Re: North American plant extinctions

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

        I agree there's no need to start another flame war on the K-T  
(the last one was pretty exhausting), but I do think Gabhan's drawn  
the wrong conclusion about the nature of the North American  
paleobotanical evidence for a K-T boundary catastrophe, perhaps  
because I didn't make myself clear enough in the first place. There  
IS a fern spike, and it is above "background" levels for changes in  
the relative abundance of ferns such as those you would see in the  
wake of "normal" forest fires. However, the important point has  
nothing to do with fern spikes or "floral graveyards" (here I'm not  
sure what Gabhan is talking about, exactly), but rather with the fact  
that a bona fide mass extinction of angiosperms really did occur.

        The best data are from a boundary section near Marmarth,  
North Dakota, where over 8,400 leaf specimens from within 50 m of the  
boundary were recovered by Kirk Johnson et al. (1989: Nature  
340,708). Shocked quartz and a 25-fold iridium anomaly were found at  
the local boundary, which is the formational contact between the Hell  
Creek and Fort Union. Pollen samples were taken from the same section  
and showed that of 90 taxa present just below the boundary, 30  
disappear, i.e., a 30% extinction event in a 1.8 meter section!  
Considering that pollen "species" often correspond to megafloral  
genera, it comes as no surprise that the 79% of the megafloral taxa  
(generally species) from the very highest Cretaceous floral zone go  
extinct at the boundary, and none of the dominant species (those  
representing 5% or more at any level below the boundary) survive the  
event.

        On top of that, the floral data clearly indicate a  
substantial warming trend in the period leading up to the boundary,  
based on the presences of palms and the angiosperm family Laurales,  
which they say is "thermophilous," and on an increased percentage of  
entire-margined angiosperm leaves (the classic paleobotanical  
indicator of warm climates) from 35% to 49%. This doesn't match at  
all with climate-driven theories for extinction (a "minny ice  
age-ette" or whatever you want to call it), at least as far as the  
Western Interior of North America is concerned - and that's the only  
place we have a really good terrestrial K-T fossil record. There is a  
more recent and presumably more detailed paper by Johnson (1992:  
Cretaceous Research 13,91) on the megafloral data, but I haven't  
gotten a look at it yet. I hope this doesn't start a flame war; I'm  
just trying to shed a little light on the issue...