Ahh. The history of science. I've always wondered how these naming
conventions get decided. Who is the authority on what gets named after
who? Historically, it seems to vary a lot.
- When Patterson published his incredibly useful map he called it the
"F-square synthesis". Does anyone NOT call it a Patterson map?
- When Rontgen discovered a new kind of light, he called it "x-rays".
Now only the Germans call them Rontgen rays.
- When the largest protein ever was discovered, it was called
"connectin", but a subsequent paper called it "titin" and the second
name has stuck. I actually can't remember who the "connectin" guy was ...
- When Joseph Fourier discovered that heat radiated from the earth could
be reflected back by gasses in the atmosphere, he simply named it by
describing it (in French). Now this is (incorrectly) called the
"greenhouse effect". Why not the Fourier effect? Fortunately for
Fourier, a mathematical series was named after him, although he neither
discovered it (Budan did that), nor implemented it (Navier did that).
All Fourier did was present a theorem based on a flawed premise that
turned out to be right anyway.
So, I decided to look up Friedel and Bijvoet in the Undisputed Source of
All Human Knowledge (wikipedia) and found that Friedel's Law "... is a
property of Fourier transforms
<http://en.wikipedia.org/wiki/Fourier_transform> of real functions."
I am willing to believe that. And considering this origin I would think
it appropriate to call (hkl) and (-h-k-l) a "Friedel pair" (or
"Friedel's pair" as it is described in the USAHK). G. Friedel was
indeed a crystallographer, but I doubt he considered more than this
simple centrosymmetric property. Who would care in 1913 which is F+ and
F-? The atomic scattering factors had not yet been worked out at that
time. Ewald may have predicted it, but anomalous scattering was not
shown to exist until the classic work of Koster, Knol and Prins (1930).
I guess that goes to show that if you want something named after you...
keep it at one or two authors.
Perhaps it has to do with the original paper getting old enough that it
gets too hard to find. I'm sure in G. Friedel's paper in 1913 he cited
Joseph Fourier's Paper from 1822. Or did he? I wonder if they were
already calling it a Fourier Transform at that time?
Okay, so what, exactly did Bijvoet do? Everyone cites his Nature paper
(1951), but one thing that I was NOT KIDDING about in my April Fool's
joke was that this paper (like so many other high-profile papers)
contains almost no information about how to reproduce the results. I
was also not kidding that boring little details like the reasoning
behind the conclusion (the hand of the microworld) were relegated to a
more obscure journal (the one in the Proc. Royal. Soc. Amsterdam). I
WAS kidding about having found and read that paper. I have never seen
it. Still, Bijvoet did the first experiment to elucidate the absolute
configuration, and he definitely deserves credit for that.
So, particularly in that light, I would agree that any pair of
reflections that would be equivalent if not for anomalous scattering
effects could be called a "Bijvoet pair". This is because they contain
the information needed to apply Bijvoet's technique.
Something that has always eluded me is who decided which is F+ and F-?
After all, the reciprocal lattice is very very nearly centrosymmetric.
You cannot tell by looking at a single diffraction image whether that
spot at a given X,Y pixel coordinate is F+ or F-, you need to know the
axis convention of the camera. At some point in writing the CCP4
libraries with their asymmetric unit definitions, someone must have
established a convention. What is it? To me, the reasoning behind
these assignments is, in fact, they key to assigning the absolute
configuration, not the anomalous scattering effect itself. So, who
worked this out? Should we really be calling them Dodson pairs?
-James Holton
MAD Scientist
