloral changes across the

contact.

Hsu: Hickey (in Kerr 1980) has stated that there was a great floral change in northern
temperate or arctic regions, but a minor change in lower, more tropical latitudes.

What is your impression?

Jarzen: The figures cited by Dr. Hickey imply that perhaps 70 to 80% of the species in
western North America became extinct. I do not believe that. The last time I spoke
with Dr. Tschudy he felt that the extinctions were less significant than that. The

citation is either a misprint or we were misquoted.

Hsu: He quoted you with reference to the extinctions in the tropics.

Jarzen: I don't know about the extent of the extinctions in the tropics.

McLean: Dr. Hickey accepts Leffingwell's (1971) pollen break as the Cretaceous-Tertiary
boundary. Leffingwell in turn suspects that hiatus control or missing section is the

cause of this pollen break and is due to tectonism.

Russell: In deltaic sections in Montana and North Dakota, Moore (1976) noted some minor
scouring near the transition into Tertiary strata. With a possibility of wave activity
and flooding of marginal lands as a consequence of an oceanic impact event, perhaps this
should be examined more carefully. However, a hiatus has not been detected in the
lithostratigraphic sequence in the eastern Fort Peck Reservoir region, where there is
an impressive continuum of change in the highest levels of the Hell Creek Formation.

The badlands there contain large-scale, cross-bedded riverine sands within a matrix of
finer overbank and floodplain deposits. The sands contain debris of different kinds of
dinosaurs and a relatively large quantity of bones of smaller vertebrates. The highest
level of occurrence of Triceratops coincides with a change in the quality of deposition,
the visual impression of which is very powerful. Patterns of sedimentation gradually

become more laminar, the scale of the cross-bedding decreases and within about five

49

meters the sequence has changed into the laminated sandstones and coals typical of basal
Tertiary deposits. There is a feeling of a massive but gradual shifting of gears from

Cretaceous to Tertiary depositional patterns.

McLean: These changes may be the result of tectonic factors. It may be well to suspect
the possibility of a hiatus until proven otherwise, because a hiatus occurs in so many
places. Which datum in terrestrial sediments - the Z coal or the palynofloral break -
corresponds to the marine Cretaceous-Tertiary boundary? There nay be a significant
time differential between the various reference horizons that radiometric dating could

resolve.

Hsu: Radiometric dating of rocks of this age are only accurate to within plus or minus

one million years. Paleomagnetic resolution is better.

McLean: Lerbekmo and his collaborators (1979), using magnetostratigraphic evidence, view
the dinoaurian and marine extinctions as synchronous, while Butler and his colleagues
(1977, 1978, 1979) using the same kind of evidence, interpret the dinosaurian

extinctions as occurring much later than the marine extinctions.

Hsu: At most, the discrepancy could not exceed one million years. Dating techniques,
whether radiometric or magnetostratigraphic, have their limitations. For example,
radiometrically datable ash beds have not been located at the boundary in marine

sections. Incidentally, where does the iridium anomaly occur in the Montana section?

Asaro: It was on the platinum engagement ring of the person who prepared our samples. We
had found an 1l-peak iridium anomaly in Montana and became concerned that iridium was
so easy to detect. The iridium was found on the aluminum foil used to wrap the samples,
and correlated with the order in which the samples were wrapped. The person who wrapped
our samples had platinum wedding and engagement rings. We measured the platinum to
iridium to gold ratio on the aluminum foils that touched her rings and found they
conformed, within experimental counting errors, with the platinum to gold to iridium
abundances in the contamination. I must add that we were overjoyed to hear that the
Los Alamos group had discovered a continental iridium anomaly in New Mexico. None of
the measurements that we have published were affected. Those of you who associate

at one time or another with people who are married may be interested to know that if a

50

platinum wedding ring loses 10 percent of its mass in 30 years, the average erosion
per minute is about two orders of magnitude higher than our sensitivity for measuring
iridium. We re-examined our previously detected gold anomalies and discovered these

could be correlated 100% with the gold wedding rings worn by another person.

W. Alvarez: Returning to the paleobotanial record, how does one interpret the conclusion
of certin workers that 90% of the palynofloral taxa become extinct in the
Aqutlapollenites province, but there was no change in the flora at the end of the

Cretaceous?

Jarzen: Aqutlapollenites represents perhaps 10% of the total flora in the Agutlapollenites
province (Oltz 1969). Some 90% of Aqutlapollenites species are no longer in the
Tertiary. One or two species are found in Paleocene strata and some have recorded the
genus from the Eocene, so the form did not become extinct. The same is true of the
Normapolles lineage or group in eastern North America. More than 80 or 90% of the
taxa became extinct, but studies by Tschudy (1975) indicate that several Cretaceous
genera continue through the boundary and occur as late as the Eocene in North America
and Europe. There were normal evolutionary changes through this horizon. If one
looks at numbers and at just Aqutlapollenites, a good case could be made for mass
extinctions. If one looks at the entire flora, including angiosperms, gymnosperms,
sporophytes, freshwater plants, and dinoflagellates, the extinctions do not appear so
great. In Saskatchewan with a sampling interval of 5 to 15 cm, many of the Cretaceous _

forms seem to survive (Jarzen 1977A). The sediments cannot have been reworked,

because one of the forms that almost always occurs in all the samples is a freshwater

fern called Azgolla. A massula or body in Azgolla encloses the spores and this very

delicate structure would be broken by reworking. A degradation of the preservational

quality of the pollen, for reasons unknown, does occur, but not mass extinctions.

McLean: Would you comment on the abundance of Aquilapollenites before the Cretaceous-

Tertiary boundary.

Jarzen: Aquilapollenttes probably reaches a peak in species diversity during Campanian

time, then begins a decline which stabilizes through the lower and middle Maastrichtian,

Si

Hsu:

to show another decline at the end of the Maastrichtian.

What is the ecological significance of the two groups of plants which did suffer a

higher proportion of extinctions?

Jarzen: Agutlapollenttes was related to the families Santalaceae and Loranthaceae, which

are very common today in the tropics where they are parasitic on other angiosperms
(Jarzen 1977B). The Wormapolles group contains lines of many botanical affinities.
They are loosely united into a morphologic group because they possess a particular pore

structure.

Dilcher: We are speaking about a group, the angiosperms or flowering plants, which have

Hsu:

their most probable origin in the lower Cretaceous. They became progressively more
important in the Cretaceous and their modernization occurred in the Tertiary. The
evolutionary theme is one of continuity, unlike in the case of the dinosaurs. The
major groups continue. Species level studies, such as those of Krassilov (1978) on
gymnosperms, show evidence of extinctions at the Cretaceous-Tertiary boundary. One
group, the cycadeoides, became extinct earlier. The flowering plants carry across time
boundaries and there seem to be no breaks in their record. The change that is present
on a species level could be a manifestation of the regression of the epicontinental

sea in mid-North America, and a mixing of the components of previously distinct floral
provinces to the east and west. As an aside, I remember that paleobotanists once
denied that fossil plants showed any evidence of continental drift. When plate
tectonics became geophysically understandable a few years later, the same paleobotanists
demonstrated how paleobotanical evidence supported plate tectonic theory. Are we ina
similar situation, where if a good mechanism were provided, data could be found to
support it? We can now explain the paleobotanic evidence in various ways, but there

is no evidence that demands a catastrophe as an explanation.

Why should the northern temperate plants have suffered more than those in the

tropics?

Dilcher: Hickey has suggested, in an unpublished manuscript, that a global cooling would

52

affect boreal regions before it would low latitudes.

Feldman: An atmospheric CO» increase by a factor of five has been modelled by Manabe and
Wetherald (1980), who predict a temperature increase of about 3 to 12°C. This would
be accompanied by a decline in rainfall in the 35° to 40° northern latitude zone and

the melting of the Arctic but not the Antartic ice cap.

Hsu: There were no ice caps in existence at the end of the Cretaceous. Because the
tropics are already saturated with respect to the greenhouse effect, a CO» increase
would produce a warming in high latitudes but not in the tropics. Hickey's
observation is in conformity with this prediction. Incidentally, I would like to point
out that the overall Cenozoic cooling trend should not be confused with the thermal

perturbations across the Cretaceous-Tertiary boundary.

Dilcher: The structure and size of leaves generally reflect the climate under which the
trees grow. During the middle Cretaceous, leaf sizes are what would be expected in a
warm deciduous forest. There seems to be a cooling in the Cretaceous, a warming again
in the Tertiary until the middle to late Eocene, and then a cooling. Some have
suggested that mid-Cretaceous angiosperms mark the advent of deciduousness of
flowering plants that had already been in existence for several millions of years.

They interpret the leaf record as evidence of the deciduous habit.

Russell: Returning to the marine record, could we hear a résumé of the high resolution

stratigraphy at the Caravaca section in southern Spain?

Smit: During the last five years we have attempted to integrate all the data on

lithology, paleontology, trace elements, and stable isotopes in a section which is
very well constrained in terms of time (Fig. 4, see also Smit 1981). The last
planktic foraminiferal biozone (the Abathomphalus mayaroensts Zone) of the Cretaceous
is about 110 meters thick and represents one to two million years. Sedimentation rates
of up to 11 centimeters per thousand years are thus indicated for a Cretaceous marl
which is composed essentially of coccoliths and planktic foraminifera, and contains
only 1 to 2% material of benthic foraminifera and ostrocods.

The marls contain a clastic component of about 20% detrital clay, and planktic
or benthic foraminifera can easily be removed and studied three-dimensionally under a

scanning electron microscope. Counts of 200 specimens each were made through the

53

v= rs mn
BR Sb gb
Be 3 285
se 8 S55
Be @ EES
ia 3 320
GUBBIO CARAVACA
( Premoli Silva1977)
A Bic DIE F G H J K L M

2 n
CE
3 BE
$ ge =
= Myr Myr. 35 ES
s -6212 623 eo |aa 3
5 SE —
| £ w > Q
© LC]
5 es e/g £
= = £ © 5
3 S | = s 5
© @ a4 : 3 5
5 l seers EL bs 8
= Q € 2 © a ©
= 3 3S * = = À 3 2
Ss 3 2 5 & ÈS ü 5
o = 2 = © o
ey (Ls mean te £
28 Py = S 68 sue o g
© = NT Es È =
vv GS s = 2 LS
8 3 3 ¢ fs && act À
® È 2 3 o Sie Gens bulloides | $
3 s 3 Somes Re FE $
3 3 Bi || Be 5 gescs a 3
S i = A Se 8
3 |29 6413 6434 633 ES Sie Eos Es È s
a g82 253523 8
v 509$ stabs g
3 DÉS FESREs
ë SxS SK@EZG lecretacea|
re i
TERTIARY K'eug Ba A maya-
oS [29 ” roensis
CRETACEOUS 5 à estes À 8 !
c-m554 6537 "1.7 S à 5 ÿSs Ses
i [ae è stee,33 3
! E, §sae 8 à
1 D 2ae5 2E €
1 ge SSTa 48 È
ce SEsl se =
! Be S688 sss
3
so
°
30 r+ $ 2
2 $ B | 3
2 5 5 re
& > 20m > 2
° oi ê € [
re = €
S 10 x !
> <
E |
< 0
-6623 66.76 Le2e8 5 532
/ = Zz
27 all 6684 188.4 3 wee
1 LE
31 F, + Le}
2 , 55° IT

FIGURE 4. Established magneto- and biozonations of the Barranco del Gredero section,
compared with bio- and magnetozonation of Gubbio (Alvarez et al., 1977).

The stratigraphie range of important planktie spectes ts indicated.

Linear scale is used for the geomagnetie polarity colums of Gubbio and
Caravaca. Base anomaly F,+ and the top of anomaly Ls+ are placed at the
same level.

A) Biozonation of the Gubbio section after Premolt Silva (1977).
B) Sea floor geomagnetic anomaly.

C) Reversal pattern of the Gubbto section (Alvarez et al. 1977).
D) Gubbio magnetozones (tbid.).

E) Reversal pattern of the Gredero section (G. Brunsmann, pers. comm.).
F) Numerical ages tn millions of years.

G) Biozonation of the Gredero section.

H) Distance tn meters from the K-T boundary.

J) Biozonation after v. Hillebrandt (1974).

K) Nannofosstl zonation (Rometn 1979).

L) Ranges of selected planktte foraminifera.

M) Biozonation in detail around the K-T boundary.

54

section and virtually no change was noted until the last millimeter of the Cretaceous
at the top of the zone. Only one species, Globotruncana ganssert, disappeared earlier
from the samples, and it was probably a rare species in the remainder of the section.
Then, of the 55 species of planktic foraminifera, 50 disappear within this topmost
millimeter. Only those species which are very rare in the smaller fractions of the
Cretaceous samples are relatively abundant in the boundary clay layer. If it were a
question of reworking only smaller species into the boundary clay, one would expect to
see relative abundances similar to those in the smaller Cretaceous fraction, which is
definitely not the case. On top of the 10 cm clay layer occur the first Paleocene
species. None of these five to seven species are in fact represented in the smaller
fraction of the upper Cretaceous biota. The lowest 40 cm of the Paleocene marl is
facially identical to the Cretaceous marl, and also yields a planktic-benthic ratio
of about 0.9 - 0.98 as well as a comparable clay content of about 18.5%. Facies and
lithology are thus exactly the same below and above the boundary clay. The oldest

or "Globtgerina" eugubina biozone of the Tertiary is 40 cm thick. Coccolith high-
resolution biostratigraphy is less precise because of the possibility of reworking.
Seventy meters for the first four Paleocene biozones versus 110 meters for the latest
Cretaceous biozone probably represents a significant drop in sedimentation rate.

The normal polarity interval 30 was identified, and 16.5 meters of Anomaly 29
negative which extends 11.5 meters into the top of the Cretaceous and 5 m into the
Paleocene, well into the Globigerina pseudobulloides zone. Thus about 11.5 and 5
meters were deposited in about 480 thousand years, or the duration of 29 negative, as
was already pointed out by Dr. Hsu. These 16.5 m, yielding sedimentation rates
extrapolated linearly according to the durations of the magnetostratigraphic and
biostratigraphic subdivisions, already account for about 500 thousand years. There
is thus no evidence of a hiatus in deposition. All of the Globotrunecana and
associated forams become extinct at a very sharply defined level. In the boundary
clay and in the eugubina zone only one species (Guembelitria aff. cretacea) remains
from the Cretaceous assemblage, a species which occurs today in the Persian Gulf and
which was probably living during uppermost Jurrasic time. It is a very hardy,
triserial species and as far as I know is the only planktonic foraminifer to survive

the extinction interval. This species, which has an abundance of less than one per

55

56

mille in the upper Cretaceous, becins to bloom in the upper part of the boundary clay
layer and reaches its peak at the base of the true Paleocene. This peak is followed by
a bloom of the earliest eugubina, then the bloom of another planktic species and so on.
These nearly monospecific blooms of several species, one after another, occur over an
interval of about 15 thousand years, or the estimated duration of the eugubina zone.
The first "stable" Globigerina species, G. pseudobullotdes, then appears. To
recapitulate, a sudden extinction event is followed by a clay layer in which some
survivors have been preserved. If all of the clay is detrital, an interval of
5 thousand to 10 thousand years is indicated. Then, a rapid succession of new
opportunistic species occurs, which are very small and thin shelled. They are
succeeded by a stable fauna which is more or less still living today.

The sediments were analyzed chemically to detect trace element changes (Fig. 5).
The upper Cretaceous pattern remains stable to a point 30 millimeters below the clay
layer. Nickel, chromium, antimony, selenium, arsenic, cobalt, iridium and osmium
abundances suddenly increase at the base of the clay layer, coinciding with a
significant decrease of all of the rare earth elements. This peak occurs in the basal
millimeter of the 10 centimeter boundary clay layer. Most of these elements decline
within a half centimeter to normal Paleocene levels. The iridium peak is very sharp
but does not decline so abruptly upwards. Perhaps the trail of the iridium is due to
secondary effects and the deposition of reworked material in the upper part of the clay.
The rare earth elements correlate well with the insoluble residue content throughout the
entire section. Only in the basal millimeters of the clay can a significant decrease
of the rare earth elements be detected. We cannot account for the arsenic anomaly,
which may be associated with peculiar environmental conditions analogous to those
promoting the formation of pyrite. Perhaps 900 PPM of arsenic would be sufficient to
constitute a biologic hazard. All of these data are based on whole rock analyses,
recalculated on a carbonate free base, without chemical purification. Carbon 12/13
and oxygen 18/16 isotope ratios were drawn from coccolith platelets, with the benthic
debris removed. Cretaceous values prevail to a level one half centimeter above the
iridium peak, where a shift occurs toward the lighter isotopes both of carbon and
oxygen. A return to former values occurs in the remainder of the clay layer. The
total oxygen excursion is on the order of 2 parts per mille, representing an increase

of about 8°C.

en (So) Ni

FIGURE 5. Distribution of 'meteoritic' elements Cr, Co, Ni and Ir, and the REE
across the K-T boundary tn the Barranco del Gredero section. All
plotted on a carbonate free base, the REE on a logarithmic scale.

A = distance from K-T boundary tn em, B = ltthology, C = sample
number, > = well-preserved Zoophycos burrows, $ = tectonic

distortion (slicken sides).

57

Sanidine spherules measuring 0.2 - 0.9 mm are found only in the basal millimeter
of the clay layer, associated with the iridium anomaly, and were probably rained in
suddenly. They contain inclusions of nickel and chromium which have abundances of
about 10,000 PPM, and thus metallic meteoritic particles may also be present within

them.

McLean: If Globigerina eugubina first occurs at the top of the clay layer, why haven't

you drawn the Cretaceous-Tertiary boundary at this level?

Smit: This is a matter of paleontological convention. Strictly speaking, a boundary

should be drawn at the first occurrence of new species, so the Tertiary should begin
at the top of the boundary clay. However, the Cretaceous ends at the base of the

boundary clay.

Russell: We have explored some of the paleontological attributes of one mass extinction.

Several other, comparable extinctions are known in the stratigraphic record, and the

timing of those events has been a matter for some speculation.

Tucker: When one considers possible mechanisms for mass extinctions, whether they be

58

crater impacts or supernovas, the probability of the occurrence of a mechanism must
correspond to the frequency of extinctions in the record. We noted the occurrence of
extinctions beginning 542 million years ago, another at 500, another at 435, another
at 355, another at 230 and the one we have been discussing at 65 million years ago.
The interval between the extinctions is increasing regularly, commensurate with a
factor of 21 million years. The elapsed time between the oldest and subsequent
extinctions are two times, five times, nine times, 15 times and 23 times 21

million years,within about + 10%. The interval between one extinction and another
plotted as a function of time describes a regular linear curve with the interval
between extinctions increasing proportional to elapsed time. This tends to indicate
that there is some overlying regularity. The Sun moves around the galaxy and
encounters close spiral density waves about every 100 million years. It also bobs up
and down out of the galactic plane with a period of between 40 and 80 million years.
These time-scales do not resemble the intervals between the extinctions. Neither does

the pattern resemble that of random occurrences of supernova outbursts, or random

impacts from cometary or Apollo object populations. If the time between extinctions

is converted into distances between extinction-producing encounters, the pattern is
reminiscent of the distance between planets in the Solar System. The solar nebula

must have consisted of rings that were increasingly separated as distance from the

Sun increased. Dust and gas, unlike planets, will spiral in under the Porting-Robinson
effect. The passage of these rings across the orbit of the Earth might then occur with
a regularity resembling the occurrence of the extinctions. The Earth would make a few
thousand orbits through the ring during its passage, and the time-scale for an
extinction could not be less than a few thousand years. Within any astrophysical
system, clouds do tend to break up into rings. The material could have been left over
from the primordial nebula or acquired through a passage of the Solar System through
an interstellar cloud. The latter is not a very probable event, but could be expected
to happen every 300 to 500 million years. During passage through such a cloud,
material cannot fall onto the Sun if it has any angular momentum. It must spiral

down into a disc and then the disc can accrete onto the Sun in the plane of the Sun's
rotation, which coincides with the orbital planes of the planets. In the case of
either primary or captured material, the cloud would be located at about a tenth of a
light year from the Sun. This is about the area from which comets are derived. The
relationship between extinctions and time would imply that the next passage will not
be for another 140 million years or so. In this model the clouds are diffuse, the
densities just enough to yield an optical thickness of about one. The matter is
probably accreted on Earth in the form of interstellar grains, which would be heated to

perhaps 10 or 20 thousand degrees and vapourized.

L. Alveraz: From a statistical point of view these numbers, 65, 230, 355, 435, 500 and
542 million years, plot on a straight line, with no indication of a quadratic

distribution. There are insufficient numbers to demonstrate non-linearity.

Tucker: The interval between extinctions seems to be increasing linearly with time. This
indicates an underlying regularity. Unfortunately, the statistics are based on very

few numbers.

L. Alvarez: In the case of an asteroid, a collision could be expected, on the average,

every 100 million years with an object about 10 kilometers in diameter. This can

59

be deduced either from the cratering record on the Earth or Moon, on the one hand, or
from the number of Apollo asteroids in space. This is consistent with the numbers that

you have presented.

McLean: Approximately 10 thousand years ago, there were great mammalian extinctions in the
Northern Hemisphere, and 100 million years ago, coinciding with massive submarine
volcanism, there was a truncation of ranges of foraminifera in the oceans. The
so-called Permo-Triassic extinctions may be an illusion because of gaps in the sections

simply due to the destruction of record.

Russell: The causes of one mass extinction may not be the same as those of another. The
dates to which Dr. Tucker referred are based on the assumption, which may or may not
be true, that the Cretaceous-Tertiary, Permo-Triassic, late Devonian, terminal
Ordovician, basal Ordovician, and end of lower Cambrian extinctions are the result of
the same kind of processes. Colleagues who are familiar with Paleozoic extinctions
(T.E. Bolton, W.H. Fritz, R. Ludvigsen, personal communications) inform me that the

extinctions cited appear to be similar in scope and effect.
Dilcher: How would one discriminate between the effects of clouds versus those of asteroids?

L. Alvarez: Dr. Smit has noted that the extinctions occurred within a one millimeter layer
that took about 50 years to be deposited. No one has suggested that enough material

could be accreted from a cloud in 50 years.

Feldman: That is correct. The densest kinds of clouds in the galaxy are similar to Lynds
134N, with central-core number densities of about 10° hydrogen molecules cm-*. The mean
free path of gas in these clouds is larger than the accretion radius parameter defined by
Bondi and Hoyle (1944). The problem then is a classic one of accretion onto an impact
parameter of the Earth's radius. The matter density in clouds of this kind is about 100
parts of gas to one part dust. It is possible to accrete 10 ngm cm-? of iridium from
dust grains with chondritic elemental abundances in passing through such a dense inter-

stellar cloud which is approximately one parsec (3.26 light-years) in diameter.

The transit time would be about 100 thousand to perhaps 8 thousand years,

60

corresponding to a random relative velocity of ~ 10 km oo in the former case, and

to an overtaking relative velocity (with respect to a spiral arm) of ~ 125 km s7) in
the latter case. A density enhancement of the kind suggested by Tucker, resulting from
the Sun's increased accretion of material, could occur and contribute to the |
production of the iridium enhancements observed. However, a substantial interval of
time would still be required, which seems incompatible with the stratigraphic data
presented by Smit.

A major part of the work I do is on the radio astronomy of interstellar molecules
in these dark clouds, the principal constituents of which are CO, HCN, its isomer HNC,
and other highly toxic molecular species. The injurious characteristics of these
gases are well known. On applying the mass fractions for these gases with respect to
molecular hydrogen, and scaling the hydrogen to what would have to be present as dust
to produce the iridium enhancement, cyanide and iso-cyanide concentrations in the
atmosphere would be about four orders of magnitude lower than those required for a
toxic dose. Concentrations of CO would be too low by two orders of magnitude. The
possibility of dense interstellar clouds poisoning the biosphere thus seems very

slight. I might be inclined to change my mind if the time scale for the enhancements

begins to approach a thousand years.

Dilcher: If a catastrophic model is considered, then steady-state conditions no longer
apply, at least during the event. How were the estimates of five thousand years and

50 years, with respect to Cretaceous-Tertiary boundary events, calculated?

Smit: The boundary clay is in my opinion detrital, and I assumed a more or less constant
influx of hemipelagic clay in this interval. The lowest (Paleocene) sedimentation rate
was applied to the clay layer, yielding a maximum figure of 15 thousand years and a more
reasonable figure of five to six thousand years. If the hemipelagic component is
deposited at the normal Cretaceous rate, only two thousand to two thousand five hundred
years would be required. The major accretion event is preserved in one millimeter of the
whole 10-centimeter-thick clay layer. Any increase in sedimentation of the detrital
component will only reduce the time for deposition of the entire clay layer. Another
approach has been made through paleomagnetic evidence. The impact event coincides

with only the base of the clay layer in Caravaca and in Denmark. In the Danish clay,

61

after removing the pyrite, the remaining clay fraction in the basal millimeter
contained the most depleted rare earth signals. I think my figure of 50 years, or

at the most 250 years, describes the maximum duration of the event accurately.

