yet more K-T drivel

[John Alroy <jack@homebrew.uchicago.edu>]

First I will comment on Schwimmer and then Friesen.

Tornillo: I haven't seen the Lehman article (nor do I know what the  
reference is) so I will wait to comment. However, I want to point out  
that I wasn't claiming there was an uncomformity at the boundary; I  
just said there was a 35 m interval devoid of terrestrial vertebrates  
that was bracketed an unambiguous Paleocene mammal fauna above and a  
pterosaur below.

Duration of the "2 m gap": I will grant that it is dangerous to  
interpolate sedimentation rates based on the known age of an entire  
formation. However, the point here is that the best estimate for the  
interval is 40,000 years and not 100,000 - 250,000 years (the figure  
I contested in the first place), unless hiatuses in that interval are  
substantially greater than average for the formation and similar  
formations elsewhere. If the 2 m represent a single depostitional  
package, as is likely, they easily could represent a depositional  
event on the scale of a few thousand years.
I did earlier discuss this with Ray Rogers, who has published  
extensively on the stratigraphy, sedimentology, and taphonomy of the  
Two Medicine and Judith River formations. He urged me to make the  
above argument that the 2 m zone could easily represent a very short  
time interval.

Polls: I agree that science isn't a democracy. However, my point was  
that the anti-impact camp was already a minority by 1984 (based on  
Raup's book) and has now become a very small one (based on McLean's  
data), which agrees with my essentially sociological model for the  
acceptance of new theories. Schwimmer has a good point about VP'ers:  
I think in general there has been a startling lack of agreement  
between geologists and marine paleontologists, on the one hand, and  
terrestrial VP'ers, on the other. Because the gradualist VP'er  
arguments center on things like the "2 m gap" and the now-falsified  
claim for gradual dinosaur extinction through a substantial stretch  
of the Maastricthian, this looks to me like a strange sociological  
artifact more than anything. I suspect if you took a poll now the  
VP'ers would be much more inclined to accept the idea that an impact  
did occur. The Rapid City meeting was indeed in 1985. I wasn't there  
but I have attended all eight meetings since then, and I have the  
impression that anti-impact feelings are no longer strong in the SVP.

Manson: I think it was pretty clear back in the mid-80's that Manson  
was just too small to cause a global catastrophe; that's what I meant  
by a "decent" candidate crater. I realize that literature as recent  
as 1992 (a paper by Anderson and Hartung I haven't seen yet) argue  
the contrary. However, I'm not clear what Dr. Schwimmer means by  
saying Manson was "subsequently discredited." I seem to recall that  
the latest radioisotopic dates for the crater are still within error  
of many other K-T boundary dates, but again I haven't seen a  
reference yet (apparently Hartung et al. 1990 GSA Special Paper 247  
reported dates of this kind).

Meyerhoff: I've dug through some literature and it's complicated. The  
original Chicxulub paper (Hildebrand et al. 1991: Geology 19,867) did  
address the available 1960's drillhole data and saw no contradiction  
of the impact model in these data. Meyerhoff et al. simply repeat  
data for a well (Y-6) that was published many years before and  
figured in the Hildebrand paper. I'm not competent to judge their  
geological arguments that the lower part of this well is  
volcanogenic, but Hildebrand et al. don't think so and went over the  
same data. I really think we've got a problem here that needs more  
research. However, when it comes to the faunal data - the supposed  
Late Cretaceous age for material within the crater - I think  
Meyerhoff et al. have a very weak argument. They describe exactly one  
level in one core as including "interbedded limestone with Late  
Cretaceous microfossils, volcanic breccias, and andesite." That's ALL  
they say about the supposed Cretaceous fauna, which comes from a  
rather deep level in the drillhole (1590 m out of about 1640).  
Hildebrand et al. discuss other fossil material from the same  
drillhole but a higher level (1000 m) that was said to be Late  
Cretaceous in the original writeup of the drillhole. However, Keller  
and Sliter are then cited as having reidentified the fauna as late  
Paleocene! This indicates to me that no conclusions should be drawn  
from the supposed Late Cretaceous fauna until it is restudied. As for  
radioisotopic dates, none greater than the age of the K-T have been  
reported despite multiple efforts, and several of the available dates  
are indistinguishable from the age of the K-T (e.g., Swisher et al.  
1992: Science 257,954). These dates are higher than the 1590 m fauna,  
of course, but my point is that Meyerhoff et al.'s argument is  
completely equivocal from a paleontological point of view.

