Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site astrovax.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!princeton!astrovax!elt
From: elt@astrovax.UUCP (Ed Turner)
Newsgroups: net.astro.expert
Subject: Re: nemesis and the end of the world
Message-ID: <355@astrovax.UUCP>
Date: Fri, 25-May-84 10:51:02 EDT
Article-I.D.: astrovax.355
Posted: Fri May 25 10:51:02 1984
Date-Received: Thu, 31-May-84 23:49:41 EDT
References: <16@utastro.UUCP>
Organization: Princeton Univ. Astrophysics
Lines: 44

Ethan has recently expounded the Nemesis or Death Star theory for the
apparent periodic mass extinctions on Earth.  Since I have them in front of
me, I thought I might add a few references and then indulge myself with a
couple of comments.

1) paleomtological evidence of periodicity: D. M. Raup and J. J. Sepkoski
   in PROC. NAT. ACAD. ACI. late 1983 or 1984.

2) suggestion of Nemesis hypothesis: M. Davis, P. Hut, and R. Muller in
   NATURE, 1984.

3) confirmation of the periodicity in crater age data: W. Alvarez and
   R. Muller in NATURE, 1984.

4) evaluation of stability of wide binary clock: P. Hut in NATURE, in press
   I think.

My comments:

Computer simulations by Hut described in 4 above seem to indicate a period
stability of the order of 10% accuracy over the required time scales; this
is sufficient given the large errors in crater aging data.

I believe the apparent periodicity in the crater age data is more
impressive than Dr. Nather's recent comments imply.  It is true that the
data are few and uncertain, but it is also true that a *preexisting* crater
age data set gives essentially the same period and *the same phase* as the
fossil record when subjected to a quite straightforward analysis.  Also I
think that Alvarez and Muller (#3 above) have been reasonably careful and
conservative in their analysis.  Their paper (which anyone interested
should certainly read) describes a number of checks and tests, including
Monte-Carlo simulations they used to estimate the significance of their
result.  Their 99.5% confidence result looks reasonable to me.

It is amusing to note that if the Nemesis hypothesis is correct, the period of 
order 10^6 yrs following a close passage would be distinguihed by having the
night sky filled with comets.  from the figures given in reference 2 above I
calculate an average of roughly 10^3 comets would be visible at any given
time!  Perhaps the old superstitions about comets being the "harbringers of
doom" is an ancient genetic memory that associates this spectacle with the
unpleasantries associated with impacts.  :-)

Ed Turner
astrovax!elt