Dilcher: This record is from marine deposits under three thousand meters of ocean water?

Smit: They were fully marine, fully pelagic environments. Spain at that time was well

out in the Atlantic Ocean, not within a restricted Tethyan environment.

Hsu: The Spanish deposits are on top of a continental crust; considerations of isostatsy
suggest that the sea was not thousands of meters deep there. The Danish section is

usually considered to have accumulated at water depths of 500 meters or less.

Smit: The enhancements so far have been identified only in pelagic sections with the

exception of the Danish section and a continental section in New Mexico.

McLean: According to some of the definitions of the top of the Cretaceous and of the
bottom of the Tertiary, or base of the Globigerina eugubina zone, the boundary clay

does not fit into any available time era.

Hsu: In the case I was describing, the first appearance of Tertiary nannofossils in that
section was below the last appearance of the Cretaceous forms. Thus the "top" of the
Cretaceous was above the "bottom" of the Tertiary. But criteria must be selected,
and in this instance the bottom of Tertiary has been defined conventionally by a first

appearance of Tertiary taxa. Implicitly, this also defines the top of the Cretaceous.

Smit: Was the clay layer missing in your section?

Hsu: The clay layer was present, but from our definition the clay is the top of the
Cretaceous. Where to draw the boundary is an arbitrary convention. In fact, according
to our definition, the event indicated by the iridium spike occurred during the last
few hundred years of the Cretaceous. Perhaps the iridium anomaly will prove to be

the best definition of the stratigraphic boundary.

Feldman: Can you explain why the iridium "tails off" through that clay, and the other

enhancements do not?

62

Smit

Hsu:

: It is most likely a question of the nobility of the metal iridium. Iridium was
the only platinum metal we analyzed through the section. There were two effects, a
direct influx from the dust clouds and a secondary influx from the continents where
iridium was also deposited. In integrating whole peaks, two effects are combined.
Like Dr. Hsu, I prefer to plot only the maximum concentration in any particular
section. The clay layer in Caravaca is laminated and there is no evidence of
bioturbational mixing. In the transport from the continents through the oceans,

oceanic residence times for nickel and cobalt are probably longer than for iridium.

Bioturbation of bottom sediments suddenly ceases at the horizon of mass-mortality
at D.S.D.P. site 524. The laminated sediments above the boundary clay suggest that
the biomass of benthic forams was inadequate to disturb the sediments to the same

degree that it had previously.

Smit: Bioturbation does not turn over clays near the Cretaceous-Tertiary boundary, for

otherwise sharp enhancement phenomena would be destroyed.

L. Alvarez: This would seem to indicate there was a very sudden killing, affecting

benthonic organisms at the same time that the iridium was deposited. The benthonic

organisms must have been absent for a significant span of time.

Smit: It is the only evidence available for extinctions of deep-water organisms, for

the fossil record does not show significant extinctions of bottom-dwelling organisms.

Kyte: With regard to Dr. Smit's comment on secondary influx from the continents above

Hsu:

the initial millimeter, if one millimeter represents the amount of material which is
deposited by the accretionary event at Caravaca, and everything else is detrital
material that was washed in, then 1% or 2% of the iridium in the entire horizon was
deposited by that event. About 50 times more iridium must be derived from other
places around the world. If a similar record occurs over 30% of the Earth's surface,

then a mass balance problem is created.

How long can the iridium dust stay in suspension?

Kyte: It depends on where it is coming from. If it was generated by an impact even I

would suspect that it would fall out in a few weeks to a few months.

63

Ls

Alvarez: It cannot be transported world-wide through the atmosphere in a few weeks. A
world-wide distribution indicates that it stayed up more than a few weeks, unless the
major transport was by ballistic trajectories, as tektites are believed to be

distributed.

Jaworski: If the material is in the form of a hot vapour, when it contacts an aqueous

phase it would most probably start to dissolve. This could allow for a considerable
length of time before it was finally deposited. If it entered the biosphere, it
would appear in the sedimentary record only with the eventual deposition of the

organic material. This could produce the tailing effect seen in the geologic record.

Alvarez: Can the mechanism that you are describing distribute the material world-

wide?

Jaworski: No. Once it has been distributed through the atmosphere it would fall and

come into contact with water. The original global transport would be through the

atmosphere.

Alvarez: In the present atmosphere, which is not nearly as turbulent as the atmosphere
would be after an impact event, it typically takes one year to transport material from
the Northern Hemisphere to the Southern. That was discovered from the distribution

of carbon 14 that was generated in the extensive Soviet bomb tests.

McLean: Considering the slow sinking time of most clay particles in the oceans, marine

circulation would have to be a distributing agent as well.

Thierstein: That depends on what the transport mechanism is. Clay particles that settle

64

in the deep ocean must be transported down through fecal pellet transport at high
velocities, because the settling time of a two micrometer glass sphere would be on the
order of 50 years. Lateral current in the deep ocean would therefore homogeneously
distribute all the coccoliths and the clay minerals throughout the world ocean. The
average grain-size of dust particles after the Mount Agung explosion in 1963 was

about two micrometers at 20 to 30 km in the stratosphere (Mossop 1964). Junge and
others (1961) calculated the settling-rates of particles based on measured grain-size

versus altitude distributions of aerosol particles. The relationship is not linear

with respect to altitude in the stratosphere. At an altitude of 10 kilometers a two
micrometer particle sinks at a velocity of 4 x 10 ? cm s l. At 20 kilometers
altitude the sinking velocity is 6 x 10 2 cm s 4, or 1.5 kilometers per month. There
is no allowance for coagulation processes which might result because of high particle
densities, and the velocities were calculated according to Stoke's Law. If the
average elevation were about 15 kilometers, within ten months virtually all of the
particles would have settled below 10 kilometers where they would be rained out within

a matter of days. It appears to be impossible to keep considerable concentrations of

dust in suspension over extended periods of time at these altitudes.

L. Alvarez: The velocities that you mention are very small compared to the turbulent
vertical velocities, which would then be the controlling velocities. The question
should be referred to a specialist, but the principle applies to the ocean where
particles are held in suspension by currents in spite of the settling effects of

Stoke's Law.

Béland: I would like to present a few thoughts in an ecological mood. One is that
ecosystems react in the same way to any drastic stress; they become simpler. For
example, the effects of overgrazing by elephants in Africa and the effects of exposing
a forest in North America to radiation are similar in that a degradation in the
structure of the ecosystem takes place. An observer might not be able to tell whether
too many elephants were present or a scientist was trying to prove some point.

Another thought is that a list of generic extinctions is less meaningful than a list
of species extinctions. Another thought is that once a planet has caught that disease
that we call life, it is very difficult to cure it. Another thought is that the
catastrophe cannot have been too drastic because the new species that appear in the
Tertiary were derived from species that survived even though they have not been
detected in the fossil record. A catastrophic event was followed by a diversification
event. An event that would cut off light sufficiently to reduce phytoplankton
production would have to block a very high percentage of light. That leads me to the
last thought, which is that production should be considered, not biomass. Other

links in the food-chain depend on phytoplankton production, not biomass.

65

66

What would happen to phytoplankton if light were severely attenuated? If it
were completely cut off for a very long period of time, I think everything would
have died, and this did not happen. If light-reduction were drastic enough, the
resulting instantarieous stoppage of production would ensure that the remaining cells
would be consumed by herbivores so rapidly that, in geological terms, an instantaneous
event would occur. Several species in several groups, such as diatoms, thallophytes
and xanthophytes, would survive through in encystment. This is common in temperate
and high latitudes. How long the cysts would remain viable is not known, but it is
probablyon the order of two to four years. The stimulus that causes these cells to
return into production may not be the return of light but the quality of the water or
the temperature.

The timing of the event could be important. For example, the deleterious
effects would be immediately felt in the tropics whatever the time of the year. In
the high latitudes, if the event were to occur today in the fall, it would probably
not matter at all because production normally ceases in the fall. During the
Cretaceous the differences between high latitudes and the tropics were not as great
as now, but differences between the patterns of production in high latitudes and low
latitudes must have existed. The decay of cells would probably produce an increase in
bacterial production. Whether this production could be used by other organisms down
the food-chain is a matter of speculation. Single bacteria would be too small and
groups of bacteria clustered around dead cells would be too large to be filtered out
by specialized zooplankton carnivores, and most of these animals would die. Many
zooplankters, including fish larvae, rely on light to catch their prey. They are
unable to feed in the dark even if food is available. Some zooplankters could
survive by effectively encysting. For example, cladocerans have both sexual and
parthenogenetic modes of reproduction which they adopt according to prevailing
conditions. They also produce durable cyst-like eggs which can survive for months
and perhaps for years. It is possible that some ostracods and copepods also produce
durable eggs. Further down the food-chain, many pelagic fish feed on zooplankton.
Herring and mackerel depend heavily on the bloom in phytoplankton during May and June.
They feed heavily, and larvae are recruited into the adult population. Ifa

catastrophe happened in the spring, the effects would be more drastic than in the fall.

Perhaps surprisingly, the larger predators, such as sharks and other large fish,

would survive longer, because they could feed on decaying animals or temporarily
become dormant. Most of these animals are ectothermic, and in colder water their
metabolism would be further depressed. Under such conditions they could survive much.
longer than terrestrial mammals, for example. The turnover-rate of food in deep-sea
bottom-dwelling organisms is very slow and ultimately depends on surface productivity,
but with a time lag. There is even a possibility that the rain of dead matter from
the surface would sustain the deep-sea benthos for a substantial period of time.

Marine ecosystems could probably survive pretty well if the blockage of light
lasted from six to eight months, or perhaps a year. They would probably not survive a
blockage of two to two and one-half years. McKay and Milne (1980) estimate that
phytoplankton cells would die after two days and zooplankton after about 30 days. The
latter may be an overestimate, for most zooplankton would not survive more than a
couple of days unless the event happened to coincide with the autumn in high latitudes.
A survival of 15 days for small fish is reasonable, as is a survival of 150 to 200 days
for large fish. Blocking the light for longer than about eight months would make it

very difficult for anything to survive into Tertiary time.

Hsu: Why should any phytoplankton survive at all if the blockage lasted more than two

days?

Béland: About two days would be required for the production of phytoplankton to completely
stop. Individual cells tend to sink continuously and are brought to the surface by
currents where they would probably rapidly be eaten. Most cells would not survive

longer than a matter of days.

Asaro: In a private communication to Dr. Milne, Dr. Bernard M. Oliver stated that there
would be tails on these distributions. Dr. Milne replied that he agreed, and the
question is how low does the density have to become before the population cannot renew

itself.

Béland: Some species could survive through encystment.

Brasch: One can envisage a rapid and total darkening halting all photosynthetic activity,

which is then followed by a rapid clearing. An extinction of some species occurs, but

67

not others that can survive seasonally adverse conditions. One can also envisage
gradual darkening, producing a slow reduction in primary food production, followed by
a gradual clearing. In this case some species might be able to adjust to the changing
conditions. The link between photoperiodism and reproductive cycles must not be
overlooked. If normal photoperiodic changes were interrupted for a period of two or
three years, survival might be jeopardized not by a lack of food but by an inability
to reproduce. The biosphere is buffered against a number of possible catastrophic
scenarios, but only within certain time constraints, which are critical and must be

defined.

Béland: Many zooplankters in high latitudes are very well synchronized to reproduce in
the spring. This would be the last thing to do in the absence of phytoplankton

productivity, but the organisms would probably be unable to switch off that pattern.

Theirstein: In experiments in culture it has been shown that one of the dominant living
coccolithophorids, Emiltanta huxleyt can survive by feeding on dissolved organic
matter for longer than four months in complete darkness (Blankley 1971). If the
concentration of dissolved organic matter is low, a minimum illumination of about 400
lux, or 0.3% of the flux incident at the surface of the ocean is needed. This is
equivalent to light intensities at a depth of 150 meters in clear ocean water.
Coccolithophorids are known to live and survive at depths of 200 meters, so that one
would have to assume 99% or more of the light was blocked for an extended period of

time to produce phytoplankton extinctions.

Béland: Although individual cells might survive up to a year in total darkness, the
effects of the blockage of production at the surface on the food chain would be so

drastic that it is difficult to believe that blockage of production lasted this long.

Dilcher: What is the record of fish kills at this time?

Russell: According to Danilchenko (1978) the number of recorded fish genera declined from

38 to 10 from terminal Cretaceous (Maastrichtian) to basal Tertiary (Danian) time.

Thierstein: According to Dr. Patricia Doyle (personal communication), the record of the

teeth of bathypelagic fishes shows a comparable evolutionary turnover, although the

68

resolution is only within + 10 million years.

Kyte: The resolution may be somewhat better than that. Doyle (1980) was able to locate
the Cretaceous-Tertiary boundary within 20 centimeters and iridium was detected,
implying a resolution of about 100 thousand years. However, she had no confidence in
distinguishing Danian from Upper Maastrichtian time, because she could not find a

change in ichthyoliths across the boundary.

Hsu: Nautiloids and ammonites have a different record. Are there any physiological

reasons to think they occupied different habitats?

Thierstein: Only nautiloids can now be studied. Nautilus pompilius is known to dive as

deep as 500 meters (Stenzel 1957).

McLean: Some plant groups were in existence during Cretaceous-Tertiary transition time
that might have had very little storage tissue either in the sporophyte or in the
spores. Two of these groups that are also living today are the Equisitales and
Osmundales. They seem not to have suffered greatly in the transition, as far as I

know. A darkening lasting several years should have devastated those groups.

Pirozynski: Perhaps marine habitats are relatively well buffered against a catastrophe.
Land habitats are very vulnerable. Plants are growing systems with symbiotic fungi
and bacteria that absolutely depend on them for a constant supply of photosynthates.
They are also attacked by pests and diseases that operate within annual life cycles.
One would expect that obligate parasites to suffer very greatly if photosynthesis
stopped for more than a year. If all the parasites, symbionts and pests were removed

a complete floral turnover would take place.

Dilcher: Could poisoning have disrupted food chains more effectively than blockage of

photosynthesis?

L. Alvarez: How could the material be transported world-wide without causing a darkening?
Two available time-constants are one year to transport material between the Northern
and the Southern Hemispheres, and two and one-half years for Krakatoa material to
fall out of the stratosphere. Other than atmospheric transport, there seems to be no

other way for landing material on both Italy and New Zealand. When one centimeter of

clay is suspended in the air, in any particle size distribution, the result is

blackness.

Thierstein: In the case of small particles and aerosols, a lot of back-scattering of
light can occur. The effect is less in the case of coarse particles. Smog reduces
visibility by impairing the transmission of direct light, whereas the light incidence

is due to back-scattered light, which is not so effectively blocked.

L. Alvarez: In passing through the equivalent of a one centimeter clay layer composed of
one micron particles, the light intensity would not decline exponentially through
scattering; it would decline linearly. The extreme cases are exponential attenuation
by perfectly absorbing particles and linear attenuation by perfectly reflecting
particles. The true situation is obviously intermediate between these extremes, but
the suspended layer must certainly have resulted in very severe attenuation of the
sunlight. The light must pass through a certain number of grams of material per
square centimeter, and it makes no difference whether it is spread out over 100

kilometers or 10 centimeters.

Thierstein: If a glass of water containing suspended coccoliths is evaporated, light will
not penetrate a clay layer of coccoliths that is one-half millimeter thick. If the
same light is directed into a glass of water with the same amount of coccoliths
distributed in suspension through the glass, the bottom of the glass will shine white

because of back-scattering.

McLean: Atmospheric particles are not necessarily opaque because light can be transmitted

through petrologic thin-sections which are thicker than the particles.

L. Alvarez: Some light will come through to be sure, but, qualitatively, it would be too

dark for plants to grow.

Dilcher: It would be useful to document in a rigorous manner that photosynthesis was
depressed to 10% or 5% of former levels for a period of a full year. The difference
between extinction proportions in terrestrial and marine systems suggests that poisons
might have been more important than the total lack of light for photosynthesis.

Poisons could be removed by fresh water flowing off of terrestrial systems into marine

70

environments where they might accumulate and produce the most damage to life.

L. Alvarez: Specialists in chemical warfare often experience difficulties because the

poisons react with rocks, water or other materials.

Hsü: Hydrogen cyanide could have been injected by 1017 grams of impacting cometary material.
Part of it would be oxidized, and the remainder within an ice conglomerate would be
dissolved in the ocean following an oceanic impact. A poison concentrated in
surficial currents would not come in contact with either benthic or terrestrial
organisms. Experiments to define lethal concentrations in the seawater for

phytoplankton are needed.
Kyte: Nickel sulphide is also toxic to some organisms (Costa and Mollenhauer 1980).

Jaworski: If one particular lethal mechanism is operative, another one by inference is
probably operative as well. The high levels of arsenic, osmium and nickel in the
very thin boundary layer at Caravaca could easily be dissolved in a surface layer
of water in the oceans and be toxic. On land, unless the poisons were directly
inhaled, the soil would act as an ion-exchanger and absorb much of these toxic metals
before they would be taken up by the roots of plants. This is one way of
differentiating between terrestrial plant toxicity and phytoplankton toxicity. The
chemical form of the elements is also significant. Some are extremely insoluble,
but others, such as oxides and ions that form soluble chloride complexes, are soluble.
Another environmentally differentiating mechanism is that, in freshwater, the complexing
anions are hydroxide and carbonate, whereas in marine systems the complexing anions
are chloride ions. Thus many elements which would normally precipitate in freshwater,
would remain in solution in salt water. It would also be useful to obtain data on
the levels postulated for some of these elements in the impacting material in order

to estimate the resulting concentrations in the biosphere.

Feldman: The platinum group elements themselves are not necessarily nontoxic. Many of
the soluble compounds of ruthenium and osmium, such as the chlorides, amino chlorides
and volatile oxides, are highly toxic and can elicit physiologic responses at
concentrations as low as 10 2 gm/ml (National Academy of Sciences, 1977). Assuming

a distribution of 10 ngm cm 2 of Pt-group elements mixed into the oceans, the

71

resulting concentrations of toxins are four orders of magnitude lower than is required
to elicit a physiological response. Two orders of magnitude can be regained by
restricting consideration to the shallow waters along the continental margins.
However, unless bioconcentration of some of these compounds occurs, their effects

would not appear to be significant.

Pirozynski: One might consider that those organisms that survived had a greater

tolerance, because they survived.

Hsu: If the poisons were concentrated in surface currents, which are usually only a few

hundred meters deep, the missing two orders of magnitude might be found.

Feldman: I had already made that assumption. The concentrations would be, at best, 10 11

gm/ml, and TOM required, for a physiologic response.

Hsü: Surface currents occupy only a part of the surface of the ocean area. Perhaps the
poisons were spread into the tropics. Another order of magnitude can be gained by

restricting them to 10% of the ocean surface.

Béland: It should be noted that many bottom-dwelling marine organisms have a phase in
their life cycle which is planktonic. They would be vulnerable to a drop in
phytoplankton productivity. Freshwater organisms living at that time of the
extinctions are not well sampled. Only those terrestrial organisms living in
specific environments coinciding with areas of sedimentation can have a fossil record.
Conditions in the ocean are more homogeneous, as is the possibility of being

preserved in the fossil record.

Clube: Certain points in the approach Napier and I (1979) have adopted might be a source
of further discussion. A suspicion has perhaps arisen that the theory we developed
was an ad hoe astronomical response to the discovery of geochemical anomalies.
However, the theory was developed prior to these discoveries and we feel it has a
comparatively secure astronomical basis. Catastrophism is a direct consequence, and
the evidence that has been discussed here fits into this kind of theoretical frame-
work. To begin, a cratering record is preserved on the inner planets and many Apollo

(Earth-crossing) asteroids have been identified. A satisfactory general relationship

72

exists between the number of Apollo asteroids and the cratering-rate of the last
three billion years. This relationship was not recognized some 10 to 20 years
ago because the number of known Apollos was too low. Discoveries of Apollo asteroids
have increased in recent years, however, following the work of Shoemaker and his
colleagues (1979). Apollos impinge on planets and produce craters, but more frequently
they just miss planets. Close encounters tend to deplete the population, ejecting
them from the Solar System altogether; the time-scale is something like 30 million
years. However, the average cratering-rate has remained approximately constant over
three billion years. There must be an astronomical mechanism, therefore, for feeding
the Apollo asteroid population and keeping it in tune with the cratering rate. Where
do these Apollos come from? The asteroid belt was once considered a natural source
for the Apollo population, when it was believed to be smaller. But there are not many
perturbers available that extract the Apollo population from the asteroid belt
(Wetherill 1976). Thus, a growing consensus amongst astronomers, views comets as the
main source of the Apollo population. Essentially, Apollos derive from the short-
period comet population which in turn derives from the spherical distribution of long-
period comets, through interaction with Jupiter. With comets identified as the main
source, a slight problem still remains. There seems to be a current overabundance of
Apollos relative to the long-period comet flux. We conclude there must have been an
enhancement in the supply of comets approximately 10 million years ago, producing the
current population of Apollo asteroids. It has been known since the turn of this
century that the distribution of the perihelia of long-period comets is not isotropic;
there are indications of concentrations which are related to the motion of the Sun
through the solar surroundings and indeed vague signs of correlation with the Milky
Way. Yabushita (1979) asked the question, "How long ago would this sort of distortion
of the surroundings have been imposed upon the comet system?" He examined relaxation
arguments and suggested something significant happened in relation to the current
comet population, again about 10 million years ago. This led us to ask, is there
anything in the solar neighbourhood that would cause one to think that something could
have happened to the Solar System approximately 10 million years ago?

Over the last few decades astronomers have recognized with increasing confidence

the existence of spiral-arms in the solar neighbourhood. Another feature of our

13

74

FIGURE 6.

perseus

axis of arm

rotation
sagittarius

Gi arm

central
spheroid

galactic nucleus

Schematte illustration of Gould's Belt. According to the most
recent estimates, the Sun's distance from the Galactte nucleus
ts ~ 7 kiloparsecs. The Sun ts moving away from Gould's Belt

in the direction of Galactic rotation at ~ 25 km s7:.

galactic neighbourhood of very great significance is the presence of some material
which has been known for nearly a century as Gould's Belt (Fig. 6). It is a feature
containing young stars, large quantities of gas and dust, and all the things that we
normally associate with spiral-arms, but it is not in fact part of the spiral-arm
system. Many theoretical models are given in the literature as to what caused it,
but they are not relevant here; all we need to know is that the feature exists. It
is a couple of hundred parsecs from the Sun, and the Sun's motion relative to the
feature is on the order of 25 kilometers per second. These numbers are
very well determined,and it is possible to conclude that the Solar System passed
through Gould's Belt about 10 million years ago. A significant possibility exists
that the long-term comets were captured or perturbed as a result of going through
Gould's Belt. This is not in accord with the prevailing view that the comets in the
Solar System are primordial, and it indicated a need for further research.
Radioastronomers, in surveying the galactic plane over the last few years,
have become increasingly aware of the existence of giant molecular clouds. There
are many of them, they are very massive and they present to the galactic astronomer
an environment for discussing the Oort Cloud that is entirely different from the
one in which the Oort Cloud of comets was originally conceived. When the Solar System
and its Oort Cloud pass close to a giant molecular cloud the latter is disturbed by
tidal action and comets are dispersed into the surroundings. Thus, it is now
possible to demonstrate that the Oort Cloud,of necessity, cannot be primordial. If
the primordial Oort Cloud is no longer tenable, but comets are present, the
probability that they are captured is high. Once comets are captured, many of them
fall into the inner part of the Solar System where they can produce the craters and
the possibility of episodic catastrophism dictated by the rate at which they are
captured from the surroundings. If impacts are responsible for triggering many of
the phenomena registered by the geophysical record, the prediction of episodicity is of
fundamental importance and one that is provided by no other theory (Clube and Napier,
in press). It is clearly important to know whether a catastrophe, such as the one
we have been discussing, was caused by the impact of a comet or a meteorite, although
from some points of view, the distinction may not be significant. As the Solar System

passes a giant molecular cloud, comets would be captured, but what is already present

75

in the Solar System, such as the primordial family of meteorites, would also be
perturbed. These objects will be fed into the inner parts of the Solar System at

the same time as comets are captured. Thus, both families respond to interaction
with giant molecular clouds, and both families are capable of impacting and producing

craters. The ratio of comet to meteorite impacts is of importance in our theory.

W. Alvarez: If the Oort Cloud is not primordial, then some meteorite debris from the
Antarctic ice sheet should have originated at a different time than did the Solar
System. The isotopic ages are all similar, and must then be interpreted either as

coincidental or wrong.

Clube: This material is undoubtedly meteoritic and is basically of Solar System origin.

W. Alvarez: Would comet debris never be found on the Earth? Should it always be broken
up and dispersed? Some delicate, soft kinds of "meteorites" have certainly been

recovered from Antarctic ice.