Regressions: My point was that RELATIVELY equally severe and rapid  
regressions occurred several times during the Cenozoic, not that  
North America had a large inland sea for a zillion years that  
repeatedly regressed (it did not). I know full well about the age and  
extent of the Cannonball Sea. I see no reason why a regression  
eliminating an inland sea should have any greater or lesser effect on  
TERRESTRIAL biotas than a regression of comparable magnitude coming  
at a time when there was no inland sea. For example, I think we can  
agree that the regression at the Eocene-Oligocene boundary (or in the  
earliest Oligocene, depending on who you talk to) was just as or more  
extreme than the K-T event and had virtually no impact on North  
American mammal faunas, which are quite abundantly represented in the  
record in that interval (the White River Group straddles the event).  
This actually does raise an interesting point - there is a huge (50%  
at the genus level) mass extinction of mammals in Europe at the  
Eocene-Oligocene boundary. I myself am convinced this is a  
biogeographic effect - a bunch of Asian mammals crossed a short-lived  
land bridge that came up at this time and all hell broke loose.  
Nothing similar happened in Asia and North America because these  
faunas were already full-scale continental faunas (Europe was an  
island archipelago) that were "used" to inter-continental migrants.  
If this story is right, and if the K-T regression opened up land  
bridges that resulted in a lot of migration, why is there basically  
no important migration of mammals into North America at that time?  
The early Paleocene mammal fauna of North America can mostly be  
derived autochthonously.

Friesen: Most of this is agreeable.

Periodicity: I really think this is the wrong place to get into a  
debate about periodicity. For one thing, it has basically nothing to  
do with dinosaurs except insofar as the K-T fits into the periodic  
model. However, my own feeling about this is that the statistical  
data were equivocal. I'm surprised that Friesen feels so strongly  
that the statistics were flawed, and that a genus-level analysis  
wouldn't help matters, but we can discuss that elsewhere. I agree  
whole-heartedly that an improved time scale would make a huge  
difference; getting the time scale straight for my North American  
mammal data has made a huge difference in my ability to get  
statistically meaningful results.

Tsunamis: I'm not sure we disagree about anything here. I agree that  
the Deccan Traps and the regressions and so on may have had a  
negative impact on Maastrichtian faunas and may have set the world up  
for a "coup de grace." This requires more research.

Faunal lists: This is just a misunderstanding. I wasn't talking about  
the "2 m gap" in this passage, I was talking about the reliability of  
the apparently unpublished faunal data supporting a Cretaceous age  
for the Y-6 drillhole (see above).

Plants: My point was that plants in North America show a severe and  
rapid extinction at the boundary. As far as I know there is some  
evidence that the extinction wasn't as bad elsewhere, but I seriously  
doubt that any study as large as Johnson's for a good K-T section has  
been performed elsewhere in the world. It actually HELPS the impact  
model for extinction if plants took a harder hit in North America  
than, say, New Zealand.

Mammals and regression: Mammals did go through a huge extinction at  
the K-T boundary, as I'm sure Stan knows. For those in doubt, take a  
look at the Bug Creek data I distributed yesterday. Given that large  
body size is correlated with low intrinsic population growth rates  
and low population densities, it is a straightforward prediction of  
ecological theory that very large animals will do very poorly during  
major environmental perturbations of any kind. Therefore, I think the  
mammals vs. dinosaur issue is of no help at all in determining just  
what happened at the K-T. Archibald thinks otherwise, but he isn't  
here to speak for himself (at least not as far as I know). Maybe  
someone should ask him to join the list (what's his address?).