Clube: I am aware of this, and it is clearly important to decide what compositional

aspects would distinguish comets from meteorites.

L. Alvarez: The spherules that were found near Tunguska were probably derived from a
comet. If that comet came into the Solar System 100 or 600 million
years ago its origin was different from normal Solar System material. The time
clocks (for example the uranium isotope ratios) would be set differently than the

ones in the normal meteorites. Such discrepancies have not been observed.

Kyte: Analyses of stratospheric dust (Brownlee et al. 1977, Fraundorf 1981) suggest
that comets are very similar to carbonaceous chondrites, which are all very similar

in terms of non-volatile bulk chemistry.

Halliday: These are ablation products. They were derived presumably from cometary
material, but the condensate, following the melting process in the atmosphere, could

be quite different (Krinov 1966).

L. Alvarez: If the Clube-Napier theory is correct, they should not show the 4.7 billion

year age of our Solar System.

76

Kyte: I don't think anyone has tried to date the spherules from Tunguska. Certainly
if a piece of a comet were analyzed, one should be able to prove whether or not it

was from the Solar System .

Halliday: Anders et al. (1973) has shown, although Wetherill (1971, 1979) does not
agree, that carbonaceous chondrites have spent a long time in the regolith of
asteroids because they were soaked by solar winds. A comet does not live long

enough in the inner Solar System to pick up much solar wind.

Feldman: Is there any possibility that enough cometary dust would be on the surface of

the Moon to show an age signal significantly different from four and half billion

years?

Halliday: The fraction of Moon-dust that is presumably meteoritic in origin is about

one to two percent. Most of it is considered to be cometary in origin (Anders

Ge (eth, Wyss

W. Alvarez: In addition to the differences in radiometric systematics, non-Solar System
material would show differences in heavy stable isotopes, such as iridium and osmium,

which may be more easily detected using accelerator techniques.

Feldman: It might be more relevant to examine the lighter materials. Radioastronomical

techniques can be used to determine 13C to l?C ratios. This ratio differs in the
galactic centre, the galactic disc and the Solar System. Radio observations of HCN

molecules could determine the ratio in Halley's Comet.

Kyte: There is still a possibility of a sample return from Halley's Comet during its

next transit (D.E. Brownlee, personal communication).

L. Alvarez: Experience with accelerators enhances one's appreciation of how difficult
it is to inject a particle from outside into a stable orbit. It cannot be
accomplished with static fields. Similarly, gravitational potential resembles a
hyperbolic hole and it is very difficult to inject something into that hole and make
it stay there; if it goes down it should come back out again. If capture is so

difficult, how can the Solar System capture comets so efficiently?

v7

78

Napier: Perhaps I can address this question, as well as whether or not the Oort Cloud is

a primordial object. The crucial point is the survivability of the primordial cloud
in the presence of tidal forces being exerted by encounters with the giant molecular
clouds. The molecular clouds are probably distributed along the spiral-arms of the
galaxy. They are certainly associated with dust, and dust is a known spiral-arm
tracer. The Solar System would not be able to soar over or under the clouds but
would have to pass between them in order to avoid an encounter. The large mass of
these clouds, in the range of a 100 thousand to a million solar masses, would
produce a strong gravitational focussing, and the actual target-area is rather larger
than the geometric target-area. Taking the uncertainties into account, one finds
that at intervals of about 100 to 200 million years the Solar System will pass
through a giant molecular cloud. Because of the substructure within a cloud it is
quite possible for the Solar System to pass completely through it without
encountering a dense nebula. Perhaps once within Phanerozoic time, a nebula with
densities of order 10+ or 10° gm cm ? has been encountered. I have carried out
some numerical modelling, and it turns out that the tidal-effect is very important in
the neighbourhood of a dense nebula (Fig. 7). About 33 thousand comet orbits have
been calculated, using standard models for the primordial Oort Cloud, and random
encounters with these nebulae. In the outer extremities of the cloud the comets
are easily swept out of the Solar System, but a dense, inner core may remain. If
the Sun passes within 10 parsecs, 5 to 10% of the comets in the long-period region
are lost. If the encounter is closer, the effect becomes more drastic, and it is
evident that even a single passage through a giant molecular cloud has a dramatic
effect upon the primordial Oort Cloud. The cometary orbits are disturbed through
three-body encounters, or a tidal effect, basically.

The substructures in the giant molecular clouds are in the size-range of a
few parsecs, while the average dimension of the cloud is 20 parsecs. The
substructures are well separated and discrete, and facilitate the disruption of a
primordial Oort Cloud. After 500 million years in one simulation, we found that we
had almost lost the outer regions of the Oort Cloud (Fig. 8). After two thousand
five hundred million years there was really nothing left. I think the conclusion

is that it is difficult to maintain the primordial Oort Cloud hypothesis with the

escape probability

FIGURE 7.

100

à

Ul
CO

6) 5 10
al10* AU)

Illustration of the damage done to a comet due to a single
encounter with a massive nebula. In thts case the nebula has mass
20,000 solar masses, asymptotic approach speed 5 km/sec and impact
parameters as shown (1 pe = 3.26 light years). "a" represents the
semi-major axes of the comet orbits (in a.u.) which were randomly
ortented and uniformly distributed in eccentricity. The hatched
area represents the region from which long period comets are

observed to arrive.

79

FIGURE 8.

80

2 3 4

log a

Cumulative effect of massive nebulae on a hypothetical comet cloud inttially
populuted uniformly in log a. Time is measured in billions of years. Any
primordial cloud (t = 5) would have long since been dissipated apart from a core
a few 10° a.u. radius. The long period comets thus orbit in an unstable region,
implying recent capture from interstellar space.

molecular-cloud system as it is currently envisaged.

It then seems advisable to examine how comets could be captured from the
interstellar medium. Several processes, from straight-forward sedimentation to Jeans'
collapse, could result in the formation of cometary bodies in giant molecular clouds.
The figures are crude and the uncertainties are large. Condensates from typical
clouds would contain a mean mass of several 10'’ grams, which corresponds well with the
estimates of Hughes and Daniels (1980) for the mean mass of long-period comets.

There seems to be a striking resemblance between the chemical composition of comets
and that of the interstellar medium. Thus, although growth mechanisms have not been
fully developed, there seems to be no great obstacle to this mode of cometary formation.
The absence of comets coming in on hyperbolic orbits places an upper limit to the
density of field-comets, and it would not be possible to capture field-comets at a
sufficient rate to produce a quasi-equilibrium Oort Cloud. They must be captured
from dense regions. Assuming about 5% of the mass of molecular clouds is in
kilometer-sized planetesimals, the capture rate can be estimated by using simple
sphere-of-influence arguments. The figure of 5% is not arbitrary. There is a long-
standing problem with depletion of C, N and O in the interstellar medium, which
Tinsley and Cameron (1974) and Greenburg (1974) have suggested might be locked up

in interstellar objects. With the 5% figure it was found that the entire Oort Cloud
could be captured in one passage through a giant molecular cloud. It may be an
overstatement, but the Oort Cloud might be viewed as a quasi-equilibrium phenomenon.

The hypothesis is vulnerable from a number of directions. For example, if the
small-body population is controlled by the Galaxy rather than by processes internal to
the Solar System, then time-scales appropriate to the Galaxy should be sought. One
consequence would be a constant cratering-rate, because the decay time of galactic
material is long. The rings that Dr. Tucker proposed earlier would be as subject to
instability as Oort Cloud comets, but if there is anything in the numerology, then one
should look for some sort of galactic explanation. Modulations in the cratering-rate
would be expected. Periods of high risk should follow periods of low risk, because of
decay times of materials in the Solar System. Discrepancies should be apparent between
the known Apollo population and what would be expected according to the cratering-rates.

Shoemaker et al. (1979) have found a statistically significant two to one increase in

81

the cratering-rate over the last 600 million years against that over the last three
billion years, and in turn there seems to be significantly more Earth-crossing
asteroids than is consistent with the more recent cratering-rate.

Also of interest would be anomalies in isotopic abundances. Iridium or osmium
isotope ratios may not be helpful because r-process environments are poorly understood;
so many supernovas have occurred and_so much mixing of ejecta has taken place over
long time-scales. Whether striking Solar System isotope ratios being detected
indicate the Cretaceous-Tertiary boundary material is truly indigenous to the Solar
System is not, according to D. Arnett (personal communication), a settled question.
However, I would definitely look for variation in the 12¢ to 13¢ ratio in strata
deposited in temporal proximity to mass extinctions. Finally, the theory predicts
impact episodes which may include 108 or 109 megaton impacts, and these should be
associated with three to six biological catastrophes within the Phanerozoic
corresponding to spiral-arm passages. McCrea (1981) has reviewed the episodicity

question in some detail.

Grieve: The sample of well-dated craters is too small to reveal adequately changes in

the cratering-rate. The uncertainty factor is about two during Phanerozoic time.
Napier: We suspect that there was an increase, actually.

Grieve: Curves can be drawn that show a factor of two increase. The lunar record covers
the period between four to three million years ago. This record must be fitted to the
Phanerozoic terrestrial record, making adjustments for the different gravities of
the two planets. Other uncertainties include the cosmic velocity of the objects
and how large the gravitational cross-section correction will be. One can draw a
line through the points and conclude that the cratering-rate has been constant for
the last three billion years. With the uncertainties attached, the data will also
support an increase by a factor of two perhaps quite recently in the Phanerozoic. What
is the composition of these extra-Solar System cometary bodies? Impact melts match
the relative siderophile abundance patterns of known meteorite groups. Some are
Cl, others are different types of chondrites, and some are A chondrites or irons. Do

these data fit your hypothesis?

Napier: This requires a statistical answer. There will be a mixture of bodies ranging
from bona fidé interstellar objects to primordial remnants. How many impact melts

have been analyzed?

Grieve: Melts from only 10 to 15 craters have been measured. Perhaps, another 50 craters
remain to be analyzed. The present sample is small. In some cases siderophile
enrichments were not detected in impact melts, for unknown reasons. Possibly the
impacting body left a signature that is very like that of terrestrial material. It

does not stand out as a geochemical anomaly.

Halliday: There are some very interesting aspects of your theory. In discussing the
Apollo objects you may be subscribing to a view which overemphasizes the cometary
component. Shoemaker and Wetherill are of one opinion, and I suspect they are on
the minority side in this case. The kind of exotic resonance phenonmena with Jupiter
that can bring material in from the asteroid belt seems to account for most of the
ordinary chondrites arriving in a time which is consistent with their cosmic ray
exposure ages. These are usually 20 to 30 million years, and in extreme cases, up to
60. Irons have survived longer in the inner Solar System and have greater exposure
ages. Extrapolating down to the small fireballs, which are all Apollo objects, we
think with increasing confidence that the material can be identified by how well it
penetrates the Earth's atmosphere. Three main classes are identified, which are
arriving in just about the abundance that was predicted 15 years ago. The
early results from the Prairie Network suggested the rate was much lower, but this
was due to an error which gave low densities for the projectiles. Ordinary
chondrites and irons are represented in the correct abundance. A second group seems
to represent the carbonaceous chondrites. They are much more abundant than the
proportion of carbonaceous chondrites in meteorite collections, because the atmosphere
is such a severe filter. A third group appears never to penetrate the atmosphere,
and these are presumably the cometary components. Many of the objects responsible
for terrestrial cratering are associated with definable meteorite categories.

Cometary impacts must also occur, but they are probably more difficult to identify.

83

Napier: A weak point in the work at this state of development is, how many objects are
bona fidé interstellar and how many are primordial? Dr. Pillinger at Cambridge is

hyd 13 . . . .
ENEO C ratio in Brownlee, or micron-sized

developing a technique for measuring the
particles. It is probable now that the Giotto Project, which is a Halley flyby, will

attempt to measure the isotope ratio.

L. Alvarez: This theory is attractive in many ways, but presents one area of difficulty.
The isotopic abundance of uranium can be measured without a mass spectrometer. Alpha
particle emission in the two different ranges provides a simple measure of the
relative abundances of uranium 238 and uranium 235. No uranium from a natural
source has ever shown a departure from the expected ratio with one exception. That
was a sample of uranium from Gabon, which was found to contain too little uranium
235 in it, leading to the discovery of a fantastic natural reactor that operated a
couple of billion years ago when more uranium 235 was present and a homogeneous
reactor could operate with natural water. If different natural "factories" make
uranium at different times and in different places in the galaxy, then the ratio of
uranium isotopes will also differ from region to region. Comets arriving from these

different regions of space should not all have a ratio typical of the Solar System.

Napier: One of the problems may be the speculative nature of current nuclear synthesis
theories. For example, Rees (1978) has suggested that supermassive, or very massive
objects,might generate heavy nuclei in a pregalactic phase: the material is then
already present and mixed. The time taken to recycle material in the Galaxy is
of the order of a third of the age of the Galaxy. If supernovas were the "factories"
and produceda certain isotope ratio which is taken to vary by a factor of two from
one supernova to another, then over one-third of the lifetime of the Galaxy
several 108 "factories" will produce and mix this material. It is possible then
that two divided by the square root of 108 would be the expected characteristic

variations in an average galactic locality.
L. Alvarez: Supernovas would then probably be eliminated as the source of uranium.

Tucker: The consensus was that uranium is made in supernovas, because no other suitable

source has been identified. Cowan and his associates (in press) now suggest that the

84

r-process occurs in the cores of red giants. Their populations are 10'° instead

of 108, so the fluctuations are further depressed. If the Solar System passes

through a cloud, whether comets are present or not, a very large amount of interstellar
material will be swept up. This would depend on the geometry and mechanics of the
accretion episode, but as much could be accreted as is observed in this boundary layer.
If a catastrophe were not caused by an impact event, could it still occur in terms of

shutting off the light?

Kyte: The Napier and Clube hypothesis is testable,as it should produce iridium anomalies
in the stratigraphic record. We plan in the near future to construct an iridium
profile for the Cenozoic, and should we find a sharp peak at about 10 million years,
we will let you know. This hypothesis predicts that we should find a sharp increase,
say two or three times, in the flux of cometary material, which should then tail off over
a period of several million years. It should be measurable, but an increase on the

order of the Cretaceous-Tertiary enrichments should not occur.

W. Alvarez: If the Apollo objects cannot be derived from the asteroid belt, and the

Oort Cloud is periodically overturned, where do the iron meteorites come from?

Napier: A different scenario may not be necessary for the iron meteorites; they could
be derived fromthe main-belt asteroids. The transfer rates are adequate for them,
but Wetherill (1978) finds that for kilometer-sized objects transfer rates fail by a

factor of 10 to 100 in deriving objects from the main belt.

Feldman: I want to point out that a large accretion of dust will probably not occur in
passing through a diffuse giant molecular cloud unless there is a direct passage
through one of the dense molecular clouds associated with it. Relative to a dense
molecular cloud, the size scale of a diffuse cloud is expanded by a factor of 10, but
the density scale is decreased by about three orders of magnitude. If the dust-to-gas
ratio remains constant, there would be two orders of magnitude less material than
has been identified in the boundary layer, assuming chondritic abundances. The short
time-scale signature observed in the geologic record does not resemble that expected
from a giant molecular cloud, unless there was a close encounter with a very dense

condensation, with n(Ho) - lol.) ems =

85

Asaro: With respect to isotopic ratios, was the material in these molecular clouds made

at essentially the exact time that the Solar System was made?

Napier: No.

Asaro: If different "factories" made uranium isotopes in the same ratio, but the uranium

was made at times that differed significantly compared to the half life of TG
which is about 700 million years, then identifiable differences in the isotopic ratios
would be detectable. If the isotopic ratios of uranium in material captured from
such a cloud have a Solar System value, then they must not only have been made in
the same way, but the length of time that has subsequently elapsed must also be the

same.

Clube: We are discussing heavily-diluted observations of uranium, the interpretation of

which is very theory dependent. Large molecular clouds with a different composition
from those associated with the origin of the Solar System may well exist. But that
raises the question as to whether or not galactic processes are adequately understood
at present. Does galactic material remain in one region and undergo continual
chemical recycling? Is spiral-arm material being displaced from another part of the
Galaxy? These questions are related to how spiral-arms are formed. Are they created
by density waves, traversing the intergalactic medium and reprocessing material? Are
they periodically produced in some other way with characteristic compositions? We
certainly recognize the point you are making, but do not believe the data at the

moment provide a satisfactory answer.

Halliday: How high is the temperature in molecular clouds when cometary material is

condensing?

Napier: It is usually estimated to be 10 or 20°K.

Halliday: Some condensates in carbonaceous chondrites formed at temperatures of

approximately 1600°K.

Clube: Where the condensation takes place may be integrally bound up with a satisfactory

86

theory of Solar-System formation itself.

W. Alvarez: So far no one has located a circular hole anywhere on Earth of the right
size and age to be the crater of the impacting asteroid. Some have sucyested that
the impact might have initiated volcanic activity, and that a large accumulation of
volcanic rocks should be sought instead of a crater. Dr. F.W. Whipple postulated in
a public television programme that if the impact occurred close to a mid-oceanic
ridge, where the lithospheric crust is thin and material is very close to the melting
point, then a large quantity of volcanic materials would be released. He examined
the mid-oceanic ridges for a large accumulation of volcanics and identified one that
seemed to be the correct age called Iceland. The problem is that there was no
oceanic crust in the vicinity of Iceland 65 million years ago. Greenland and Norway
were in contact until something like 55 or 60 million years ago, according to
magnetic anomaly patterns documented by Talwani and Eldholm (1977). The Cretaceous-
Tertiary boundary is very close to Anomaly 29. The earliest anomaly in the North
Atlantic is Anomaly 24, which is near the Paleocene-Eocene boundary. Seyfert and
Sirkin (1979) have considered the possibility that large impacts may trigger hot-
spot activity and generate the onset of seafloor spreading. Dr. Seyfert (personal
communication) suggests that the Cretaceous-Tertiary object, after some time delay,
caused the onset of volcanism in this area. In support of this, he notes that the
ages available on volcanic rocks in Scotland and in eastern Greenland do in fact turn
out to be very close to the age of the Cretaceous-Tertiary boundary. However,
volcanism on the west coast of Greenland is clearly associated with the somewhat
earlier openings of a rift between Baffin Island and Greenland. Volcanologists are
satisfied on petrological grounds that the lavas from east and west Greenland are
related to the same volcanic processes. In west Greenland, ages are known back to
77 million years. Furthermore, a microcontinent called Rockall Bank west of the
British Isles has similar volcanic rocks which have ages dating back to 105 million
years. This general volcanic province is significantly older than the time of the
Cretaceous-Tertiary boundary. That some volcanic ages coincide with the boundary

event is probably coincidental.

Clube: Do radiometric ages indicate a number of impacts down the Mid-Atlantic Ridge?

W. Alvarez: There are several volcanic centres which have been related to hot spots
in the mantle, extending from Jan Mayen Island in the north to Bouvet Island off
the coast of Antarctica. The onsets of volcanism extend from well below the

Cretaceous-Tertiary extinction to somewhat after it.

Hsu: Walvis Ridge volcanism commenced about 80 million years ago, and if it had anything

to do with an impact it would not be the Cretaceous-Tertiary impact. Some colleagues

working in the Scottish Highlands on basaltic ring-dykes with ages of 65 million
years half jokingly suggested that they may have something to do with the terminal

Cretaceous impact.

W. Alvarez: It is a curious coincidence that there is an extensive volcanic pulse in

India, the age of which is essentially indistinguishable from 65 million years.

McLean: There seems to have been a general increase in mantle circulation and opening
of the North Atlantic at the time of the Cretaceous-Tertiary extinctions. Iceland
has volcanics which are relatively rich in alkalis. However, can these lavas be

related to a meteorite impact?

W. Alvarez: These lavas are all much younger than the extinctions.

McLean: Schilling (1973) makes a case that they are the result of plume activity, and
related to a general opening of the North Atlantic. I think that we are in

agreement that Iceland is not a good choice for the site of meteoritic impact.

W. Alvarez: There is no evidence to indicate anything other than a normal continuation
of the process of the opening of the Atlantic. Rockall Bank itself is separated
from the British Isles by what appears to have been an aborted rifting attempt. The

processes were clearly active before the Cretaceous-Tertiary extinctions.

Hsu: Icelandic volcanism differs from mid-ocean ridge volcanism in that the light rare
earth elements are less depleted. They are supposedly derived from mantle sources
which were less differentiated than those of other oceanic basalts. Icelandic

volcanism has continued to the present.

W. Alvarez: In comparison with the angular momentum of the Earth or the energy transport

88

Hsu:

involved in plate tectonic overturn, an impact is in the level of background noise.
However, Dr. Seyfert (personal communication) suggests that the mantle may itself
be in a metastable condition, so that meteorite impacts could be triggers of much

larger energy transfers.

Hot spots do not stay stationary with respect to the plates. If one was created
as the result of an impact, its volcanic product would not have come out at the same

point of a moving plate during the last 65 million years.

W. Alvarez: Deccan Trap volcanism began as close to the extinctions as time can be

Hsu:

measured. One thousand meters of plateau basalts, including many flows, were

extruded within no more than four and possibly as few as two polarity zones. At that

time these zones were of about a half-million years duration, so that the volcanic

episode lasted only one to two million years. One would expect that melts of that sort

would be very undifferentiated chemically, as are the Columbia River basalts, but
there are substantial amounts of rhyolite and other chemically differentiated

volcanic rocks in the Deccan Traps. There may be an incongruity between the time

required for differentiation to occur and the duration of the entire episode. Perhaps

this is evidence that the material did not simply flood out of a hole suddenly
created in the crust. Iceland is not the site of an impact. The Deccan Traps might
be, but a strong case cannot be made for it at present. Perhaps the most important
conclusion might be that something caused a volcanic-tectonic disturbance about 65

million years ago.

Were the Deccan basalts deposited on a flat surface?

W. Alvarez: That would appear to have been the case.

Hsu:

Can the occurrence of craters be excluded?

W. Alvarez: The pre-Deccan Trap surface has not been sufficiently well explored. It

extends over a larger area than a one- or two-hundred kilometer crater. The 65-
million-year level in adjacent oceanic areas has not yet been penetrated by D.S.D.P.

cores.

Feldman: Where were the Deccan Traps 65 million years ago?

89

W. Alvarez: They were probably south of the equator.

Thierstein: How do they compare volumetrically with the Columbia River basalts?

W. Alvarez: I believe they are substantially larger.

McLean: The Deccan Trap lavas are today concentrated in western and central India, and

cover about 500 thousand square kilometers. Along the west coast near Bombay they

are up to three kilometers thick. Isolated outliers of basalt on the east coast of
India suggest that before erosion the lavas probably covered at least a million

square kilometers. Pascoe (1964) considers basalts in Baluchistan to be the same age.
Reunion, a volcanic island east of Madagascar, is considered to have been the hot spot
then. Submarine volcanism occurred in what is now the Arabian Sea. The lavas in
western and central India today are only a fraction of the total volcanism of that
time. The Brito-Alaskan flows may be younger, but extensive volcanism may be involved

in the Cretaceous-Tertiary transition scenario.

W. Alvarez: Some of the volcanic outliers in eastern India are of late Jurassic or

early Cretaceous age.

McLean: India was drifting very rapidly at a rate of up to 17 to 18 centimeters per
year. Morgan (in press), and Chandrasekharam and Parthasarathy (1978) suggest that
convection from under the Indian plate facilitated its rapid drift to the north.
Mantle-plume activity may thus have been partly responsible. India does seem to
have been domed up, and the lavas emerged from fissure and crack systems. The lavas
were quite fluid, and individual flows can be traced for 60 or 70 miles. Eruption
was not continuous, nor would I postulate that the extinctions spanned the entire
five million year period of Deccan volcanism. Massive outpourings may have coincided
with the early stages of this volcanism. Deep origin volcanism could bring large
quantities of primordial CO; to the surface (Wiley and Huang 1976). Because
latitudinal temperature gradients were less, oceanic circulation would have been
slower during Cretaceous time than at present. Carbon dioxide would also have been
taken up more slowly by the oceans. Volcanic CO» would accumulate in the atmosphere

and then, by diffusing into the oceanic surficial water, produce a lowering of pH.

Calcareous planktonic organisms indeed were severely affected in the terminal
Cretaceous extinctions. Perhaps the siderophile enhancement does reveal information
on mantle processes. The plumes may be cited as a potential source of the siderophiles,
through which platinum-group elements were transported to the crust where they would
be scattered by volcanic exhalations into strata of about the same age as the
extinctions. I postulate that the primary source of biotic stress was a global
warming. There is a drop in 180 values across the contact that can be taken as a
warming signal. Volcanism began approximately contemporaneously with the marine
extinctions. There is some evidence that the extinctions of so-called Cretaceous
organisms in the marine realm spanned several tens of thousands of years.
Magnetostratigraphic techniques enable us to correlate more precisely terrestrial and
marine sediments. Butler and others (1977, 1978, 1979), basing their interpretations
on these techniques, would have the extinction of the dinosaurs occurring at least

a half a million years later than the marine extinction. Perhaps a progressive
accumulation of CO» in the atmosphere had by then resulted in a warming sufficient

to disrupt reproduction among the large reptiles. Late Pleistocene mammalian
extinctions may have also occurred as a result of an abrupt postglacial warming and
disruption of reproductive physiology. Many modern reptiles in desert areas cannot
reproduce during the summer because with the increasing heat, reproductive tissue
degenerates. So with a global warming, dinosaurs would progressively experience
reproductive problems. Erben and his associates (1979) showed progressive thinning
of dinosaurian egg shells from below the boundary up to the Cretaceous-Tertiary
contact. Interestingly, during hot summer months the shells of chicken eggs begin

to thin until the hens stop laying (Wolfenson et al. 1979). The extinctions can

be seen as a slow, progressive phenomenon, particularly if the abrupt termination

of ranges is due to hiatus control as a consequence of a regression of the seas.

W. Alvarez: The regression is a valid explanation for simultaneous terminations of ranges
in shallow water sediments, but at a depth of two thousand meters in the oceans a
hiatus need not occur. If there is a gap in the record of deep-sea sediments, it
is extremely short. The geochemical anomaly, magnetic stratigraphy, and calculated

sedimentation rates are strong evidence that if a break does occur in sedimentation,

91

it is very short. The enormous effect of a regression in shallow water will not be

there.

McLean: In the Indian Ocean there are deep water hiatuses that span several tens of

millions of years.

Hsu: Yes, but they are not relevant to the problem because they do not coincide with

the Cretaceous-Tertiary boundary.

Russell: The biostratigraphy of the dinosaur egg shells in southern France has not
been well studied. No one has attempted to examine carefully the distribution of
different egg morphotypes through the lithostratigraphic units. Changes have been
seen near the boundary; what kind of changes occur lower in the section? The data
available are not of the same quality as microfossil data from pelagic carbonate
sequences, such as those considered earlier. The surface-volume proportions of
dinosaurs have generated some discussion on how much activity they could sustain and
still balance their thermal budgets. With regard to a warming-trend through the
Cretaceous-Tertiary boundary, it seems likely that in the case of gradual changes
refugia might persist in some areas of the globe which would be climatically

compatible with the survival of some dinosaurs.

McLean: In the late Pleistocene mammalian extinctions, refugia were created in the

general warming of the globe because all areas did not warm equally. With a CO»

increase, the Northern Hemisphere warms more than the southern, and certain regions

of North America and northern Eurasia warm far above the Northern Hemisphere

average. With respect to the definition of the Cretaceous-Tertiary boundary, can

we draw precision from imprecision? The crux of the issue is that these secondarily-
truncated ranges have been defined as the top of the Cretaceous. In a time sense

the K-T boundary would have been higher in the section. Then a clay is deposited on
this truncated surface, which has nothing to do whatever with the true extinctions

at the end of Cretaceous. Let us postulate a variety of ranges of organisms in

time whose remains are preserved in shallow-marine sediments. The seas then retreated

to their basins, and the sediments were removed through erosion down to the new base

level. All that remains in the record are a series of decapitated ranges.

Russell: Is there any place on Earth where we should look for a complete section?

McLean: On the basis of some recent drilling, some zones were apparently discovered
which have not been identified before. This would indicate that some of the sections
that have been studied are not complete. It is known that a large volume of rock

was stripped away after the great terminal Cretaceous regression in many areas.

Smit: Where is the large volume of rock that has been stripped away now? It must be

deposited somewhere.

McLean: All along the east coast of North America are great thicknesses of Mesozoic

and Cenozoic sediments that came from the erosion of the continent.

L. Alvarez: Dr. Butler's observation that the dinosaurs disappeared in a positive
magnetic sequence and the marine organisms disappeared in a negative magnetic
sequence clearly implies that these extinctions were not synchronous. This point
was addressed in our paper. I recently spoke with Dr. Butler on this important
problem. He indicated that he is reviewing the evidence and is ne longer certain

of his earlier conclusions. I believe that he would say that if he were here.

Asaro: With respect to the enormous amounts of material that are presumed to have been
eroded from all of the boundary sections that have been measured, the abundances of
the rare earth elements can be measured with high precision and their ratios do
vary somewhat from point to point. In one section in Italy the abundances just
below and just above the clay layer could not be distinguished from each other using
the highest precision techniques available. We could not measure the time that
elapsed between the deposition of the sediments sampled, but there could not have

been an enormous amount of missing or eroded rock.

McLean: In many, but not all cases, there is a considerable hiatus in terrestrial
sediments. In one section on the east coast, Upper Cretaceous strata are absent
and Tertiary deposits lie on Lower Cretaceous strata. Thus, in many regions a hiatus

of often indeterminate magnitude can be identified in marine sediments. In response

98

94

to any phenomenon all parts of the world do not change exactly the same way. For
example, calcium carbonate dissolution occurs at certain depths, which may change
with latitude. It would be very useful if a section could be found across the

boundary that could be proven to be complete.

W. Alvarez: Every responsible stratigrapher is well aware of the effects of erosional
hiatuses. Anyone who is concerned with the Cretaceous-Tertiary extinctions thinks
about it constantly. It is very difficult to imagine any way by which a sequence can
be shown to be absolutely complete. It is clear in a number of sections, particularly
the one at Caravaca, that if any time is missing it is less than the thousands of

years range. This is all we can hope for.

McLean: In some localities in Europe, the end of Cretaceous has been defined by marine
extinctions. The base of the Tertiary is defined by the appearance of G, eugubtina-

Between these levels is a time-gap which should be discussed.

Hsu: How are we to define an incomplete section? By the presence of a 10-day hiatus

or a million-year hiatus?

Smit: The clay layer at Caravaca was deposited in 2,500 to 10,000 years.

Hsu: According to minimal pelagic sedimentation rates, it is highly unlikely that this

clay layer represents more than two million years.

McLean: Was it deposited uniformly from top to bottom, or was it dumped in as a storm

deposit with the top few centimeters representing a few millions of years?

W. Alvarez: An argument can be made from paleomagnetic stratigraphy. We know that all
of the reversed and normal polarity zones are recorded at Gubbio because the match
to the seafloor spreading record is unmistakeable. The duration of each of these
polarity zones is known with some degree of accuracy. We also know how thick they
are in the section at Gubbio. A rate of sedimentation can therefore be calculated.
These rates can vary by factors of three from zone to zone, but not by factors of 10.
If the Cretaceous-Tertiary boundary is assumed to represent more and more time, then

less time will be represented by sediments above and below it in the reversed magnetic

polarity zone that contains the boundary. Sedimentation rates are driven up to
approach unacceptable values. Yet the sediments in adjacent polarity zones remain
similar, with identical nannofossil matrices containing 5 to 10% foraminifera. There
is no change in lithology to suggest that sedimentation rate changes of major magnitude

occurred.

L. Alvarez: It has been stated that a general environmental change need not produce
uniform sedimentological consequences all over the world. However, the iridium

anomaly has a world-wide distribution, with only minor differences between localities.

McLean: It could be interpreted as evidence of volcanism.

Hsu: Where are the data showing trace element similarities between the clay layer and
the Deccan volcanics? This question has already been posed, and the materials are
shown to have different trace-element chemistries. Unless new data are recovered
which prove that the other data were wrong, the question need not be discussed
further. What is a large amount of CO»? How many megatons of CO» would be
introduced into the atmosphere by Deccan volcanism, and how does this amount compare
to the amount of CO» already present in the atmosphere? The earth sciences today
are more quantitative than in the 19th century, and unless we can speak in terms

of numbers our arguments lose their force.

McLean: The linking of volcanism to the Cretaceous-Tertiary transition scenario is
new, and the consequences are not yet as well known as they would be if the issue
had been under consideration for some time. Has anyone examined the strontium 87
to 86 ratios in the boundary clay? In the Deccan volcanics the ratios are low, but

at least some chondritic material has a relatively high strontium 87 to 86 ratio.

Thierstein: We have examined the thicknesses of the biostratigraphic zones in various
carbonate sections. The sedimentation-rate in any one section seems to be reasonably
constant because the individual biostratigraphic events tend to be spaced in
proportion to the sedimentation-rates calculated for the total thickness of the
section. Parenthetically, sedimentation-rates would have to be corrected for density

and porosity differences between deep-sea and epicontinental sediments in order for

95

the numbers to be comparable. There is no indication of a large or disproportionate
gap in the record in any of the sites considered, which are the best ones known.

The quality of preservation of coccoliths does not decline across the boundary
(Thierstein, in press A). The decrease in the carbonate content of the sediments

is presumably due to a decline in the productivity of calcareous phytoplankton. The
preservation of coccoliths in many earliest Tertiary (basal Danian) assemblages is

actually better than in the lastest Maastrichtian.

McLean: Perhaps pH changes in the ocean depressed phytoplankton productivity.

Thierstein: The preservational patterns we observe in recent coccolith assemblages

from deep-sea sediments are not determined by pH in the surface waters, but by the
dissolution gradient in the deep-sea. The Cretaceous-Tertiary boundary is not
associated with a dissolution event in the Atlantic Ocean, although it may be in

the Pacific. However, it is often difficult to distinguish a dissolution event from

a diagenetic feature. Dissolution in low carbonate, high organic carbon sediments

of middle Cretaceous age has been shown by imprints of coccoliths, which are no longer
preserved, in black shales (Thierstein et al., in press). Many of the preservational
changes observed in the Mesozoic record are probably diagenetic rather than

primary.

McLean: In volcanic eminations, carbon gases are second in abundance to water-vapour.

Carbon dioxide could enter the water by simple diffusion over a period of time,

independent of movements of the carbonate compensation depth.

Thierstein: Two additional points can be made in reference to changing the CO» content

96

of the atmosphere. The first is that living coccolithophorid cultures can be

placed in a sea-water medium with a pH as low as five (Wilbur and Watabe 1963). The
effect on Emiltania huxleyt, for example, is that it will shed its coccoliths and
survive naked for as long as two to three years. If the coccolithophores are
returned to higher pH environments with nutrients they will produce coccoliths as
formerly. A second point is that it is unlikely that much CO» can accumulate in

the atmosphere without it ultimately entering the ocean. This would produce a

global dissolution event in basal Tertiary sediments which is not seen in some sections.

Hsü: In the drilling on the Walvis Ridge, it was very easy to locate the Cretaceous-

Tertiary boundary because it coincided with a red clay. Perhaps the change in
sedimentation is due to a decrease in coccolith productivity, but there are also
indications of increased dissolution. At D.S.D.P. site 356 the red clay is absent,
as was one of the nannozones across the Cretaceous-Tertiary boundary. Dr. Perch-
Nielsen insists that the zone should have been there, although you disagree. There
is a possibility that the zone and the red clay were removed by mechanical erosion.
I do not think that you can say that there is no indication of increased CaCcO3

dissolution at the end of the Cretaceous.

Thierstein: Most sedimentary sequences show a change in lithology which is related to a

Hsu:

L.

carbonate decrease. This can be achieved by decreasing the supply of carbonate
from the photic zone or by increased dissolution. The preservation state of

individual fossil assemblages often improves across the boundary.

If a concentration gradient exists in which the water is greatly undersaturated,
sediments tend to dissolve at the water-sediment interface. In most of the D.S.D.P.
sites, the Cretaceous Globotruncana have undergone a great deal of dissolution.
There is thus evidence of increased dissolution in the sea floor at the end of the

Cretaceous, but conditions were improving when the Tertiary began.

Alvarez: Two separate arguments have been presented that the material in the
boundary layer is extra-terrestrial in origin; one in our original paper, and
another by Ganapathy (1980) and by Asaro earlier in this conference. I would like

to ask Dr. McLean in what way he feels these arguments are invalid.

McLean: I have approached the Cretaceous-Tertiary transition from the point of view

of trying to understand a broad spectrum of Earth processes. This issue could be
addressed without discussing meteorites or siderophiles at all, but I want to be
complete. The dinosaurs can be seen as disappearing in a climatic change which is
related to very broad and sweeping mantle processes, which have been in operation
throughout geologic time. These mechanisms may provide an alternative source for
the siderophilic enhancements. At this stage it is wise to consider as many

alternatives as possible, and keep examining the data.

S)7,

L. Alvarez: Two separate pieces of data have been cited which exclude some material

in the boundary clay from being derived from volcanos. In what way is this incorrect?

McLean: The distribution of siderophiles throughout the entire Earth is not well known.
Deccan volcanism is real. It began at almost precisely the same time that the marine
extinctions began. Fluctuations in the carbon-cycle may have produced the observed
paleontological scenario. Basaltic ultrabasics, particularly those from greater
depths, are generally rich in siderophiles. If the Deccan volcanism resulted from
mantle-plume activity, perhaps the siderophiles have an origin from one or several
levels of the mantle. A mantle-plume mechanism is proposed as a terrestrial

alternative that is worthy of consideration.

L. Alvarez: Why is a mechanism that has never been observed before to be preferred
over one concerning which much information exists and which would produce the

observed siderophile signal?

McLean: I cannot say that a meteorite did hit the Earth. I cannot say that a meteorite
did not hit the Earth. Its role in the extinctions needs to be explored. The role
of massive volcanism needs to be explored. The role of massive flood basalt

volcanism in climatic change needs to be explored. The siderophiles are incidental.

Asaro: Do you think it would be useful to carry out chemical analyses of the basalts of

the Deccan Traps?

McLean: It could well be if the right ones were studied. In some volcanic episodes the
siderophile-enriched materials seem to come up as an early pulse. I have no figures

on the platinum-group elements in the Deccans.

Grieve: From the point of view of cratering statistics, the possibility exists that a
large impact event could have occurred 60 to 70 million years ago. As far as the
geochemical anomalies are concerned, from the patterns of siderophile enrichments
in material associated with impact craters, it should not be a matter of concern that
the siderophile fingerprint is slightly different from site to site. Projectiles
vapourize on impact and fractionation takes place in the vapour, as well as in various

chemical interactions with the target materials. Fractionation will be better

98

understood when the projectile and target materials are more accurately known.

A

projectile of the size proposed will not be affected by the ocean in a physical sense,

although water will be thrown into the atmosphere. The volume of material excavated

relative to the volume of the projectile varies with impact conditions, but is probably

on the order of several hundred to a thousand times. The bulk of that material is

thrown out of the crater in ballistic trajectories and forms an ejecta blanket, which

is a continuous cover out to two crater radii, although some ejecta may extend out as

far as 10. This material originates from a position which is relatively distant from

the projectile and contains little or no projectile component. Of more interest is the

material that is ejected as the projectile is penetrating the ground. As a projectile

enters the ground, the flow-field is partially directly radially into the target but a

component is also directed upwards. The latter material contains a large projectile

component. At this point the projectile is a super-heated vapour, with a temperature

perhaps as high as 10,000°C. The proportion of projectile to target material in the

vertical flow-field varies between 0.01 to 1.0 with impact conditions (O'Keefe and

Ahrens 1981). This is the material that would reach high levels in the atmosphere,

and could be deposited as a global clay layer. Thus there may be no dilution problem

with respect to an asteroidal impact model, if only this initial high speed ejecta is

considered.

Kyte: We were concerned about this dilution problem, and tried to devise a few
methods to obtain the enrichments of 20% which were observed in the Danish
One hypothesis involved an asteroidal core which will also have chondritic

metal abundances. In this respect it is difficult to determine whether or

alternative

section.

noble

not a

chondritic or an iron-core object were involved. There appears to be a chromium

enrichment in the horizon and chromium is normally depeleted in iron meteorites.

a chromium component were identified, perhaps an iron meteoroid source could be

ruled out. Another alternative hypothesis involved an interstellar dust cloud.

itag

We

came to the conclusion that the siderophile anomaly would be distributed through too

great a thickness of sediment, and left that question open. Our third hypothesis

involved an accretionary event without an impact. We suggested the possibility of

tidal disruption of cometary material and accretion of numerous small fragments, which

would be destroyed in the upper atmosphere and produce little or no ejecta.

The

data

99

do not uniquely constrain a chondritic asteroid impact, but an accretionary event
must have occurred.

We have identified an iridium anomaly in an Antarctic core which is only a
few million years old. The noble metal carrier in that horizon is apparently
chondritic material and there is no evidence of impact ejecta in that horizon. These
particles are quite unstable in the marine environment, and only large particles
have been recovered. All have severely etched surfaces, and if this horizon were to
survive for 65 million years these particles would be completely dissolved. The
primary carrier of the noble metals would no longer be recognizable. Perhaps similar
material served as the carrier for the Cretaceous-Tertiary material. We are
examining some microtectite horizons which are believed to be derived from ejecta,
but have not found associated iridium enrichments. It would be useful to further
constrain these alternative hypotheses. An advantage of a tidal-disruption model is
that the accretionary event would probably last for a period of weeks to months.
Virtually all of the energy available would be transmitted to ecosystems, rather

than being buried by an impact, so the extinction effect may be enhanced.

Thierstein: I have been wondering about alternative processes to enrich these sediments
with noble elements, and would like to pose three questions. This first consideration
is whether these enrichments could be a consequence of the mass extinction? If some
of these elements are enriched in present-day biomass in similar proportions, then
the extermination of that biomass would produce similar enrichments in the sediments.

Does any evidence exist to rule out such a possibility?

Rucklidge: In the process of studying the iridium anomalies in the Danish section we
made various separations. Silvana de Gasparis prepared some organic extractions and
these were measured directly using the tandem accelerator-mass spectrometer method.
The organic extractions had extremely high platinum and iridium values, in the
parts per million range. We suspected that, in the process of extraction, platinum
metals were also concentrated into this organic fraction because of the chemistry
involved. Using physical methods for extraction the same type of enrichments were
again detected. However, only a small fraction of the total amount of iridium present

in the section could be contained in the organic fraction.

100

Thierstein: In this example the iridium is related to organic material still preserved.
If all of the organic material had decayed, the residue would contain the element
enrichments. A second kind of consideration is: could the enhancements be related to
inorganic processes caused by oxygen deficiency, since noble elements tend to be
enriched in sulphides (Wedepohl 1970)? We (Thierstein and Berger 1978) have speculated
on the possibility of developing an oxygen-minimum layer below a freshwater lid.

There has been a debate concerning an absence of pyrite. However, heavy metals are
precipitated from the water-column during anaerobic periods, and buried. When oxygen
is recirculated into the water-column, metal sulphides may oxidize, so that an absence
of pyrite does not necessarily imply the absence of an anoxic event. The possibility
still exists that there was an oxygen deficiency at depth across the Cretaceous-Tertiary
boundary. Present-day anoxic environments could thus be examined to see if a
mechanism exists whereby noble elements are precipitated from seawater inorganically.
The third consideration is: if an asteroid impact provides a mechanism for the
extinctions at the Cretaceous-Tertiary boundary, what makes that mechanism so unique?
The extinction in my opinion remains unique both in magnitude and in timing, although
it is true that less is known of the other mass extinctions. If an asteroid impact
has such devastating effects, where are the paleobiologic effects of smaller, but more
numerous impacts? This question may in part be answered by finding an iridium anomaly
at another time-interval. If the iridium concentration at 2.3 million years occurs in
other localities, where are the biotic extinctions? It must be deleterious for some of

the phytoplankton if only 60% of the sunlight were blocked.

Asaro: Concerning the first point, could there actually be less organic material associated
with the extinction event in the sedimentary record? An extinction would depress
production, and reduce the concentration of an element that might be deposited with

the organic material.

Thierstein: This is a question concerning global cycles of various elements. How much
carbon is used up at any one time in the biotic reservoir, and how much leaves the
reservoir into the sediments? We would have to make some assumptions on what the
biomass was at that time and what the flux-rates were. For simplicity, perhaps the

present-day oceans should be considered as an analogue, the amount of oceanic biomass

101

measured, and if this were destroyed, the trace element signal buried in the sediment
should be calculated. If the carbonate decrease across the boundary is due to a
productivity decrease rather than a dissolution event, then the carbon that was
formerly in the biomass, assuming that the biomass was considerably smaller after
the extinction event, must be dissipated. A chemical trace should occur in the

sediment.

Russell: The biomass of dinosaurs prior to the extinction event was probably about one

one-hundreth that calculated for an impacting asteroid.

Béland: The fact that biomass is not accumulating can simply imply that most of the

additional biomass produced in one year by all sources is used up by the biomass itself.
Individual growth and births equal total deaths, although production is continuing.
The productivity is consumed to be used as kinetic energy or metabolism, and transformed

into heat.

Thierstein: This is not quite in agreement with global steady-state model assumptions

Hsu:

102

where the carbon supply into the ocean is held to be approximately equal to the carbon
that leaves the ocean reservoir, both in carbonate and in organic carbon. My
understanding is that the annual carbon supply is relatively small compared to existing

biomass.

The annual carbon supply is 8.6 x 1014 grams of carbon atoms and the present ocean
has 3.5 x 10!% grams of carbon atoms in dissolved carbonates (Hsü et al., ms.). With
respect to your first two considerations, why should either organic or inorganic
precipitation of platinum-group metals produce ratios similar to those of chondrites?
It would be too much of a coincidence. With respect to your third consideration
concerning the uniqueness of the Cretaceous-Tertiary extinction, I suggest that the
uniqueness may be explained by assuming cometary instead of asteroid impact. The three
agents of stress which have been postulated are dust, poison and heat. Perhaps only a
comet would provide enough poison. How do particular groups of organisms
differentially react to dust, poison and heat? The record of these groups across the
Cretaceous-Tertiary boundary might reveal which stresses were more lethal during the

interval of crisis. A cometary poison model might be unique because comets do not

impact as often as do stony meteorites. Further, toxic materials in asteroids may be
largely buried as a solid body after the impact, while cometary material could more
easily be dispersed to some extent by the atmosphere. Heavy metals could form compounds
soluble in seawater which could be recycled, and become agents, more toxic than cyanide.

Again, experiments should be carried out on the toxicological resistance of
particular groups of organisms and compared to the record of the group through the
extinction interval. For example, it is very difficult to locate the most important
extinction-interval across the Cretaceous-Tertiary boundary on the basis of some groups
of echinoderms. Are they equally resistant to all of these poisons and to agents of
physical stress? More attention should be given to the biological patterns of

extinction.

L. Alvarez: At the Radiation Laboratory in Berkeley, when large amounts of radioactivity
were first available, we became very interested in the problem of surviving radiation.
When radiation dose (or biological insult) is plotted against survival, the curve is
initially nearly flat, but then declines sharply through the point of a 50% lethal
dose (LD50). The curve then flattens out into a very long tail, like that of an
exponential curve. When the LD50 is exceeded by a number of probable errors or standard
deviations, and then exceeded by twice that, the number of survivors declines by an
enormous factor. This may be a key to understanding why the Cretaceous-Tertiary
extinctions were so severe. Had the stresses been twice as severe we would not be
here to talk about them. For example, a large population of organisms will produce
survivors as long as it is not depleted below a thousand members or whatever number is
required to reproduce sexually. A controlling factor that decides whether a species
survives or not may be whether or not its population, multiplied by the fraction that

can survive a given insult, exceeds this critical number of one thousand.

Brasch: The LD50 can be defined quantitatively. However, it is also important to determine
consequences to the food-chain. Applying these statistics to a mixed population of
interacting species, many can easily survive a given insult, but if a key primary-
producer cannot, then the others will ultimately be removed by starvation. Thus,if a
global darkness is extended beyond a certain time, then the food-chain is destroyed,

LD50s notwithstanding. If the darkness is too short there will be no long-term effect.

103

The action spectrum for photosynthetic activity under given models of atmospheric

darkening should be studied because some photosynthetic organisms will react primarily

to red or far-red light, others to green, and so forth.

L. Alvarez: Whether the stress is due to darkness, cyanide or some other agent, the

survivorship curves tend to be similar. If the Cretaceous-Tertiary insult were at such

a level that 20% of the genera survived, and then this insult were doubled, then nothing

would survive. Doubling the size of an asteroid implies changing the diameter only by

the cube root of two.

Brasch: In connection with a toxicity model, if we envisage toxins like cyanide and heavy

metals such as nickel and osmium, which are severe metabolic poisons, it again becomes

difficult to rationalize the heterogeneous extinction pattern observed. At certain

levels the common metabolic enzymes of all organisms would be affected.

Jaworski: This may depend on the environment in which an organism resides. The

availability of a particular toxin will depend on the chemistry of the aquatic or

terrestrial environment. Also even among fish species, differences in sensitivity

of two or three orders of magnitude can exist in LD50 to the same toxin in the same

water.

Brasch: This is true, but an upper-limit can be assigned nonetheless to the concentration

of a given toxin in the oceans if any species survives.

Hsu: In the case of poisons, selectivity can be geographic as well as physiologic. In

the marine realm, poisons could be concentrated in surface currents. The areas of

equatorial current convergence should thus be peculiarly vulnerable to large-scale
poisoning (Hsu 1980). Assuming poisoning, organisms not requiring photosynthesis can
be eliminated without invoking starvation. The total darkness does not seem to be

flexible enough in its effect to explain the selective extinction.

McLean: Terrestrial plants do not generally seem to have been greatly affected in the

extinctions, in contrast to calcareous microplankton in the ocean. How do the effects

of a meteoritic impact discriminate between these two groups?

104

Russell: Terrestrial palynofloras of late Cretaceous age often include on the order of
200 species. Tschudy (1971) has demonstrated a decline to on the order of one-third

Cretaceous levels in Montana, and to about one-half in the Mississippi embayment.

McLean: Some of those declines were taking place before the end of the Cretaceous. The
event at the Cretaceous-Tertiary contact was of short duration. There was a decline
in global temperature, throughout late Cretaceous time, the effects of which can be

seen in terrestrial floras.

Jaworski: Perhaps there were some organisms in a decline that were eliminated in the

extinction event.

McLean: At the Cretaceous-Tertiary contact calcareous phytoplankton were more severely

affected than terrestrial floras.

Hsu: Microplankton could have been destroyed by chemical poisoning. The extinction could
alter ocean chemistry to the point of changing the solubility of some heavy metals,
increasing the toxic effect. Why were calcareous organisms more affected? Perhaps a
change in the CO, content in the ocean resulted in a stress environment for organisms

which secrete calcareous skeletons.

McLean: How is a poisoning of the water compatible with the generic diversity figures

cited in Table 1 of, for diatoms, 10 before and 10 after; for silicoflagellates, 4

before and 3 after; and for radiolarians many of which live in the photic zone, 63

before and 63 after?

Hsu: Diatoms tend to populate cooler waters more densely. Poisons would be concentrated

in the equatorial currents.
McLean: Diatoms are caught by the billions in plankton nets off South Carolina.

Hsu: Those areas in which diatoms suffer only partial mortality would provide a source of

migrants into disaster areas. Those you mention do not secrete CaCO3.
Jaworski: Each of the groups mentioned has a mineral component in its exoskeleton.

McLean: In diatoms it is opaline silica.

105

Jaworski: Shrimp and various other organisms can concentrate many elements in the
exoskeleton, thus reducing concentrations in the cytoplasm and reducing the possibility

of toxicity.

McLean: Does cytoplasm not communicate with the seawater through the pores in diatoms?

Jaworski: There are freshwater algae that are able to survive rather high levels of

heavy metal concentration by concentrating the metal in the cell wall.

McLean: How can an organism concentrate material in its wall, when the wall construction

is due to protoplasmic activity?

Dilcher: Individual cells can pack the vacuole full of metabolic wastes. If
differentially permeable membranes are not denatured by toxins, the organisms (plants)

can usually cope with them.

Asaro: Mechanisms of extinction are likely to be discussed in increasing detail during
the next few years. It would be very useful to have a detailed list such as the one
discussed here (Table 1), but more expanded to show the numbers of species which
became extinct, and where they became extinct. For those species which did not
become extinct, but whose populations renewed themselves, as a function of geography,
the information might be of assistance in locating the impact-site of the object that

caused the extinctions.

McLean: To record a species diversity of 40, for example, below the boundary, and 40
above does not clarify the problem if all of the species are different. There is a

need for more detail.

Russell: In Table 1, generic turnover across the boundary is often masked, or diversity

decreases are obscured, by a rapid post-extinction diversification.

Béland: Assuming that such a list of species were available, and that toxicity levels
were determined in related living organisms, how could one be confident that the

latter results could be applied to the more ancient species?

106

Jaworski: It would be necessary to argue by analogy. The chemistry of the aquatic
situation at the close of Cretaceous time is not known in sufficient detail. An
overriding consideration is that whether at a particular pH, carbonate or chloride

concentration, the uptake of toxins by organisms is permitted.

Béland: Complex organisms, such as vertebrates and large molluscs, show very diverse

reactions to given products.

Jaworski: Perhaps what was suggested was that two separate groups of organisms be
identified that are quite different in their physiological characteristics, and then
used as test cases in relation to the biological insults which may have been operative

at that time.

Béland: Perhaps some priority should be given to the primary-producers, because of the
possible implications on food-chain. The reaction of unicellulars may be more constant
than that of the higher invertebrates. For example, forams today and forams living 30
million years ago, belonging to the same taxonomic categories, would probably react to

a stress in essentially the same way.

Thierstein: When it became apparent to me that an instantaneous extinction of phytoplankton
occurred at the end of the Cretaceous, I was initially disappointed because models of
ecologic succession or phyletic gradualism could not be applied to the paleontological
phenomena. Knowing that four or five species were present in earliest Tertiary strata,
an obvious question was what their ecological preferences were during late Cretaceous
time. The data I have assembled on late Cretaceous biogeography may be the most
complete available because of sample-sizes including thousands of fossils (Thierstein,
in press B). There is a tendency for coccoliths, which are typical of early Tertiary
time, to occur in late Cretaceous sediments in higher latitude sites rather than
tropical sites, supporting the thesis that whatever lived in high-latitudes in the
marine environments was somewhat better adapted to the extinction than forms living in
low-latitude environments. The relative abundances of Tertiary taxa in the late
Cretaceous sediments, assuming no contamination, are on the order of a tenth to a
hundredth or less of a percent, making it impossible to discern an evolutionary

pattern such as one of phyletic gradualism. The same applies to the appearance of

107

Emiliania huxleyt, which is the youngest coccolithophorid living today (Thierstein

et al. 1977). It appeared in the geographical record about 250 thousand years ago and
remained in incredibly low abundances for 80 to 100 thousand years. Then it began to
bloom, and this is when it would conventionally be found in the geological record. The
point is, with an almost perfect fossil record and large sample sizes, it is not
possible to determine what the diversity of the earliest Tertiary phytoplankton was in
the latest Cretaceous. Even their point of origin cannot be determined because they are
initially so incredibly rare. How can one be sure that all of the species of macro-
fossils have been recognized with small sample sizes? Even if a list is made ona

species level, true changes in diversity may not be apparent.

Hsu: Perhaps high latitudes were a Cretaceous refugium, from which survivors could
migrate into new niches after tropical waters were detoxified. There they could

bloom vigourously, because the competitors were gone.

Thierstein: Perhaps they were adapted to darkness because of high-latitude seasonality.

Hsu: In any event, evidence concerning biogeographic provinciality may be more important

than that relating to diversity changes.

Russell: The Aquilapollenites province seems to have suffered the greatest diversity
declines, although it occurs in high latitudes. If Aquilapollenttes were parasitic,
it was a boreal parasite, although existing vegetational gradients suggest that

epiphytic plants are more common in the tropics.

Dilcher: There is no question that some species extinctions and/or replacements occur
across the Cretaceous-Tertiary boundary. But when examined on a generic or family
level these changes are masked, so perhaps a closer look at this boundary from
several different botanical aspects (e.g. leaves, pollen, wood, fruits and seeds) is
justified. Much of the plant evidence is not based on an entire organism. Some parts
of angiosperms are more diagnostic than others. There is an ongoing effort to
understand the relationships of organs across time boundaries. Techniques such as
palynology, detailed leaf analyses and cuticular analyses have helped to resolve

these (Dilcher 1974). Crepet (1981) presents an interesting scenario where the earliest

108

flowering plants have generally been assumed to be insect-pollinated, and we see a
radiation of wind-pollinated plants following the Cretaceous-Tertiary extinctions. He
has proposed that there was strong competition among plants for insect pollinators.
This could have been a result of the extinction of insects in terrestrial environments’
during the terminal Cretaceous event, or as Crepet suggests, the fact that they hadn't
radiated or were not abundant. Surviving plants became facultatively, and later

obligately, wind-pollinated.

Russell: It is in the quiet air on the forest floor that plants tend to rely more on
insect pollination. Would this imply that forest herbs suffered more than emergent

trees?

Jarzen: In the tropics, mode of pollination does not depend on an arborescent versus
herbaceous form. It depends on the spacing of plants and on wind velocities at the

top and in the lower canopy.

McLean: Tectonism was involved in Montana, Wyoming and Colorado, and one would expect
insects to migrate in and out of areas in response to changing environmental

conditions.

Dilcher: Crepet does not raise the possibility of catastrophic events playing any role
in the extinctions. He points out that there is radiation of wind-pollinated forms
at this time, perhaps in relationship to migrations of plants into seasonally dry

habitats and temperate areas.

Hsu: Terrestrial organisms were beyond the effects of poisoned seawater, but may have
been more affected by temperature changes. Temperature changes would in turn affect
environmental conditions in polar regions far more than in tropical regions. I can
imagine the possibility of latitudinally different extinction patterns in terrestrial

and marine organisms.

Jarzen: The possible duration of dormancy in seeds is also a factor to be considered. The
forms growing in the western interior of Canada were ecologically more similar to living
tropical rather than temperate species. Germination studies of northern plants such as

commercial trees and agricultural crops are available, but information on tropical trees

109

is difficult to obtain. Ng (1978) studied germination patterns and timing in 180
species of Malayan trees, or about 4% of the tree-species in Malayan forests. Of
these, 65% would germinate within an average of 12 weeks, 7% within 22 to 68 weeks,
and 28% within 6 to 27 weeks. Of those germinating within the 22 to 68 week interval,
some took as long as 158 weeks, or approximately three years, for germination. Thus a
period of darkness lasting up to a year might be compatible with the survival of the
seeds of some species but those of others would have become non-viable. If the

cloud of dust was circulated more rapidly, perhaps because of turbulence following
the impact, and if 10% of the light returned within a one year interval, then many
tree-species may have survived. Ina rain forest near Leticia, Colombia, I found
that the darkest portion of the forest floor received approximately 10% of the light
falling on an open area near a stream bank. Plants such as Selaginella, mosses and
liverworts were growing in the dim light. These measurements were casually made, but
they do suggest that many plants could survive under conditions of low illumination

for periods of time of up to one year.

Dilcher: I know of no well-described terrestrial sequence across the Cretaceous-Tertiary
boundary in a tropical, low-latitude area. Trees in tropical areas experience even
temperatures and only small changes in day length. Many of them are evergreens; they
can flower at any time, and the seeds usually germinate and grow when dispersed. Plants
that live at high latitudes often have deciduous vegetation, experience months of long
darkness and have physiological mechanisms to delay seed germination. I would predict
that the boundary effects in tropical areas, if the stresses were the result of a

dust model, would be more drastic than in high-latitude areas.

Pirozynski: The suite of adaptations that promoted survival may offer a clue as to what
happened during the extinction period. For example, the rise of the wind-pollinated
amentiferae in forests dates roughly from that time. These systems are adapted to
acid soils, and the trees are often deciduous. Both birds and mammals evolved very
elaborate systems whereby the young animal eats different food from the adult. These
kinds of traits might be conspicuous by their absence in the organisms that were

eliminated.

110

McLean: A marine regression was taking place as well as tectonic changes and volcanism.
Perhaps a meteorite impacted as well. All of these factors should be integrated into

models which seek to account for the paleontological data.

Feldman: Among these factors is the discovery by Dr. Smit of considerable concentrations
of volatiles in the basal millimeter of the Caravaca section. Arsenic, selenium,
antimony and other elements can be toxic in sufficient concentrations. The boundary
material seems to be impossible to derive from terrestrial rock sources within the
very narrow time space of 100 years. Here is therefore another thing that
has to be explained in the context of some kind of accretionary event, and which
appears to rule out an ordinary meteorite. Can we come to some kind of consensus
that, if these data are correct, there is evidence for a cometary object that brought

volatile poisons into the terrestrial environment?

Smit: Many would correlate the arsenic with the pyrite horizon within the boundary
layers, and a small amount of pyrite is also present in the basal millimeter at
Caravaca. These are pyritic residues of probably organic origin. The sulphur
content is lower than the arsenic content in the horizon. It is difficult to imagine
an oceanic mechanism to deposit only arsenic here. It would be preferable to derive
this element from the projectile itself, but this is inconsistent with existing data
from meteorites. Perhaps comets contain these arsenic elements, as Dr. Hsu has

suggested.

Feldman: It is probably as important for the Berkeley group to deal with this problem,
as it is for Dr. McLean to deal with the abundance anomalies of the platinum-group

metals.

L. Alvarez: I can assure you that we will consider this problem.

Kyte: These same enrichments occur in the Danish section, but the arsenic to iridium

ratio is 150 times chondritic.

Smit: Arsenic is so volatile that it may be present in higher concentrations in live

comets than in extinct ones.

ED

Kyte: Of course this is possible; perhaps there may even be sanidine meteorites. But it is

difficult to imply the existence of a type of meteorite for which there is no evidence.

Smit: None of the meteoritic hypotheses can account for the sanidine. Only one iron
meteorite has been found to contain sanidine, and iron meteorites are depleted in

chromium. This is a report of K20 rich spherules in the Tunguska event (Glass 1969).

Kyte: There is a slight enrichment of selenium and arsenic in the D.S.D.P. site 465A
horizon, but not in our abyssal clay section. These elements probably have a

terrestrial source.

Jaworski: Is there a compendium of trace-element work on boundary core samples?

Kyte: There are very few data in the literature at present. We published all our
data on the Danish section and on two samples from site 465A. The Berkeley group
is publishing their data on 465A now, but about a dozen sections have been analyzed

for which the data are not yet available in the literature.

Hsu: We also have some data for arsenic and other rare elements, which can be made

available.

Smit: It might be significant that the arsenic is only present in those layers which
show the highest concentrations of iridium. If these layers are not present, an

arsenic anomaly will also be absent, although an iridium anomaly may remain detectable.

Kyte: We found the selenium anomaly to be much sharper near the base than that for

arsenic.

Jaworski: Have nickel and vanadium concentrations been measured? These are markers for

certain benthic organisms.

Smit: Vanadium correlates perfectly with the detrital clay.

Asaro: We have analyzed a number of Cretaceous-Tertiary boundary samples. Sometimes there
are enhancements of many elements and sometimes there are none. In the seven centimeter

New Zealand boundary layer, for example, the middle section contains 20 PPB of iridium

112

and approximately 5,000 PPM of arsenic. In the clay immediately above and below, the
iridium content declines. It is evident from this pattern that the iridium and other
elements are peaking in slightly different horizons. Of the 50 elements examined, the
abundances of 30 to 40 do not correlate with those of iridium. A lack of correlation
means that discrepancies will occur in the ratios of elemental abundances. The
elements are displaced stratigraphically from each other to some extent. In anoxic
materials from another age, we detected enormous enhancements of zinc, for example.
Many other enhancements were detected in the anoxic sediment sample which correlate
approximately with abundances in seawater. There was no enhancement of iron and

nickel.

Jaworski: Using iridium as an internal standard, do other elements line up with the

maxima of iridium?

Asaro: There is a general correlation, for example, in D.S.D.P. core 465A between the
clay elements and the iridium. As mentioned earlier, the Ni/Ir ratio is smaller than
in Cl chondrites in the detrital clay, but is larger than in Cl chondrites in the
pyrite-bearing silicate (which may be an authigenic clay). However, the patterns in

different localities appear to be sensitive to local conditions.

Jaworski: If there was a single event, and the particulate matter did not dissolve, then
the maxima and ratios would not vary. If dissolution was occurring in aquatic
environments, then the resonance times of the various elements are not similar. Their
deposition will not be simultaneous, and the observed differentiation would simply
reflect the chemistry that has occurred between the initial entering of the aquatic

phase and deposition.

Asaro: In D.S.D.P. site 465A the iridium seems to be associated with the clay and the
pyrite but some kind of elemental displacement may be present. Why is the iridium
distributed through abroad peak considerably above the level where the boundary might
be, and why is a considerable amount of detrital clay distributed over the same region?
The Gubbio type of iridium pattern with a fairly well-defined peak occurs in some
localities. In others it is spread out over 50 centimeters. Examples of multiple-

peaking occur, and it is not known whether the pattern is in the sediments or if it is

14

a function of the coring operation. The patterns are often not as simple as one

iridium peak which tails off apparently exponentially.

L. Alvarez: Could the separation depend on the depth of the ocean? At what depth were

the sediments in the Caravaca section deposited?

Smit: Our estimates, from the few larger echinoids present in the sediments, range between

400 and 2000 meters.

Hsu: The record in the 465A core may be the result of resedimentation phenomena, in

which the peaks were blurred.

Thierstein: There is no evidence for coccolith resedimentation at 465A. The site is in

a foram sand, and winnowing may have occurred rather than redeposition.

Hsu: Winnowing implies the movement of bottom materials, and particle-size sorting

effects.

Thierstein: The most drastic of the documented extinctions occurred in the photic zone
in open-ocean environments, which is one of the areas most vulnerable to the effects
of heat, poison or even radiation. One of the best places to escape these effects
would be in the lower part of the photic zone. Impact mechanisms imply that these
stresses would be distributed globally in the atmosphere and should affect terrestrial

organisms much more severely than those in open-marine environments.

Jaworski: Heavy metals do not move very much through the soil column. The reason why
we have not destroyed ourselves from sludge containing industrial wastes is because
the soil column has a high ion exchange capacity. This is lacking in the water

column because there are fewer cation exchangers.

Thierstein: Why did nearshore organisms suffer less than oceanic plankton?

Hsu: Ocean surface currents did not enter shelf areas everywhere.

Smit: Why not combine a slight heating with a slight blocking of sunlight? If the
photic zone is restricted to the top 50 or 40 meters, and if the atmosphere is heated

to 200°C and some of the heat dissipated into the ocean, the combined effects would be

114

McLean:

W. Alvarez:

Hsu:

more lethal than either in isolation.

The reproductive cells of plants can be as vulnerable to heat as those of

animals. For example, the fern gametophyte is a small plantlet-bearing antheridia

that produce sperm cells and archegonia that produce eggs. The sperm swim in a film

of water from an antheridium to an archegonium.

would have caused major reproductive disruptions amongst some of these plants.

Cesare Emiliani expressed regret that he would not be able to be present and

asked me to summarize a paper authored by himself, Eric Krause and Eugene Shoemaker,

which has been accepted by Earth and Planetary Science Letters. Dr. Emiliani finds

Heat over an extended period of time

that many organisms are living now and were living during the Cretaceous very close to

their maximum temperature tolerances. An increase of a few degrees would be lethal

to many of these organisms. Animals tend to be somewhat closer to their maximum

tolerance than plants. Dr. Shoemaker presents a detailed quantitative estimate of

what would happen during a large meteor impact into the ocean. Dr. Krause considers

the possible atmospheric effects. Immediately after impact in the ocean a moderate

temperature increase would take place due to the passage of hot oceanic lithosphere

fragments through the air, and adiabatic
content of the rocks thrown into the air
energy of the bolide. These two effects

The dust-albedo effect is then likely to

compression of the atmosphere. The heat
could represent up to 20% of the kinetic

are on the order of one to two degrees each.

produce a rather short decline in temperature.

Water injected into the stratosphere rains out quickly and carries the dust with it.

An enormous amount of water vapour remains in the air, dramatically increasing the

greenhouse effect. If approximately two

centimeters of precipitable water is

injected into the stratosphere, and even one hundredth of a centimeter remains,

temperature rises of the order of a few degrees, and up to 10 degrees could be

expected. These are evidently conservative estimates.

Many factors could affect oxygen isotope composition, and the signals from the deep

sea cores may not be as systematic as we

would like. However, we did detect a

temperature decrease about of 4 or 5 degrees immediately after the impact event. If

that signal is real, it may reflect the presence of a dust cloud.

The predicted third

1S

stage of warming was also suggested to me by Peter Signer (personal communication)

who noted that the water vapour from the ocean would increase the greenhouse effect.
Our data show an increase of about 10 degrees following the decline of 4 to 5 degrees.
Our data do not contradict the predictions of Emiliani and others, but we prefer to

assume the greenhouse effects of increased CO, in the atmosphere.

Smit: Are there only a few data points for the decrease and subsequent increase in

Hsu:

temperature?

Ye'si:

Smit: Many locations have now been identified where there is a thick accumulation of

Hsu:

sediment across the Cretaceous-Tertiary boundary, with the same facies above and

below the boundary. In all of those basins the sediments are intercalated with
turbidites. Yet nowhere in the world, to my knowledge, has a turbidite been

deposited exactly on the Cretaceous-Tertiary boundary. An impact in the ocean of the
magnitude usually postulated would disturb the whole marine environment by tidal waves.
A crater in the ocean some 300 kilometers in diameter would lower sea-level by 6 to

8 meters. Some kind of violent sedimentary influx should be detectable in turbidite
basins, such as in Zumaya in northern Spain, where every thousand to two thousand years

a turbidite deposit is recorded.

The absence of such evidence is in contradiction with a violent oceanic impact.

I have never found turbidites at the Cretaceous-Tertiary boundary.

McLean: Tidal waves coming ashore should produce a mixing of marine and terrestrial

Hsu:

116

sediments. Assuming that large amounts of seawater were thrown into the atmosphere,
marine fossils should be rained down at the Cretaceous-Tertiary contact throughout

terrestrial areas.

Two of the three Soviet craters which have been dated at 65 million years measured
25 and 3 kilometers in diameter, and were flooded by seawater after they were
formed. The third, in Siberia, was a terrestrial impact which left behind a crater
of 60 kilometers. The earliest sediments in those craters might reveal local

environmental conditions, and the impact melts might reveal the nature of projectile.

The findings of Soviet researchers will be most interesting in these areas.

W. Alvarez: Oil companies are beginning to find craters by multichannel seismic reflection
profiling techniques. A few have been detected in the northern plains (Williston
Basin) in the subsurface, which are not of the same age as the Cretaceous-Tertiary

boundary, but they do produce oil.

Hsu: A fragmented projectile could produce more than two or three craters. In the Scaglia
Rossa near Ancona in Italy, a crater-like feature was recently located in pelagic

sediments of Maastrichtian age, and covered with Tertiary sediments.

Smit: It may well not be of the correct age.

Hsu: if a cometary object is disrupted through tidal effects the fragments need not

impact simultaneously.

Kyte: Using the tidal disruption model the way we envision, numerous fragments could
enter elliptical orbits around the Earth. Opik (1972) has described a similar
mechanism for the possible capture of the Moon. In this, a proto-Moon is tidally
disrupted and reformed from fragments elliptically orbiting the Earth with perigee
at the point of encounter and apogees extending to a distance of several hundred
Earth radii. Only about half of the fragments are actually captured. Cometary
objects would probably be initially less scattered, but the gravitational effects of
the Moon could hurl them onto the Earth's surface. These scattering effects,
combined with Earth's rotation could easily spread the impacts over 100 degrees of
latitude. However, as we stated earlier, in order to have a disruption and a capture,
a low encounter velocity is implied, meaning that the object must have an orbit
similar to that of the Earth. There are a few Apollos which have apogees and perigees

within about 10% of the Earth's orbit.

L. Alvarez: What is your capture mechanism?

Kyte: This has been described by Opik (1972). Perhaps someone familiar with this type of

problem could model it and determine its feasibility.

117

L. Alvarez: The asteroid could conceivably pass through the atmosphere and lose enough

velocity to be captured and broken apart. However, it would only decelerate by
about 1 g, and the cross-section is inadequate for such a close pass. Direct

asteroidal impacts are more likely.

Kyte: Opik (1966) lists the probability of ejection, collision and tidal encounter with

Earth for the comets Encke, Apollo and 1948E. The probability of ejection from the
Solar System for these is about 10% or less, the probability of a collision with Earth
ranges between 88 and 97%, and the probability of tidal encounter is 100%. If the
assumption is made that cometary objects are extremely fragile and if a tidal
encounter occurs before a collision, most of the fragments would not be large objects

at the time they are finally accreted.

Béland: Returning to biological considerations, northern land plants could accommodate

118

changes in temperature and a darkening of the skies better than low-latitude plants.
Deciduous trees, for example, that had lost their leaves might survive an extension
of adverse conditions for an additional six months, whereas plants in humid tropical
environments could not sustain a comparable stress. Savanna plants, such as those

in Africa or Australia, that lose their leaves during the dry season could perhaps
also survive a reduction in light for an extended period of time. Did plants growing
in high latitudes and in seasonally dry tropical regions actually survive the
extinctions better than those in equable environments?

Pollen grains may not reveal the habitat preferences of the parent plant. For
example, there are 600 different species of eucalypt trees in Australia living ina
wide diversity of habitats. If the pollen grains show close taxonomic resemblances,
very little could thereby be deduced about the ecological requirements of different
species. It is probably also true that the pollen entering the fossil record is
biased towards wind-pollinated plants, and pollen from insect-pollinated plants is
not so well represented. Perhaps wind-pollinated plants were highly diversified
after the Cretaceous-Tertiary boundary, but what was their status before this time?
Would they not always be over-represented in the pollen record? When it is said that
the pollen record does not show significant changes in diversity, such a small

percentage of the possible spectrum of plants and habitats may be represented that the

generalization is without meaning.

Dilcher: The germ line which passes through the male gametophyte (angiosperm pollen)
has been the object of extensive selective pressure. Many extant genera and families
have been mapped according to the morphology of that cell and, although many extinct
types of pollen occur in the fossil record, certain unique forms are found only in
particular lines of extant plants today which can be traced back in the fossil
record. Pollen grains (wind and a few insect-pollinated forms) are distributed more
widely in basins of accumulating sediments than are leaves, wood, fruits or seeds.

The pollen record is thus a more accessible part of the fossil plant record.

Béland: This is true. However, consider Eucalyptus saligna, a very large tree that
grows at the border of rain forests, and another eucalypt that lives as a tiny shrub
in very dry areas. Their pollen grains would probably be similar. Applying the
analogy to the record of Cretaceous plants, if half of a given type disappears, how

would one know that moist habitats were more disturbed than dry ones?

Jarzen: Ecological conclusions cannot be drawn on the basis of some pollen types. For
example, pollen of the lily family is abundant in Cretaceous strata. However, modern
lilies occur in northern and equatorial latitudes, and the occurrence alone of the
family in Cretaceous strata would not provide grounds for environmental speculation.
Fortunately other pollen-types can be related to genera or to tribes within families
which do have restricted ecological requirements. Among these, Gumnera, Pandanus and
Nyssa may be cited. Ecological conclusions are drawn based on the occurrence of
these forms.

Graham (1976) demonstrated that in tropical areas insect-pollinated plants are
well represented in the pollen record. Outwashing of tropical rains around fresh or
brackish water basins carry insect-pollinated pollen in abundance into the sedimentary
record. For example the pandans are largely insect-pollinated, at least in those
forms that have so far been studied. Yet in some sections in western Canada 68% of
the pollen belongs to Pandanus, while the wind-pollinated pines and podocarps are
poorly represented. We can postulate that this basin was receiving a great outwash of

pandan pollen because the pandans were growing in close proximity and were washed right

119

into the section. Palynofossils are often abundantly preserved, and in great
morphological diversity. Some can be identified, but many in every sample cannot be

accurately identified at present.

Russell: Based on the palynofloral record, what conclusions can be drawn concerning the

nature of the stresses on the Cretaceous-Tertiary boundary?

Jarzen: The stresses cannot at present be identified by the pollen. There is a general
trend across the boundary towards greater diversity and numbers of pollen typical of

wind-pollinated plants, but there are exceptions.

McLean: Are you referring to the boundary coal of Brown, or the horizon of palynofloral

change of Leffingwell?

Jarzen: The reference horizon is the Number 1 coal in western Canada.

Béland: Some statements have appeared to the effect that "...but land plants do not show

any change across the boundary." This suggestion has not been convincingly supported.

Jarzen: Land plants do not show evidence of drastic extinctions across the Z coal

boundary, but they do show evidence of change.

Béland: The number of species represented in the sampled sections is small relative to
the perhaps 500 species then in existence in the vicinity of the depositional site.

The sampling may be inadequate to reveal important changes in diversity.

Jarzen: Two hundred palynomorph species can be identified in one section. The total
number of identified late Cretaceous terrestrial and marine palynomorph species is in

the thousands. How many species of dinosaurs occur in one section?
Russell: Would it be useful to examine pollen residues for their trace element content?

Rucklidge: Our method could detect the presence of anomalies of platinum and iridium in
this material. Unaltered spores should be analyzed from a different locality as a
control because there is always a fear of contamination from other components in the

clay, particularly when it is known to contain trace element anomalies.

120

Béland: If the plants that were producing pollen at the end of the Cretaceous stopped

growing and producing pollen for two years, could this be detected?

Hsu: Intervals of two years cannot normally be resolved in the sedimentary record.

Béland: The effects on herbivore populations could nonetheless be catastrophic.

Pirozynski: Parasitic insects and fungi should show similar declines. Pollinating insects
would also be unable to survive a two-year halt in flower production. After their

extinction, wind would be the sole remaining pollination mechanism.

Béland: Insects are an extremely important component in terrestrial ecosystems, but their

fossil record is rudimentary.

Hsu: If the impacting projectile had a significant nitrogen component, perhaps the cosmic

nitrogen 15 to 14 ratio could be detected in pollen and spore residues.

W. Alvarez: Would the signal not be overwhelmed by that of atmospheric nitrogen?

Hsu: It might be detectable in remains of aquatic organisms if the nitrogen were mixed

directly into the water without passing through the atmosphere.

W. Alvarez: With respect to the problem of the apparent absence of turbidites associated
with the Cretaceous-Tertiary boundary, it is my experience in the Italian sections
that fossil zones are commonly missing either immediately below or above the extinction
horizon. In probably half of the outcrops, the boundary is not preserved. The same
seems to be true of deep-sea drilling cores. This would appear to support a violent

physical event of some kind in the oceans, but the absence of turbidites is puzzling.

Smit: Nevertheless in the deepest basins neither a hiatus or a turbidite can be found.
There is no evidence of any violent sedimentological event. Pelagic sections in
Europe and in many deep-sea cores show a significant decrease in sedimentation rates
from the Cretaceous to the Paleocene, beginning at the Cretaceous-Tertiary boundary.
The detrital component, or the clay which has been derived from the continent, also
accumulates less rapidly. At Zumaya, Gubbio and Caravaca, sedimentation rates fall by

a factor of 3 to 10, implying a significant decrease in continental rates of erosion.

IAAL

Continental sections should show changes which reflect a decreased rate of sediment

transport to ocean basins.

Russell: This does not seem to have occurred in the northern interior of North America,
where, if anything, rates of deposition increase. Certainly the grain-size of clastic
sediments increases and lignitic deposits become more abundant above the Cretaceous-
Tertiary boundary. It would be interesting to learn if similar changes occur at this
time in comparable sedimentary environments in Argentina. Are the changes due to

deforestation in adjacent highlands?

Smit: Perhaps it was due to a change in the balance between herbivores and the vegetation.
When herbivores are removed on a large scale, such as the removal of elephants from

stressed systems in eastern Africa, the vegetative cover becomes heavier.

W. Alvarez: Drs. Emiliani, Krause and Shoemaker note in their manuscript that the damage
in the Aquilapollenites province surrounding the North Pacific may identify the area
of impact of the object. Perhaps the Mediterranean cores were shielded from oceanic

disturbance.

Smit: Turbidites might be expected to occur around the North Pacific basin.

Thierstein: Many Pacific cores contain long range hiatuses extending from various horizons
in the late Cretaceous into the Paleocene, but in most that I have examined the hiatus

terminates at the very base of the Paleocene and the latest Cretaceous is missing.

Hsu: At the Tunguska site the trees were burned within a small area, but if a larger
cometary object impacted, would the effects account for the damage observed in boreal

terminal Cretaceous floras?

Dilcher: It would be difficult to explain floral changes over the entire boreal zone by
the impact of an object in the middle of one of two floral provinces existing at
that time. Dr. Hickey suggests that the changes are due to a withdrawal of an interior
sea that divided North America, which allowed the two floral provinces to mix and
elements within them to compete. This explanation seems to be most plausible. Around

the margins of an impact area one would expect reasonable survival rates.

122

Feldman: Rainfall may be important, as indicated in the manuscript by Dr. Emiliani and
his associates, in redepositing atmospheric dust. Is enough known from the
palynological record to identify regions of high rainfall? Can a link be established
between climatic patterns 65 million years ago and the removal of dust from the

atmosphere?

Dilcher: Yes, sedimentation of the material with rain would take place over a geologically
very short period of time. No physical evidence of a resulting pattern has been
identified in terrestrial strata. I find the most convincing arguments in favour of
a catastrophic event at the Cretaceous-Tertiary boundary to be the noble metal
fingerprinting and the extinction of marine micro-organisms. Terrestrial evidence is

much more ambiguous.

Hsu; Poisons introduced into the air in the vicinity of a mercury mine produce devastating
effects on animals and plants in the neighbourhood. The effects could be extrapolated

to the scale of a floral province.

Dilcher: During terminal Cretaceous time the western floral province was larger than the
boundaries that separated it from the eastern floral province. Its centre may have
been devastated but the margins would remain nearly intact. If the latter were also
devasted the devastation would have spread right into the eastern province as well.
In fact, the eastern province suffered less. A number of plants such as the palms,
and ancestral birches and oaks survive. Among the groups becoming extinct in the
eastern province is an interesting plant called Liriophyllum. A number of mid-
Cretaceous plants survive up to the Cretaceous-Tertiary boundary and become extinct.

There are extinctions in the eastern province as well.

McLean: Do you refer to the Z coal or the level of palynofloral change as the Cretaceous-

Tertiary boundary?

Dilcher: The forms cited occur in eastern North America, but the boundary there could

probably be correlated with some horizon of palynofloral change in the western interior.

Hsü: Were the extinct forms insect-pollinated?

Dilcher: Both insect- and wind-pollinated forms became extinct at the boundary.

123

Russell: When we were in East Africa several years ago, it was interesting to observe

how large vertebrates bioturbated alluvial sediments. Along the banks of the Luangwa
in Zambia, or around Tendaguru Hill in Tanzania the effects of the passage of elephant
through soft substrate were profound. In similar Jurassic deposits at Tendaguru a
string of tail vertebrae of a brontosaur was interrupted by a footprint of another
dinosaur. One animal was mired upright and died with a forefoot plunged two meters
into the substrate. In the Hell Creek Formation in the western interior, the strata
also have a massive appearance, and do not become finely laminated until after the
dinosaurs disappear from the sedimentary record. The lithologic change may well be
linked to the absence of dinosaurs. Fossil leaves are easily recovered from the
fissile strata of basal Paleocene age. The absence of lamination in the older strata
containing dinosaur bones has discouraged paleobotanists from attempting to recover

leaf fossils from the floras which predate the extinctions, to our loss.

McLean: There are always preservational problems, and only small amounts of the pollen

that is produced are ever preserved. Many fossil palynomorphs resemble modern ones.
However, many living plants produce the same kind of pollen grains, and identifications
of fossil pollens are uncertain. Compounding these uncertainties are large gaps in

the fossil record near the Cretaceous-Tertiary boundary in the western interior. Yet
Nichols and Ott (1978) have not been able to detect major changes in the palynofloral

record.

Russell: Within the Hell Creek and Bug Creek areas of eastern Montana where might the

large gaps be seen on the Cretaceous-Tertiary boundary?

McLean: What formula should be used to define the Cretaceous-Tertiary contact?

Russell: Use whichever one you like.

McLean: This raises the need for good radiometric dates over a broad area. Using the

Z coal as the boundary defines it on the basis of tectonic change. The palynofloral

change some 20 meters below probably reflects the presence of a hiatus.

Russell: The deposition of the marine Cannonball Formation in North Dakota spanned at

124

least a large part of the two-million-year duration of the Globigerina pseudobulloides

zone (Fox and Olsson 1969). The Cannonball is separated by a thin layer of lignite
from a bentonite which is at the top of the Hell Creek Formation and the highest
occurrence of the Triceratops dinosaur fauna. The pseudobullotdes fauna follows the
eugubina zone which occurred immediately after a marine extinction and is also in close
physical proximity to the highest terrestrial dinosaur fauna in the interior of

North America. This biostratigraphic evidence implies simultaneity in the marine and

terrestrial extinctions.

McLean: The resolution is inadequate to discriminate short-time intervals.

W. Alvarez: Radiometric dates from 65 million years ago also have plus and minuses
associated with them of one to two million years. This is one standard deviation, not
the absolute range. Resolution in that part of the time-scale is better with

paleomagnetic and micropaleontologic techniques.

W. Tucker: It would be useful to be better informed about other extinctions so statistics
for more than one event would be available. Does an extinction event have to happen
more than five times in five hundred million years, or does it just have to happen
once? What are the time intervals between extinctions? Are these extinctions similar
or is there a spectrum of extinctions? Do some happen with much less intensity than
others? This information would enable us to examine the cosmic setting with

statistical means to see which hypotheses are plausible.

Hsu: Two extinctions were missing from the list you cited earlier. One was at the end
of the Triassic about 200 million years ago, when the ammonites almost died out, and
the other was at the end of the Eocene about 38 million years ago. These are almost

as important as the terminal Devonian and terminal Ordovician extinctions.

Clube: Possible declines in groups of organisms prior to the extinction event have not
been discussed extensively, but have been cited as evidence against the kind of
astronomical model that we have proposed. However, if one giant asteroid impacted and
caused a major extinction event, it would have also been associated with comets and
debris from outside the Solar System, or a gathering of lower-mass objects. It is

statistically very unlikely that the huge object would be the first one to impact, and

125

an array of earlier impacts of smaller objects would inflict some damage. It would be
very interesting to examine the nature of the prolonged declines that occur prior to

major events.

Kyte: What kind of time-scale do you expect in the passing through a cloud or collection

of such objects?

Clube: Basically one should consider the relaxation times in the Apollo asteroids, or time-

scales of several millions of years.

Smit: The paleontological record could accommodate perhaps 10 thousand years but certainly

not one million years.

Thierstein: This applies to deep, open-ocean plankton in the Northern Hemisphere. Some
paleontologists who work on molluscs, such as Kauffman (1979) and Wiedmann (personal
communication), maintain the view that there is a gradual decline in both ammonite
and inoceramid diversity through the final 10 to 20 million years of Cretaceous time.
The shallow-water evidence indicates that the problem there is the sample control and

transgressive - regressive control.

McLean: It was pointed out that Aqutlapollenites was declining for quite a long time
before the end of the Cretaceous. This is why a broad view is necessary. Start many
years before the Cretaceous-Tertiary transition, see what the world was doing; then
trace the pattern of events up to and through the extinction interval. The events will

be much more easily understood than if the extinctions are treated as an isolated event.

Smit: Throughout the history of the ammonites, at any particular time there is roughly a
50% chance they are either in a decrease or in an increase. The trends may have no

relation to the boundary.

McLean: When the number of genera are counted through time with histograms, ammonite

diversity was stepping down like a staircase.

Smit: This was during late Cretaceous time; at the beginning of the Cretaceous they were

increasing.

126

Russell: In the long view some groups are always waxing and some are always waning.
What makes the subject more controversial is that abrupt extinctions occurred in
some groups when they were already in a decline. An example are the ichthyosaurs.
Their remains occur in terminal Cretaceous deposits, but they are much more abundantly

preserved in Triassic and Jurassic strata.

127

REFERENCES

Aggarwal, H.R. and V.R. Oberbeck. 1974. Roche limit of a solid body. Astrophysical Journal
191:577-588.

Alvarez, L.W., W. Alvarez, F. Asaro and H.V. Michel. 1980. Extraterrestrial cause for the
Cretaceous-Tertiary extinction. Science 208:1095-1108.

Alvarez, W., L.W. Alvarez, F. Asaro and H.V. Michel. 1979A. Experimental evidence in support
of an extra-terrestrial trigger for the Cretaceous-Tertiary extinctions. EOS, 60:734.

Alvarez, W., L.W. Alvarez, F. Asaro and H.V. Michel. 1979B. Anomalous iridium levels at the
Cretaceous-Tertiary boundary at Gubbio Italy: negative results of tests for a supernova
origin. Geological Society of America. Abstract. Programs II:350.

Alvarez, W., M.A. Arthur, A.G. Fischer, W. Lowrie, G. Napoleoni, I. Premoli-Silva and
W.M. Roggenthen. 1977. Upper Cretaceous-Paleocene magnetic stratigraphy at Gubbio, Italy
V type section for the Late Cretaceous-Paleocene geomagnetic reversal time scale.
Geological Society of America Bulletin 88:383-389.

Alvarez, W., F. Asaro, L.W. Alvarez, H.V. Michel, M.A. Arthur, W.E. Dean, D.A. Johnson,
M. Kastner, F. Maurrasse, R.R. Revelle and D.A. Russell. New data on the Cretaceous-
Tertiary extinction. Abstract to AAAS meeting in Washington D.C. January 1982.

Anders, E., R. Ganapathy, U. Krahenbuhl and J.W. Morgan. 1973. Meteoritic material on the
moon. The Moon 8:3-24.

Arculus, R.J. and J.W. Delano. 1981. Siderophile element abundances in the upper mantle:
evidence for a sulfide signature and equilibrium with the core. Geochimica et
Cosmochimica Acta 45:1331-1343.

Arndt, N.T., D. Francis and A.J. Hynes. 1979. The field characteristics and petrology of
Archaean and Proterozoic Komatiites. Canadian Mineralogist 17:147-163.

Asaro, F., H.V. Michel, L.W. Alvarez and W. Alvarez. Results of a dating attempt - chemical
and physical measurements relevant to the cause of the Cretaceous-Tertiary extinctions.
Proceedings of the ACS Symposium on "Nuclear and Chemical Dating Techniques" March 23-28,
1980, Lawrence Berkeley Laboratory Unpublished Report LBL-11613 September (1980).

Baltz, E.H., S.R. Ash and R.Y. Anderson. 1966. History of nomenclature and stratigraphy of
rocks adjacent to the Cretaceous-Tertiary boundary, western San Juan Basin, New Mexico.
United States Geological Survey Professional Paper 524D:D1-D23.

Barker, J.L. and E. Anders. 1968. Accretion rate of cosmic matter from iridium and osmium
contents of deep sea sediments. Geochimica et Cosmochimica Acta 32:627-645.

Blankley, W.F. 1971. Auxotrophic and heterotrophic growth and calcification in
coccolithophorids. Doctoral dissertation, University of California at San Diego. 186 p.

Boersma, A., N. Shackleton, M. Hall and Q. Given. 1979. Carbon and oxygen isotope records at
DSDP 384 (North Atlantic) and carbon isotope variations in the Atlantic Ocean. Initial
reports of the Deep Sea Drilling Project 43:695-718.

Bondi, H., and Hoyle, F. 1944. On the mechanism of accretion by stars. Royal Astronomical
Society of London. Monthly Notices 104:273-282.

Bromley, R.G. 1979. Chalk and bryozoan limestone: facies, sediments and depositional
environments. In: Birkelund, T., and R.G. Bromley (eds.): Cretaceous-Tertiary boundary

events symposium. I. Proceedings. University of Copenhagen :16-32.

Brown, R.W. 1962. Paleocene flora of the Rocky Mountains and Great Plains. United States
Geological Survey Professional Paper 375, 119 p.

128

Brownlee, D.E., D.A. Tomandl and E. Olszewski. 1977. Interpianetary dust; a new source of
extraterrestrial material for laboratory studies. Proceedings 8th Lunar Planetary Science

Conference : 149-160.

Bujak, J.P. and G.L. Williams. 1979. Dinoflagellate diversity through time. Marine
Micropaleontology 4:1-12.

Butler, R.F., E.H. Lindsay, L.L. Jacobs and N.M. Johnson. 1977. Magnetostratigraphy of the
Cretaceous-Tertiary boundary in the San Juan Basin, New Mexico. Nature 267:318-323.

Carpenter, K. 1979. Vertebrate fauna of the Laramie Formation (Maestrichtian) , Weld County,
Colorado. Contributions to Geology, University of Wyoming 17:37-49.

Chamley, H. and C. Robert. 1979. Late Cretaceous to early Paleocene environmental evolution
expressed by the Atlantic clay sedimentation. In: Christensen, W.K. and T. Birkelund
(eds.): Cretaceous-Tertiary boundary events. II. Proceedings. University of Copenhagen
:71-77.

Chandrasekharam, D. and A.N. Parthasarathy. 1978. Geochemical and tectonic studies on the
coastal and inland Deccan Trap volcanics and a model for the evolution of Deccan Trap
volcanism. Neves Jahrbuch fur Mineralogie Abhandlungen 132:214-229.

Chou, C-L. 1978. Fractionation of siderophile elements in the earth's upper mantle.
Proceedings 10th Lunar Planetary Science Conference :219-230.

Clube, S.V.M. and W.M. Napier. (in press). The role of episodic bombardment in geophysics.
Earth and Planetary Science Letters.

Corliss, B.H. and C.D. Hollister. 1979. Cenozoic sedimentation in the central North Pacific.
Nature 282:707-709.

Costa, C. and H.H. Mollenhauer. 1980. Carcinogenic activity of particulate nickel compounds
is proportional to their cellular uptake. Science 209:515-517.

Cowan, J.J., A.G.W. Cameron and J.W. Truran. (in press). The thermal runaway r-process.
Astrophysical Journal.

Crocket, J.H., R.R. Keays and S. Hsieh. 1967. Precious metal abundances in some carbonaceous
and enstatite chondrites. Geochimica et Cosmochimica Acta 31:1615-1623.

Crocket, J.H. and H.Y. Kuo. 1979. Sources for gold, palladium and iridium in deep-sea
sediments. Geochimica et Cosmochimica Acta 43:831-842.

Danilchenko, P.G. 1978. Fishes, suborder Teleostei. In: Soloviev, A.N. and V.N. Shimansky:
Evolution and change of the organic realm at the Mesozoic-Cenozoic boundary. Nauka,

Moscow :17-44.

Doyle, P. 1980. Seabed disposal program: improvement of ichthyolith stratigraphy for giant
piston core 3. Progress Report for 1979-1980 for Sandia Laboratories, 14 p.

Dudley, W.C. and D.E. Goodney. 1979. Oxygen isotope content of coccoliths grown in culture.
Deep-Sea Research 26A:495-503.

Ehmann, W.D. and D.E. Gillum. 1972. Platinum and gold in chondritic meteorites. Chemical
Geology 9:1-11.

Elsasser, W.M., P. Olson and B.D. Marsh. 1979. The depth of mantle convection. Journal of
Geophysical Research 84, number B1:147-155.

Erben, H.K., J. Hoefs and K.H. Wedepohl. 1979. Paleobiological and isotopic studies of
eggshells from a declining dinosaur species. Paleobiology 5:380-414.

Fraundorf, P. 1981. Interplanetary dust in the transmission electron microscope: diverse
materials from the early solar system. Geochimica et Cosmochimica Acta 45:915-943.

129

Ganapathy, R. 1980. A major meteorite impact on the earth 65 million years ago: evidence
from the Cretaceous-Tertiary boundary clay. Science 209:921.

Ganapathy, R., S. Gartner and M. Jiang. 1981. Iridium anomaly at the Cretaceous-Tertiary
boundary in Texas. Earth and Planetary Science Letters 54:393-396.

Giauque, R.D., F.S. Goulding, J.M. Jaklivic and R.H. Pehl. 1973. Trace element determination
with semiconductor detector x-ray spectrometers. Analytical Chemistry 45:671-681.

Glass, B.P. 1969. Silicate spherules from the Tunguska impact area: electron microprobe
analysis. Science 164:547-549.

Glezer, Z.I. 1977. Silicoflagellates. In: Vakhrameev, V.A. (ed.): Evolution of floras at the
Mesozoic-Cenozoic boundary. Nauka, Moscow. (in Russian) :22-25.

Graham, A. 1976. Late Cenozoic evolution of tropical lowland vegetation, Veracruz, Mexico.
Evolution 29:723-735.

Greenberg, J.M. 1974. The interstellar depletion mystery, or where have all those atoms gone?
Astrophysical Journal 189:L81-L85.

Grieve, R.A.F. and M.J. Cintala. (in press). A method for estimating the initial conditions of
terrestrial cratering events, exemplified by its application to Brent crater, Ontario.
Proceedings 12th Lunar and Planetary Science Conference.

Hakansson, E. and J.M. Hansen. 1979. Guide to Maastrichtian and Danian boundary strata in
Jylland. In: Birkelund, T., and R.G. Bromley (eds.): Cretaceous-Tertiary boundary
events. I. The Maastrichtian and Danian of Denmark. University of Copenhagen :171-188.

Hillebrandt, A. von. 1974. Bio-estratigrafia del Paleogeno en el sureste de Espdna (Prov. de
Murcia y Alicante). Cuadernos de Geologia. Universidad de Granada. 5:135-153.

Hoefs, J. 1973. Stable Isotope Geochemistry. Springer-Verlag, Berlin 140 p.

Hsu, K.J. 1980. Terrestrial catastrophe caused by cometary impact at end of Cretaceous.
Nature 285:201-203.

Hsu, K.J., Q. He, J.A. McKenzie, H. Weissert, K. Perch-Nielsen, H. Oberhansli, K. Kelts,
J. LaBrecque, L. Tauxe, U. Krahenbuhl, S.F. Percival, R. Wright, R.Z. Poore, A.M. Gombos,
K. Pisciotto, A.M. Karpoff, N. Petersen, P. Tucker, M.F. Carman Jr., and E. Schreiber.
(in press). Environmental and evolutionary consequences of mass-mortality at the end of
Cretaceous. Science.

Hsu, K.J. and J. LaBrecque. (in press). Initial reports of the Deep Sea Drilling Project 73.

Hughes, D.W. and P.A. Daniels. 1980. The magnitude distribution of comets. Royal Astronomical
Society of London. Monthly Notices 191:511-520.

Jarzen, D.M. 1977A. Angiosperm pollen as indicators of Cretaceous-Tertiary environments.
Syllogeus 12:39-49.

- 1977B. Aquilapollenites and some santalalean genera — a botanical comparison. Grana
16:29-39.

Javoy, M. and F. Pineau. 1978. Isotope geochemistry of deep seated carbon. United States
Geological Survey Open-File Report 78-701:203.

Junge, C.E., C.W. Chagnon and J.E. Manson. 1961. Stratospheric aerosols. Jouinal of
Meteorology 18:81-108.

Kallemeyn, G.W. and J.T. Wasson. 1981. The compositional classification of chondrites - I.
The carbonaceous chondrite groups. Geochimica et Cosmochimica Acta 45:1217-1230.

130

Kaminskiy, F.V., Frantsesson, Ye. V. and V.P. Khyostoya. 1974. First data on metals of
platinum group (Pt, Pd, Rh, Ir, Os) in kimbertitic rocks. (in Russian) Doklady.
Academy of Sciences of the USSR. Earth Science Sections 219:204-207.

Kauffman, E.G. 1979. The ecology and biogeography of the Cretaceous-Tertiary extinction event.
In: Christensen, W.K. and T. Birkelund (eds.): Cretaceous-Tertiary boundary events
symposium. II. Proceedings. University of Copenhagen :29-37.

Keller, G.H. and N.L. Douglas. 1980. Variation of sediment geotechnical properties between
the greater Antilles Outer Ridge and the Nares Abyssal Plain. Marine Technology
4:125-143.

Kent, D.V. 1977. An estimate of the duration of the faunal change at the Cretaceous-Tertiary
boundary. Geology 5:769-771.

Kerr, R.A. 1980. Asteroid theory of extinctions strengthened. Science 210:514-517.
- 1981. Satellites of asteroids coming into vogue. Science 211:1333-1336.

Korde, K.B. 1977A. Charophytic algae. In: Vakhrameev, V.A. (ed.): Evolution of floras
at the Mesozoic-Cenozoic boundary. Nauka, Moscow (in Russian) :26-30.

- 1977B. Dacycladacean algae. In: Vakhrameev, V.A. (ed.): Evolution of floras at the
Mesozoic-Cenozoic boundary, Nauka, Moscow (in Russian) :31-38.

Krahenbuhl, U., J.W. Morgan, R. Ganapathy and E. Anders. 1973. Abundance of 17 trace elements
in carbonaceous chondrites. Geochimica et Cosmochimica Acta 37:1353-1370.

Kresak, L'. 1978. The Tunguska object: a fragment of comet Encke? Astronomical Institutes
of Czechoslovakia. Bulletin 29:129-134.

Krinov, E.L. 1966. Giant meteorites. Pergamon Press, Oxford, New York and Toronto :264.

Kues, B.S., T. Lehman and J.K. Rigby Jr. 1980. Teeth of Alamosaurus sanjuanensts, a late
Cretaceous sauropod. Journal of Paleontology 54:864-869.

Kyte, F.T., Z. Zhou and J.T. Wasson. 1980. Siderophile-enriched sediments from the Cretaceous-
Tertiary boundary. Nature 288:651-656.

Kyte, F.T., Z. Zhou and J.T. Wasson. 1981. High noble metal concentrations in a late Pliocene
sediment. Nature 292:417-420.

Lang, K.R. 1980. Astrophysical Formulae, 2nd Edition. Springer-Verlag, Berlin 65,

Leffingwell, H.A. 1971. Palynology of the Lance and Fort Union formations of the type Lance
area, Wyoming. Geological Society of America Special Paper 127:1-61.

Lerbekmo, J.F., M.E. Evans and H. Baadsgaard. 1979. Magnetostratigraphy, biostratigraphy and
geochronology of Cretaceous-Tertiary boundary sediments, Red Deer Valley. Nature 279:
26-30.

Lindsay, E.H., R.F. Butler and N.M. Johnson. 1979. Reply to comment. Geology 7:326-327.

Lindsay, E.H., L.L. Jacobs and R.F. Butler. 1978. Biostratigraphy and magnetostratigraphy of
Paleocene terrestrial deposits, San Juan Basin, New Mexico. Geology 6: 425-429.

Lupton, C., D. Gabriel and R.M. West. 1980. Paleobiology and depositional setting of a late
Cretaceous vertebrate locality, Hell Creek Formation, McCone County, Montana.

Contributions to Geology, University of Wyoming 18:117-126.

Manabe, S. and R.T. Wetherald. 1980. On the distribution of climatic change resulting from an
increase in C02 content of the atmosphere. Journal of Atmospheric Science 37:99-118.

Masailis, V.L. 1976. Astroblemes in the U.S.S.R. International Geological Review 18:1249-1258.

131

Masailis, V.L., A.N. Danilin, M.S. Masak, A.I. Raikhlin, T.V. Selivanovskaya and
G.M. Sedenkov. 1980. Geology of Astroblemes. Nedra, USSR. (in Russian) 231 p.

Mason, B. 1963. The carbonaceous chondrites. Space Science Reviews 1:621-646.

. 1971. Handbook of elemental abundances in meteorites. Gordon and Breach, New York.
555 p.

McCrea, W.H. 1981. Long time-scale fluctuations in the evolution of the earth. Proceedings
of the Royal Society of London, Series A, 375:1-41.

McKay, C. and D.H. Milne. 1980. Quick response to pelagic marine food chains to a global
blackout. NASA Ames Ad Hoc Committee Preliminary Report (Unpublished). 3 p.

Michel, H.V., F. Asaro, W. Alvarez and L.W. Alvarez. 1981. Distribution of iridium and other
elements near the Cretaceous-Tertiary boundary in Hole 465A: preliminary results.
DSDP Initial Reports Site 465A.

Molnar, R.E. 1978. A new theropod dinosaur from the Upper Cretaceous of Central Montana.
Journal of Paleontology 52:73-82.

Moore, W.L. 1976. The stratigraphy and environments of deposition of the Cretaceous Hell
Creek Formation and the Paleocene Ludlow Formation, southwestern North Dakota.
North Dakota Geological Survey Report of Investigations 56: 40 p.

Morgan, W.J. 1972A. Plate motions and deep mantle convection. Geological Society of America
Memoir 132:7-22.

1972B. Deep mantle convection plumes and plate motions. American Association of
Petroleum Geologists Bulletin 56:203-213.

. 1981. Hot spot tracks and the opening of the Atlantic and Indian oceans. In:
Emiliani, C. (ed.): The oceanic lithosphere. Wiley-Interscience, New York
: 443-487.

Mossop, S.C. 1964. Volcanic dust collected at an altitude of 20 km. Nature 203:824-827.

Naidin, D.P. 1979. The Cretaceous/Tertiary boundary in the U.S.S.R. In: Christensen, W.R.
and T. Birkelund (eds.): Cretaceous-Tertiary boundary events. Volume II:188-201.

Naldrett, A.J., E.L. Hoffman, A.H. Green, C. Chou, S.R. Naldrett and R.A. Alcock. 1979.
The composition of Ni-sulfide ores, with particular reference to their content of PGE
and Au. Canadian Mineralogist 17:403-415.

Naldrett, A.J. and J.M. Duke. 1980. Platinum metals in magmatic sulfide ores. Science
208:1417-1424.

Napier, W.M. and S.V.M. Clube. 1979. A theory of terrestrial catastrophism. Nature 282:
455-459,

National Academy of Sciences. 1977. Medical and biologic effects of environmental
pollutants: platinum-group metals. Washington, D.C. :167 ff

Naylor, B.G. 1978. The earliest known Wecturus (Amphibia, Urodela), from the Paleocene
Ravenscrag Formation of Saskatchewan. Journal of Herpetology 12:565-569.

Ng, F.S.P. 1978. Strategies of establishment in Malayan forest trees. In: Tomlinson, P.B. and
M.H. Zimmerman (eds.): Tropical trees as living systems. Cambridge University Press,
Cambridge :129-162.

Nichols, D.J. and H.L. Ott. 1978. Biostratigraphy and evolution of the Momipites-

Caryapollenttes lineage in the early Tertiary in the Wind River Basin, Wyoming.
Palynology 2:93-112.

132

O'Keefe, J.D. 1981. Impact mechanics of the Cretaceous-Tertiary extinction (abstract).
Lunar and Planetary Science 12:785-787.

Oltz, D.F. 1969. Numerical analyses of palynological data from Cretaceous and early Tertiary
sediments in east central Montana. Palaeontographica 128B:90-166.

Opik, E.J. 1966. The dynamical aspects of the origin of comets. Proceedings 13th International
Astrophysical Symposium, Liege, July 1965, :523-572.

. 1972. Comments on Lunar origin. Irish Astronomical Journal 10:190-238.

Orth, C.J., J.S. Gilmore, J.D. Knight, C.L. Pillmore, R.H. Tschudy and J.E. Fassett. 1981.
Iridium anomaly at the palynological Cretaceous-Tertiary boundary in New Mexico (abstract
V 202). EOS 62:438.

Orth, C.J., J.S. Gilmore, J.D. Knight, C.L. Pillmore, R.H. Tschudy and J.E. Fassett.
(in press). An iridium abundance anomaly at the palynological Cretaceous-Tertiary
boundary in northern New Mexico. Science.

Palme, H., E. Gobel and R.A.F. Grieve. 1979. The distribution of volatile and siderophile
elements in the impact melt of East Clearwater (Quebec). Proceedings 10th Lunar
Planetary Science Conference :2465-2492.

Palme, H., R.A.F. Grieve and R. Wolf. (in press). Identification of the projectile at the
Brent crater and further considerations of projectile types at tefrestrial craters.
Geochimica et Cosmochimica Acta.

Pascoe, E. 1964. A manual of geology of India and Burma, III, Government of India Press,
Calcutta :1345-1416.

Penzias, A.A. 1980. Measurements of isotopic abundances in interstellar clouds. In:
Andrew, B.H. (ed.): Interstellar molecules, Proceedings of International Astronomical
Union Symposium 87, (Mt. Tremblant, P.Q., 6-10 August 1979), Reidel, Dordrecht-Holland
: 397-404.

Perch-Nielsen, K. 1979. Calcareous nannofossil zonation at the Cretaceous/Tertiary boundary
in Denmark. In: Birkelund, T. and R.G. Bromley (eds.): Cretaceous Tertiary boundary
events. I. The Maastrichtian and Danian of Denmark. University of Copenhagen :115-135.

. (in press). Les nannofossiles calcaires à la limite Crétacé-Tertiaire près de El Kef,
Tunisie. Cahiers de Micropaléontologie.

Pettijohn, F.J. 1975. Sedimentary rocks. Harper and Row, Publishers, New York. 628 p.

Premoli-Silva, I. 1977. Upper Cretaceous - Paleocene magnetic stratigfaphy at Gubbio, Italy
II Biostratigraphy. Geological Society of America Bulletin 88:371-374.

Rasmussen, H.W. 1979. Crinoids, asteroids and ophiuroids in relation to the boundary. In:
Birkelund, T. and R.G. Bromley (eds.): Cretaceous-Tertiary boundary events, Vol. I.
765-71.

Rees, M.J. 1978. Origin of pregalactic microwave background. Nature 275:35-37.

Romein, A.J.T. 1977. Calcareous nannofossils from the Cretaceous/Tertiary boundary interval
in the Barranco del Gredero (Caravaca, Prov. Murcia, S.E. Spain). I. Roninklijke
Nederlandse Akademie van Wetenschappen Proceedings, Series B, 80:256-278.

Russell, D.A. 1977. The biotic crisis at the end of the Cretaceous period. Syllogeus 12:11-23.

Seyfert, C.K. and L.A. Sirkin. 1979. Earth history and plate tectonics. Harper and Row,
New York. 600 p.

Schilling, J.G. 1973. Afar mantle plume; rare earth evidence. Nature 242:565-571.

133

Shoemaker, E.M., J.G. Williams, E.F. Helin and R.F. Wolfe. 1979. Earth-crossing asteroids:
orbital classes, collision rates with Earth, and origin. In: Gehrels, T. (ed.):

Asteroids. University of Arizona Press, Tuscon :253-282.

Smit, J. 1981. A catastrophic event at the Cretaceous-Tertiary boundary. Doctoral
dissertation, Universiteit van Amsterdam, 138 p.

Smit, J. and J. Hertogen. 1980. An extraterrestrial event at the Cretaceous-Tertiary boundary.
Nature 285:198-200.

Sohl, N.F. 1960. Late Cretaceous gastropods in Tennessee and Mississippi. United States
Geological Survey Professional Paper 331, 344 p.

Stenestad, E. 1979. Upper Maastrichtian foraminifera from the Danish Basin. In: Birkelund, T.,
and R.G. Bromley (eds.): Cretaceous-Tertiary boundary events. I. The Maastrichtian and
Danian of Denmark. University of Copenhagen :101-107.

Stenzel, H.B. 1957. Nautilus. Geological Society of America Memoir 67, 1:1135-1142.

Talwani, M. and O. Eldholm. 1977. Evolution of the Norwegian-Greenland Sea. Geological Society
of America Bulletin 88:969-999.

Thierstein, H.R. (in press A). Late Cretaceous calcareous nannoplankton and the changes at the
Cretaceous-Tertiary boundary. In: Douglas, R.G., J. Warme and E.L. Winterer (eds.):
The Deep Sea Drilling Project: a decade of progress. Society of Economic Paleontologists
and Mineralogists, Special Publication.

- (in press B). The terminal Cretaceous extinction event and climatic stability. In:
Crowell, J.C. and W.H. Berger (eds.): Pre-Pleistocene Climates. Studies in geophysics.
National Research Council, Washington, D.C.

Thierstein, H.R. and W.H. Berger. 1978. Injection events in ocean history. Nature 276:461-466.

Thierstein, H.R., H.J. Brumsack and P.H. Roth. (in press). Mid-Cretaceous black shales and
ocean fertility. Geological Society of America, Abstracts with Programs.

Thierstein, H.R., K.R. Geitzenauer, B. Molfino and N.J. Shackleton. 1977. Global synchroneity
of late Quaternary coccolith datum levels: validation by oxygen isotopes. Geology
5: 400-404.

Tinsley, B.M. and A.G.W. Cameron. 1974. Possible influence of comets on the chemical
evolution of the Galaxy. Astrophysics and Space Science 31:31-35.

Tschudy, R.H. 1971. Palynology of the Cretaceous-Tertiary boundary in the northern Rocky
Mountain and Mississippi embayment regions. Geological Society of America Special Paper

127:65-111.

. 1975. Normapolles pollen from the Mississippi embayment. United States Geological
Survey Professional Paper 865:1-42.

Van Valen, L. 1978. The beginning of the age of mammals. Evolutionary Theory 4:45-80.
Wall, W.P. and P.M. Galton. 1979. Notes on pachycephalosaurid dinosaurs (Reptilia:Ornithischia)
from North America, with comments on their status as ornithopods. Canadian Journal of

Earth Sciences 16:1176-1186.

Wedepohl, K.H. (ed.). 1969-1970. Handbook of geochemistry. Springer-Verlag, New York.
2 volumes.

Weishampel, D.B. and J.A. Jensen. 1979. Parasaurolophus (Reptilia, Hadrosauridae) from Utah.
Journal of Paleontology 53:1422-1427.

134

Wetherill, G.W. 1971. Cometary versus asteroidal origin of chondritic meteorites. In:
Gehrels, T. (ed.): Physical studies of minor planeta. NASA SP-367, Washington, D.C.
:447-459,.

. 1976. Where do the meteorites come from? A re-evaluation of the Earth-crossing
Apollo objects as sources of chondritic meteorites. Geochimica et Cosmochimica Acta
40:1297-1317.

. 1979. Steady state populations of Apollo-Amor objects. Icarus 37:99-112.

Whetstone, K.N. 1978. A new genus of Cryptodiran turtles (Testudinoidea, Chelydridae) from
the Upper Cretaceous Hell Creek Formation of Montana. University of Kansas Science
Bulletin 51:539-563.

Wilbur, K.M. and N. Watabe. 1963. Experimental studies on calcification in molluscs and the
alga Coccolithus huxleyi. Annals New York Academy of Science 109:82-112.

Williams, G.L. 1978. Dinoflagellates, acritarchs and tasmanitids. In: Hag, B.U. and
A. Boersma (eds.): Introduction to marine micropaleontology. Elsevier, New York.
2293-326.

Wolfenson, D., Y.F. Frei, N. Snapir and A. Berman. 1979. Effect of diurnal or nocturnal
heat stress on egg formation. British Poultry Science 20(2) :167-174.

Wuk, H.B. 1956. The chemical composition of some stony meteorites. Geochimica et Cosmochimica
Acta 9:279-289.

. 1963. Unpublished data reported by B. Mason.
Wyllie, P.J. and W.L. Huang. 1976. High CO2 solubilities in mantle magmas. Geology 4:21-24.

Yabushita, S. 1979. A statistical study of the evolution of the orbits of long-period comets.
Royal Astronomical Society of London. Monthly Notices 187:445-462.

135

CONTRIBUTED ABSTRACTS

Episodic Catastrophism
S.V.M. Clube

I think it might be helpful if I say a few words about the origin of the theory of
terrestrial catastrophism!. There may be a tendency to think the astronomy has grown in an
ad hoc way out of the geochemical evidence. This is not so. In fact, for several years
now, Various problems in galactic evolution have highlighted the importance of testing
whether spiral arms should be treated as material bodies or as density waves? à 5,

It is known for example that spiral arms house giant molecular clouds and these are the
natural places for making comets — but in the density wave theory unlike material arm
theory, the comets would have to drift away from behind the spiral arms and fill the disc
of the galaxy more or less uniformly. Since the Solar System captures comets as it moves
through the galaxy and there have for some years been good reasons for believing comets
produce Apollo asteroids and Apollo asteroids through impacts produce terrestrial
catastrophes, a history of "Episodic catastrophism" (controlled by spiral arm passages) as
opposed to "regular catastrophism" (controlled by purely random encounters) was evidently
a means of distinguishing whether the galaxy contained material arms or density waves.

The proposition that Solar System comets may be recent captures from interstellar
space contradicts present views as to the origin of comets and we were led to review the
status of the so-called Oort Cloud. When the primordial Oort Cloud was first postulated®,
the existence of giant molecular clouds was not known, but as Dr. Napier will show! 5 7,
encounters with these molecular clouds are very damaging to any primordial Cloud and we now
know it is bound to be stripped away. So where do the existing long period comets come

8 that the comets could have been

from? As it happens, recent calculations had shown
captured into the Solar System ~ 10 Myr ago, and this idea turns out to provide a better
explanation of the high mass members of the Apollo asteroid population (as dead comets) than
the older idea, already in some difficulty, that they are meteorites come from the asteroid
belt. So, in the light of the theory already developed, we raised the question whether
there is spiral arm material that the Solar System went through ~ 10 Myr ago? And, as it

happens, there is: there is in the solar neighbourhood a large conspicuous spiral arm

feature known as Gould's Belt which the Solar System entered some 20 Myr ago and emerged

136

from ~ 10 Myr ago. So, mindful of the possibility that impacts may trigger glaciations, and
the knowledge that recent terrestrial history through 10-20 Myr has the character of a major
ice-age, we came to the conclusion that there was a basis in both fact and theory for
proposing "Episodic Catastrophism". Catastrophic extinctions were predicted for large
Apollo asteroids with simultaneous deposition of interstellar signatures, and the discovery
of such effects at the K-T boundary evidently bears great promise for galactic theory.

I would like to emphasise however, that the theory quite naturally produces meteorites
at the low mass end of the Apollo population at the same time as it does dead comets at
the high mass end since the former are derived from the remnants of the Solar System
primordial population which are gravitationally perturbed into such orbits as a result of
passing close to an interstellar cloud. We have tended therefore to take the view that the
proof of meteorite?, comet!% 11, or Apollo asteroid! for any particular impact producing
extinctions like the K-T is not really a fundamental problem. It is the pattern of impacts
that is more fundamental; indeed it is important now to be recognising that impacts are an
inevitable and (statistically) predictable part of the astronomical scene and to be asking
whether they are primary agents in producing the stratigraphic record generally or whether

they are minor perturbations on events that are largely of pure terrestrial origin}.

References

1 W.M. Napier and S.V.M. Clube. 1979. Nature 282:455.

2 §.V.M. Clube. 1973. Mon. Not. R. astr. Soc. 161:455.

3 S.V.M. Clube. 1978. Vistas Astr. 22:77.

4+ F.G. Watson and S.V.M. Clube. (submitted). Mon. Not. R. astr. Soc.
S §.V.M. Clube and W.M. Napier. (submitted). Q.J.R. astr. Soc.

6 J.H. Oort. 1950. Bull. Astr. Inst. Neth. 11:91.
W.M. Napier and M. Staniucha. (in press). Mon. Not. R. astr. Soc.
8 S. Yabushita. 1979. Mon. Not. R. astr. Soc. 187:445.

9 L.W. Alvarez, W. Alvarez. F. Asaro & H.V. Michel. 1980. Science 208:1095.
10 K.J. Hsu. 1980. Nature 285:201.
11 F.T. Kyte, Z. Zhou & J.T. Wasson. 1980. Nature 288:651.

12 S.v.M. Clube & W.M. Napier. (submitted). Earth Plan. Sci. Lett.

137

Looking Back on the Tunguska Event
Ian Halliday

When we consider the possible effects of a major impact on the Earth some 65 million
years ago it is probably worth spending a few moments to review what is known about the most
violent of recent impact events, the Tunguska explosion of June 30, 1908. The popular news
media enjoy circulating dramatic stories attributing the Tunguska event to the current
scientific (or science fiction) novelty -- antimatter, nuclear accidents on alien spacecraft,
neutron stars, black holes — so it is important to realize that much concrete knowledge
exists. The early research on Tunguska occupies a large portion of the volume "Giant
Meteorites" by E.L. Krinov (1966, English edition) an excellent scientific book with the
flavour of an adventure story.

Krinov's account contains interviews with eyewitnesses to the 7 a.m. fireball which
rivalled the Sun, descriptions of the sounds, the blast of radiation and the pillar of fire
and smoke reported by more distant observers. Vague accounts of fallen trees and forest
fires were reported by native hunters but it was not until the second decade after the
event that the accounts were collected and expeditions organized to this remote area of
north-central Siberia.

Dust from the explosion caused two very bright nights over northern Europe which
persisted to some degree for two months. Seismic and microbaragraph records from across
Europe suggest an energy of 1022-1023 ergs for the explosion, which produced an active
seismic record for an hour at Irkutsk, 900 kilometers from the desolate site of the event.

Due to the tireless efforts of the meteorite expert, L.A. Kulik, early expeditions to
locate the site began in 1927. Great hardships were endured in penetrating the forest
with horse-drawn wagon trains but a large area of devastated forest was located and mapped,
in which the trees were blown down radially outwards from a central region. Later surveys
established the area of this damage as 1600 km2. The central area coincided with a large
swamp in permafrost and Kulik believed the numerous pits were caused by individual
meteorite fragments, so a trench was dug to drain one 32-meter diameter pit, but no
meteorites could be found. Today it is believed that the swamp was a pre-existing feature
and no sizable fragments from the incoming object survived to strike the ground. At least
four major expeditions were conducted before the Second World War (which claimed the life
of Kulik).

138

Later expeditions included aerial surveys, geophysical surveys and forest and soil
research. Of particular significance are collections of sub-millimeter spherules of
magnetite and silicates, first collected in 1930 but extensively mapped by subsequent
expeditions. They appear to define an area up to 200 kilometers from the epicentre, over
which ablation products were distributed by the wind, concentrated to the northwest of the
explosion site. Analysis of some of these particles yields an unexpectedly high ratio of
potassium to sodium. There seems to be little doubt that the large mass of the projectile
was completely disrupted at a height of about 5 to 7 kilometers and experimental models
using an explosion along a line source can produce a very similar pattern of blast damage.

Several attempts have been made to establish the path of the Tunguska meteoroid in
order to speculate on its orbit and origin. The possibility that it was a cometary
fragment was suggested by Kulik and Astapovich in the U.S.S.R. and by F.J.W. Whipple, an
English meteorologist, as early as the years of the first expeditions to the site. This
remains the best explanation and recently it received further support in a study by
Kresak (1978) who used a recent analysis of the visual observations to conclude that the
Tunguska object was probably a fragment from Encke's Comet.

Our knowledge of the chemistry of comets is still far from satisfactory. Some
qualitative information is available from cometary spectra and the spectra of shower
meteors, whose origin can be attributed to comets, but we are not in a position to say
much about the relative abundances of minor constituents in comets vs asteroids. There is
still debate as to whether some of the more fragile classes of meteorites may have had
their origins in comets, although the fate of the Tunguska object may be an indication of
how severe a filter the Earth's atmosphere is for moderately large chunks of cometary matter.
There will be differences in the details of the impact mechanism for high-density and
low-density projectiles but it is clear from Tunguska that large amounts of material may be

spread over great distances by major impact events.

References

Kresak, L., 1978. The Tunguska object: a fragment of Comet Encke? Bull. Astron. Inst.
Czechoslovakia, 29:129-134.

Krinov, E.L., 1966. Giant Meteorites, Pergamon Press, pages xxii + 397.

139

Evolutionary and Environmental Consequences of a Terminal Cretaceous Event

Kenneth J. Hsu

I started with the premise that the occurrence of a collision event with an extra-
terrestrial object is the only viable working hypothesis at the moment to explain the K-T
boundary. The questions to be explored are: 1) the nature of the object, and 2) the scenario
for selective extinction. During the conference, I presented the work done on deep sea
drilling cores from Site 524 in the South Atlantic (30°N, 3°E) by my colleagues Q. He,

J. McKenzie, K. Perch-Nielsen, H. Oberhänsli and U. Krähenbühl in Switzerland. The data
indicated the possibility of a collision event, which caused a mass-mortality of ocean life,
followed by catastrophic environmental stresses which caused the selective extinction of

marine calcareous planktons and of dinosaurs.

The K-T boundary at Site 524

The boundary is defined by nannofossil evidence on the basis of first common occurrence
of Tertiary forms, such as Thoracosphaera spp. even though nannoplankton species which had
populated the Cretaceous constitute the bulk of the fossil assemblage (95-50%) in the first
10-centimeter thick sediment above the contact, deposited during the first three or five
thousand years after the postulated boundary event. The paleontologically defined boundary
falls within a magnetostratigraphic epoch corresponding to the Seafloor Anomaly 29 - reversed,

which spanned a time interval less than half a million years.

Mass-Mortality and Dissolution-event

A boundary clay about two centimeters thick, containing less than 5% of CaC03 is present,
straddling the boundary. The clay constituents are apparently detrital, containing little
of what could be considered extraterrestrial materials, except an iridium anomaly (3.6 ppb Ir)
is present. The decrease of carbonate-content, as has been observed elsewhere, has been
interpreted as evidence of drastically reduced fertility of ocean life during and after the
terminal Cretaceous event. Dissolution of planktonic foraminifera is remarkable and the
CacO3-content showed a 10-fold decrease, in the uppermost (~ 10 centimeters) of Cretaceous
sediments. We believed that the calcite-compensation-level of the terminal Cretaceous ocean

rose because the lack of production had minimized the chemical difference between surface

140

and deep-bottom waters.

Carbon-Isotope Excursion

Both the analyses of the bulk and of the fine fraction, consisting almost exclusively
of nannofossils, produce a shift of the 13c/12c ratio, with maximum 613¢ shifts of about
minus three pro mille. A first shift during the first few hundred years after the event
May indicate a temporary change in the isotopic compositions of the surface waters when the
production of organisms was almost completely suppressed. After returning to the normal
Cretaceous value, a second maximum of minus three pro mille shift was built up during some
30 or 50 thousand years, probably because the production and sedimentation or organisms

could not keep in pace with the influx from rivers of carbon atoms anomalous rich in 12¢,

Oxygen-Isotope Excursion

Analyses of the bulk and fine fractions also revealed drastic changes of oxygen
isotopes, indicative of temperature-changes of surface waters. The shift in the boundary
clay layer could be interpreted as a signal of5°C cooling during the first few hundred years.
Like the carbon-shift, the oxygen-shift in the earliest Tertiary sediments suggests a
temperature rise of about 10°C (after the minimum) during the next 30 or 50 thousand years.
While the initial cooling might be related to the insulation of solar energy by the ring
of ejecta in the stratosphere, the subsequent warming could be related to the greenhouse
effect, either of water molecules thrown into the air by impact, or of increased CO2 in

atmosphere after the exchange between the atmosphere and an ocean with reduced fertility.

Scenario for Terminal Cretaceous Extinction

I am suggesting a two-act drama for the boundary event. The initial collision event
was so brief, that its only record was the fallout of the siderophiles (iridium), a carbon
isotope shift indicative of a suppressed ocean-productivity and an oxygen shift suggestive
of a momentary cooling. The mass-mortality of the oceans, in my view, might have been
caused by chemical poisoning, which has been distributed by surface currents and has
affected the tropical planktons the most. The consequences of the mass-mortality was a
change toward a more acid ocean. The survivors of the "endangered species", mainly marine

calcareous planktons, became extinct in this stressed ocean environment within 50 thousand

141

years or so after the catastrophe, while an explosive evolution of the Tertiary species was
eventually to occupy the vacated ecological niches. The temperature-rise after the
catastrophe may have produced the thermal stress which caused the extinction of the dinosaurs,
which had probably become the “endangered species" on land after the holocaust.

This summer an article is to be published by myself with my Swiss colleagues and with

my shipmates on Leg 73. I am presenting our joint research as a reporter for the team.

142

Deccan Volcanism and the Cretaceous-Tertiary Transition

Scenario: a Unifying Causal Mechanism
Dewey M. McLean

The Cretaceous-Tertiary (K-T) transition Deccan flood basalt volcanism, one of the
greatest outpourings of lavas from the Earth in geological history, was synchronous with,
and accounts for via fluctuation of the carbon cycle, the K-T transition: (1) marine CaC03
dissolution and "clay" events, and the apparently linked extinctions of calcareous micro-
plankton, (2) drops in 613C ana 6180 values, and (3) via "greenhouse" conditions, the
extinctions of the dinosaurs by heat-infertility linkage. Theorized to represent mantle
plume activity originating near Earth's core-mantle boundary, this volcanism provides a
mechanism for conveying iridium and osmium from the core (where they presumably exist in
cosmical proportions) to be distributed via volcanic exhalations. Carbon gases are among the
most abundant volcanic emanations, and possibly reflect degassing of Earth's interior
of material trapped during Earth's accretion. Deep origin volcanism spanning the 5 million
years of the Deccan volcanism (65 to 60 Myr ago) would have brought vast amounts of carbon
gases to Earth's surface. Volcanic C0: and reduced carbon in igneous rocks have low 6lèc
values (-70/00 and -19 to 289/59, respectively) , accounting for the drop in 613c¢ values
across the K-T contact. Build up of atmospheric CO» would result in "greenhouse" conditions
and the drop in 6180 values (a warming signal) across the K-T contact. Acidic volcanic
gases injected into the oceans would cause CaCO3 dissolution; the holding of CaCO3 components
in solution would allow accumulation of a concentrated clay layer (e.g. the Danish fish clay)
on the Cretaceous erosional (corrosional) surface. Lowered marine pH would also trigger
calcareous microplankton extinctions; they were most severe in the Tethyan region in
proximity to the Deccan volcanism. "Greenhouse" conditions would have affected reptilian
reproduction according to size. Larger organisms have relatively small surface-volume
ratios; during warming, organisms above a critical size (10 kilograms) would have retained
excessive body heat, causing degeneration of internal germinal tissues in males and, in
females, disruption of calcium metabolism and hormonal systems. The abrupt terminations of
geological ranges of many taxa at the K-T contact seemingly reflect hiatus control of ranges
associated with the shallow marine CaCO3 dissolution event, and terrestrial hiatuses and

facies control associated with the terminal Cretaceous regression of epeiric seas off the

143

continents, that have created the illusion of simultaneous global range terminations. In
all likelihood, the K-T extinctions spanned hundreds of thousands, if not millions, of

years.

144

Effects of Solar System Passage Through Giant Molecular Clouds

W.M. Napier

It is now known that giant molecular clouds comprise about half of the mass of the
interstellar medium. They have masses 10°-10$ solar masses and diameters typically 30-300
light years. It is probable that they are concentrated in the spiral arms of the Galaxy;
certainly they are associated with dust, which is a spiral arm tracer, and they are seen to
lie along the spiral arms of the Andromeda Galaxy. The Solar System passes through spiral
arms at 50-60 million year intervals. Approaching a line of GMCs it cannot soar over or
under because its orbit is nearly in the galactic plane, and it can only avoid encounter by
passing between them; however because of the large masses of these clouds there is a
gravitational focussing effect. Putting all these factors together one finds that there is
an expectation of one encounter with a GMC every 100-200 million years, that is 3-6 times
within the Phanerozoic.

GMCs seem to be largely empty space: they comprise dense nebulae, generally a few
light years in diameter and with masses about 1o+ solar masses. During passage through a
GMC there is an expectation that the Solar System will pass within 15 light years or less
of some such substructure, and strong tidal forces are then exerted across the Oort comet
cloud, which is over a light year in diameter. The result is to perturb the orbits of the
comets around the Sun to the extent that many of them escape. This has been modelled
numerically and some typical results are shown in Fig. 7: we see that even a single
passage through a GMC has a dramatic effect on the comet cloud. A typical close encounter
will result in the loss of 20-80% of the long period comets attached to the Solar System.
In addition, the orbits of the surviving comets are expanded so that escape becomes easier
on subsequent encounter. This process has been followed by numerical simulations: one such
run is shown in Fig. 8. We see that, within the lifetime of the Solar System, any primordial
cloud of comets would have been completely destroyed beyond a few thousand astronomical units
from the Sun. And yet the observed long period comets conspicuously arrive from regions
40,000-50,000 AU distant. These comets must have been placed there recently, and so we are
led to an interstellar origin, with a capture event having taken place within say the last
10 million years.

Questions to be asked on this theory are whether comets can grow in the interstellar

environment, and whether they can be captured from it. It turns out that molecular clouds

145

are crucial in both respects. Because of their high densities and low temperatures
(T < 20K) they are ideal places for the growth of large bodies. The mechanism we have
studied involves forcing together of dust grains by differential radiation pressure. We

o18

find that bodies will coagulate out with mean mass m % 1 gm (say diameter - 10 kilometers)

1.7

and differential mass distribution nm) « m or so. Recent observational work indicates

018 -1.64+0.2.

that the long-period comets have masses a few 1 gm and n(m) « m The striking

resemblance between cometary and interstellar chemical compositions has often been remarked
on and follows from this theory. The agreement is spuriously good — the theory is not that
accurate — but it does seem that there is at least one mechanism for growing comets in the

interstellar medium and there are probably others.

There is a deficiency of heavy elements in the interstellar medium in relation to the
predictions of galactic nucleosynthesis theory, and interstellar comets have already been
proposed as a means of ‘locking up' the missing material. If 5% of the mass of a GMC is
in the form of comets, then elementary calculations, with the GMC as a third body, give an
approximate capture rate during a close encounter between Sun and GMC. It turns out that
109-1011 comets can be captured during each encounter, comparable to the number lost:
one therefore anticipates a substantial overturn of the Oort cloud comet population during
encounter. On this hypothesis the Oort cloud is in a quasi-equilibrium situation with
calculated population and dimensions much as observed, whereas on the usual "primordial"
cosmogony these are empirical quantities.

The theory is vulnerable on a number of points: (1) the Solar System cratering flux
should be determined by time constants appropriate to the Galaxy rather than the Solar
System. In particular we expect fluctuations in the rate by factors three or four to one,
or even more, at 100-200 million year intervals on average; (2) we do not expect precise
agreement between the observed cratering rate (say over the last 500 million years) and
that expected from the current circumterrestrial missile population. Again a discrepancy
by a factor up to three or four is expected; (3) we should look for chemical signatures.
The (Ir, Os) isotope ratios at the K-T boundary have Solar System values but this may not
be critical: an interstellar planetesimal represents a mix of the output from many
r-process environments over a wide range of time and space and there is no reason to expect
significant variations within the Galaxy. On the other hand 12¢/13¢ varies appreciably
from place to place in molecular clouds due to fractionation and one should look for
departures of several percent from the terrestrial value of 89 at significant boundaries.

146

Accelerator Mass Spectrometric (AMS) Measurement of Pt and Ir at the Cretaceous-Tertiary

Boundary

J.C. Rucktidge et Gif

The tandem Van de Graaff accelerator at the University of Rochester has been used as
an ultra-sensitive mass spectrometer to measure ultra-low levels of Pt and Ir in samples
from the Cretaceous-Tertiary boundary. This method, which offers sensitivity comparable
and often superior to neutron activation analysis (NAA) is described in detail in
Rucklidge et al. (1982) or more generally in Litherland (1981). While still in an early
stage of development, this method has advantages which include the use of milligramme size
samples without extensive pretreatment, and the ability to determine sub-ppb levels of Pt
and Ir in times measured in minutes rather than days or weeks. By this method, the
detection of Pt is about 1.6 times more sensitive than Ir, which is in stark contrast to
NAA where Ir is about 1000 times more sensitive than Pt. This latter fact is the reason
that Ir is the element for which the anomaly is noted in so many analyses of the K-T
boundary, while in fact there may be equally valuable information in the Pt data. The
work summarized here is the first attempt to apply AMS to a geological problem, and the
data presented may not be directly comparable with NAA data from similar samples. This is
because the volume of material sampled by the two methods differs by a factor of 10* to 10°.
The AMS sample may very well not be representative of the bulk material, and the results will
be very sensitive to the particular mineral grains which present themselves to the primary
ion beam. Thus a heterogeneous distribution of Pt and Ir within a sample could result in
widely disparate readings which would have to be interpreted accordingly.

While a study of Ir in four samples across the boundary at Stevns Klint gives a profile
(Fig. 1) which is similar to that found by Alvarez et al. (1980) the Pt profile of the same
samples (Fig. 2) is extremely erratic, showing large variations in count rate with time.

The sample showing the largest variation is also the most pyritiferous, and the changes in
Pt count rate may be ascribed to the presence or absence of pyrite grains under the primary
ion beam. It would appear from this that Pt is concentrated in the sulphide phase, which
might imply that some marine geochemical process has been at work to achieve this. The same
erratic behaviour is not apparent in the Ir profiles, but there the count rate was much

smaller, and the statistics gave less confidence.

147

T T

IRIDIUM

COUNTS PER MINUTE
mn
©
fo)
T

194

Pt in 4 samples from
Danish Cretaceous - Tertiary Boundary

Cretaceous Fish

Chalk
991

Tertiary Clay

100
CRETACEOUS
CHALK
BED 3
acl; 08 SAMPLE 2
‘ 9827
LL +'atiempstitant- = = =
Ss ee Ê cS oc 0 10 20 30 40 50 60 70 60 90 100 110 120 130 140 150
COUNTING TIME (minutes)
TIME (MINUTES)
Fig. l. 1937, count rate plotted against Fig. 2. As for Fig. 1 but for 194, ,
time while analysing four samples across
the Cretaceous-Tertiary Boundary,
Stevns Klint, Denmark.
Table 1 Table 3

Ir and Pt by AMS, Stevns Klint. Dec 1980

| Sample | Ir(ppb)|Pt(ppb) |
|Tertiary |Bryozoan Lste| 0.13 | 1601
JE SSSEEE= FESSES ESS ESSSESS= EST |
| |Br°on Cllay93=8| 2.7 | 253]
| | 92=4/7 9537 1 438m)
| Fish | BI kK Clay 93=/ SC | OM ||
| | OS Aa | | 2691
| Clay | 92-2*| 14 IRAIES NI
| | Pyrite 92-2) 9.3 | Sik
| 10-20cm | rich 93-6*| 87 i 452) 7]
| | Clay 93=6+] 12 | 14 |
| |Grey Clay93-5| 2.9 | 2.6|
PES | ==SSs=ss=—S=2\|====S== fsa |
|Cretaceous|Bryozoan Chlk| 0.14 | 0.3]

*heavy fraction +light fraction

Table 2

Ir and Pt by AMS, Glendive,Mont.Dec 1980

| Sample |Ir(ppb)|Pt (ppb) |
10168/9 3"above Z contact| 0.07 | om
a l======= |======= |
10168/8 at Z contact" 0.17 | GO |
|| HHS RS SSS SSS SS SSS === [ESS lee |
10168/7 3"below Z contact| 0.13 | AAs ||

University of Toronto Palynological
collection numbers should all be

of the form 15nn-—-n.
The order in the tables reflects the

stratigraphic relationships.

148

Ir and Pt by AMS, Danish sections. May 1981

| Sample | Ir(ppb) |Pt (ppb) |Enrich* |
|Stevns Klint | | | |
| Section 1 92-6 fh ob | Wok | 476 |
| 92-5 Osa SA | 22/40
| Heavy 92-4 1109 | 667 | 476 |
| fractions 92-3 | BA | 106 | 524]
| 92-2 | 6s iP nO) | 2
| | | | |
| Section 2 93-9 | 46 | 4.8 | 4650 |
| 93-8 | Oe f) AB | 200 |
| Heavy 93-7 fh Woes} ff 13} | 25370
| fractions 93-6 | dot IS SYA | Gall
| 93-5 | | 72 | 245 |
| | | | |
| Jutland | | | |
| Nye Klgr (whole rock) | 0.17 | 0.004| 476 |
| Kjolby Gard (heavy) | | 0.96) IN 62508)
| Kjolby Gard (whole) | Ws5@re)s}|] Was) || 1
| |

|
*Enrichment=(Wt rock sample)/(Wt heavy extract)

Table 4

Ir and Pt in Danish organic extracts. May 1981

| Sample | Ir(ppb) |Pt (ppb) |Enrich* |
| | | | |
| 93-10 Grey Clay | Solo 1 EEE) || 16268}
| 92-3 Black Clay DNS CN | 348 |
| 93-7 Black Clay Lu CU LYS | 90 |
| 93-5 Grey Clay ils 3) | 39 | 1140 |
|

*Enrichment=(Wt rock sample)/(Wt organic extract)

To establish whether concentration of Pt group elements (PGE) in sulphides has taken
place, some of the more pyritiferous samples were separated into heavy and light fractions,
and analysed individually. The results certainly seem to demonstrate a strong affinity of
Pt and Ir for the chalcophile fraction, as can be seen from Tables 1 and 3. The other type
of separation which was made was to extract only the organic component, and subject this
to analysis for Pt and Ir. In preparing these organic extracts from the Danish Fish Clay,
physical methods were used as far as possible to avoid any possibility that PGE could be
concentrated chemically. Glauber's salt was used to break the rock down, and the resulting
grains were passed through 30 mesh, 150 mesh and 40 micron sieves. The fractions retained
by these three sieves were treated with heavy liquid (tetrabromoethane) and the light
fractions from this stage were treated with 10% nitric acid and 35% hydrofluoric acid to
extract the organic residue. Only milligramme quantities of organic material were obtained,
but the analytical method has such sensitivity that these were analysed with ease, and were
found to contain impressively large amounts of Pt and Ir. However, when the enrichment
factor is taken into account, there is insufficient Pt and Ir in the organic fraction to
account for the concentrations in the bulk material. It is interesting to note that the Ir
anomaly reported by Orth et al. (1981) occurred in a coal seam, so the association of PGE
and organic remains may be one which should be investigated in more detail. The analytical
data for samples from Stevns Klint, N. Jutland, and Montana are shown in Tables 1 to 4.

We thank T.F.D. Nielsen and H.J. Hansen for the Danish samples, and D.A. Russell for the

Montana samples.

References

Alvarez, L.W., W. Alvarez, F. Asaro, H. Michel. 1980. Extraterrestrial cause for the
Cretaceous-Tertiary extinction. Science 208:1095-1108.

Litherland, A.E. 1980. Ultrasensitive Mass Spectrometry with Accelerators. Ann. Rev. Nucl.
Part. Sci. 30:437-473.

Orth, C.J., J.S. Gilmore, J.D. Knight, C.L. Pillmore, R.H. Tschudy and J.E. Fassett. 1981.
Iridium anomaly at the palynological Cretaceous-Tertiary boundary in New Mexico.
EOS 62:438 (abstract).

Rucklidge, J.C., M.P. Gorton, G.C. Wilson, L.R. Kilius, A.E. Litherland, D. Elmore and

H.E. Gove. (in press). Measurement of Pt and Ir at sub-ppb levels using accelerator
mass spectrometry. Can. Mineral.

149

PARTICIPANTS

Luis Alvarez

Walter Alvarez

Frank Asaro

Pierre Béland

Klaus Brasch

Victor Clube

David Dilcher

Paul Feldman

Richard Grieve

Ian Halliday

Kenneth Hsü

David Jarzen

John Jaworski

Frank Kyte

Ted Litherland

Dewey McLean

Helen Michel

William Napier

Kris Pirozynski

John Rucklidge

Lawrence Berkeley Laboratory
University of California

Department of Geology and Geophysics
University of California

Lawrence Berkeley Laboratory
University of California

Péches et Océans
Québec, P.Q.

Department of Biology
Queens University, Kingston

Royal Observatory, Blackford Hill
Scotland

Departments of Biology and Geology
Indiana University

Herzberg Institute of Astrophysics
National Research Council of Canada

Earth Physics Branch

Department of Energy, Mines and Resources (Canada)

Herzberg Institute of Astrophysics
National Research Council of Canada

Swiss Federal Institute of Technology
Zurich

Paleobiology Division
National Museum of Natural Sciences (Canada)

Environmental Secretariat
National Research Council of Canada

Institute of Geophysics and Planetary Physics
University of California, Los Angeles

McLennan Physical Laboratory
University of Toronto

Department of Geological Sciences
Virginia Polytechnic Institute

Lawrence Berkeley Laboratory
University of California

Royal Observatory, Blackford Hill
Scotland

Paleobiology Division
National Museum of Natural Sciences (Canada)

Department of Geology
University of Toronto

Dale Russell Paleobiology Division
National Museum of Natural Sciences (Canada)

Jan Smit Geological Institute
University of Amsterdam

Hans Thierstein Geological Research Division
Scripps Institution of Oceanography

Wallace Tucker Center for Astrophysics

Harvard College Observatory,
Smithsonian Astrophysical Observatory.

151

RECENT SYLLOGEUS TITLES / TITRES RECENTS DANS LA COLLECTION SYLLOGEUS

No. 24
No. 25
No. 26
No. 27
No. 28
No. 29
No. 30
No. Si
No. 32
No. 33
No. 34
No. 35
No. 36
No 37
No. 38

Haber, Erich and James H. Soper (1979)
VASCULAR PLANTS OF GLACIER NATIONAL PARK, BRITISH COLUMBIA, CANADA.

Danks, H.V. (1980)
ARTHROPODS OF POLAR BEAR PASS, BATHURST ISLAND, ARCTIC CANADA.

11, 68 D:

Harington, C.R. (ed.) (1980)
CLIMATIC CHANGE IN CANADA. 246 p.

White, David J. and Karen L. Johnson (1980)
THE RARE VASCULAR PLANTS OF MANITOBA. / LES PLANTES VASCULAIRES RARES DU MANITOBA.
52) Ds / OSD.

Douglas, George W., George W. Argus, H. Loney Dickson, & Daniel F. Brunton (1981)
THE RARE VASCULAR PLANTS OF THE YUKON. / LES PLANTES VASCULAIRES RARES DU YUKON.
(HS SAC pe

Brodo, I.M. (1981)
LICHENS OF THE OTTAWA REGION. (English edition) 137 p.
LICHENS DE LA REGION D'OTTAWA. (L'édition française)

Manning, T.H. (1981)
BIRDS OF TWIN ISLANDS, JAMES BAY, N.W.T., CANADA. 50 p.

Ferguson, R.S. (1981)
SUMMER BIRDS OF THE NORTHWEST ANGLE PROVINCIAL FOREST AND ADJACENT SOUTHEASTERN
MANITOBA. 23 p.

Ouellet, H. et F.R. Cook (1981)
LES NOMS FRANCAIS DES AMPHIBIENS ET REPTILES DU CANADA: UNE LISTE PROVISOIRE. 7 p.

Harington, C.R. (ed.) (1981)
CLIMATIC CHANGE IN CANADA 2. 220 p.

Bousfield, E.L. and N.E. Jarrett (1981)
STATION LISTS OF MARINE BIOLOGICAL EXPEDITIONS OF THE NATIONAL MUSEUM OF NATURAL
SCIENCES IN THE NORTH AMERICAN PACIFIC COASTAL REGION, 1966 TO 1980. 66 p.

Coad, B.W. (1981)
A BIBLIOGRAPHY OF THE STICKLEBACKS (GASTEROSTEIDAE:OSTEICHTHYES). 142 p.

Pendergast, J.F. (1982)
THE ORIGIN OF MAPLE SUGAR.

Russell, D.A. and R. Séquin (1982)
RECONSTRUCTION OF THE SMALL CRETACEOUS THEROPOD STENONYCHOSAURUS INEQUALIS AND A

HYPOTHETICAL DINOSAUROID. 43 p.

Jarzen, David M (1982)
PALYNOLOGY OF DINOSAUR PROVINCIAL PARK (CAMPANIAN) ALBERTA. 69 p.

das. AUD TETE TOP
CALIF ACAD OF SCIENCES LIBRARY

Oe UE Be 4

bh Meee a 3 1853 10004 4671
Crete Gita ip 5 EH

EX
PRE el). ARS

v LA
= |
14
re
x
a
, F
ñ
i
+ >
t
j