Bjoern's challenge
zoe_althrop
Quote:
>Zoe, I've challenged you before; I do it again here:
>1) Choose oldD/Di (the same for all samples).
>2) Choose an age (the same for all samples).
>3) Choose original P/Di for, say, three samples (different values,
>please).
>4) Calculate present P/Di from original P/Di and the age.
>5) Calculate newD/Di from original P/Di and the age.
>6) Calculate totalD/Di from oldD/Di and newD/Di.
>7) Give to us *only* the three values of totalD/Di and the three values
>of present P/Di.
>8) Let's see if we can derive the values you chose for oldD/Di, the age
>and original P/Di from that, o.k.?
End quote.
Okay. I believe my figures fulfil the above requirements. I'm going
to give you the figures as they would be if measured today, without
reducing the measurements. And for good measure, I'm giving you four
samples, not three.
A B C D
Total D/Di 50:4 85:8 120:12 195:24
Present P/Di 10:4 15:8 20:12 25:24
and now you'll tell me oldD/Di, the age, newD/Di, and original P/Di,
right? If you do, I surrender.
----
zoe
Jon Fleming
wrote:
Piece o' cake.
To avoid poisoning the well, I've emailed my answers to Zoe and Zoe
alone.
Sven Silow
>Bjoern, I accept your challenge, which you phrased as follows:
>
>Quote:
>
>>Zoe, I've challenged you before; I do it again here:
>>1) Choose oldD/Di (the same for all samples).
>>2) Choose an age (the same for all samples).
>>3) Choose original P/Di for, say, three samples (different values,
>>please).
>>4) Calculate present P/Di from original P/Di and the age.
>>5) Calculate newD/Di from original P/Di and the age.
>>6) Calculate totalD/Di from oldD/Di and newD/Di.
>>7) Give to us *only* the three values of totalD/Di and the three values
>>of present P/Di.
>>8) Let's see if we can derive the values you chose for oldD/Di, the age
>>and original P/Di from that, o.k.?
>
>End quote.
>
>Okay. I believe my figures fulfil the above requirements. I'm going
>to give you the figures as they would be if measured today, without
>reducing the measurements. And for good measure, I'm giving you four
>samples, not three.
>
> A B C D
>
>Total D/Di 50:4 85:8 120:12 195:24
>
>Present P/Di 10:4 15:8 20:12 25:24
(As I "challenged" you with this first I don't feel ashamed of using
these figures also for my "challenge".)
Normalizing (i.e. getting the denominator the same in all ratios)
i.e. multiply with: 6:6 3:3 2:2 1:1
Total D/Di 300:24 255:24 240:24 195:24
Present P/Di 60:24 45:24 40:24 25:24
For normalized figures (Di = 24) we have:
D:P 300:60 255:45 240:40 195:25
300-D:60-P 0:0 45:15 60:20 105:35
Slope (300-D)/(60-P) n/a 3 3 3
D - 3*P 120 120 120 120
(Hey! Your figures works!)
Thus:
Equation of isochron for Di=24: D = 120 + 3*P
which gives us the general expression: D/Di = 5 + 3*P/Di
>and now you'll tell me oldD/Di,
This corresponds to the term 5 in the general expression, thus:
old_D/Di = 5:1 (or 120:24) = 5.000
>the age,
The slope is 3, which gives us that the age is equal to the length of
two half lives. [2log(3+1) = 2log(4) = 2]
>newD/Di,
Is given by newD/Di = 3*P/Di, thus for the samples:
A: P/Di = 10:4 => newD/Di = 30:4 = 7.500
B: P/Di = 15:8 => newD/Di = 45:8 = 5.625
C: P/Di = 20:12 => newD/Di = 60:12 = 5.000
D: P/Di = 25:24 => newD/Di = 75:24 = 3.125
>and original P/Di,
Is given by originalP/Di = P/Di + newD/Di, thus for the samples:
A: 10:4 + 30:4 = 40:4 = 10.000
B: 15:8 + 45:8 = 60:8 = 7.500
C: 20:12 + 60:12 = 80:12 = 6.666...
D: 25:24 + 75:24 = 100:24 = 4.166...
>right? If you do, I surrender.
Finally!
And, Zoe, congratulations to finding a correct set of numbers for a
working isochron. It really took its time...
And, especially, congratulations Björn!
(Handing over virtual bottle of wine.) Cheers! ;-)
Sven
Jon Fleming
wrote:
Check.
>>the age,
>
>The slope is 3, which gives us that the age is equal to the length of
>two half lives. [2log(3+1) = 2log(4) = 2]
Check.
>>newD/Di,
>
>Is given by newD/Di = 3*P/Di, thus for the samples:
>
>A: P/Di = 10:4 => newD/Di = 30:4 = 7.500
>B: P/Di = 15:8 => newD/Di = 45:8 = 5.625
>C: P/Di = 20:12 => newD/Di = 60:12 = 5.000
>D: P/Di = 25:24 => newD/Di = 75:24 = 3.125
Check, check, check, check.
>>and original P/Di,
>
>Is given by originalP/Di = P/Di + newD/Di, thus for the samples:
>
>A: 10:4 + 30:4 = 40:4 = 10.000
>B: 15:8 + 45:8 = 60:8 = 7.500
>C: 20:12 + 60:12 = 80:12 = 6.666...
>D: 25:24 + 75:24 = 100:24 = 4.166...
Check, check, check, check.
>>right? If you do, I surrender.
>
>Finally!
Any side bets on whether she'll retract her surrender?
zoe_althrop
<jo...@fleming-nospam.com> wrote:
>On Fri, 9 Aug 2002 17:32:38 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>>
>>Er, no, the ratio of totalD/Di (not oldD/Di) ranged from
>>approximately 1/10 to 8/10. ...
>
>Oops, I meant to say " the ratio of totalD/Di (not oldD/Di) ranged
>from approximately 3/10 to 8/10".
oh, okay.
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
zoe wrote:
>>Or in reverse, take, for example, a post of Wayne Hoxsie's,
>
>I assume you mean <http://makeashorterlink.com/?D2CF53A71>
yes, and for convenience, here's a cut-and-paste:
Sample Lu (ppm) Hf (ppm) 176Lu/177Hf 176Hf/177Hf
---------------------------------------------------------
zrc-1 174.3 15,330 0.001614 0.282363
zrc-2 74.22 10,430 0.001010 0.282538
biot-2 0.7598 14.41 0.007484 0.282597
wr-1 1243 50.61 3.529 0.349371
wr-2 1315 73.46 2.564 0.330715
xt-1 2948 16.20 28.35 0.816178
xt-2 3373 19.82 26.36 0.779948
P : 176Lu
D : 176Hf
Di: 177Hf
176Lu beta decays to 176Hf (stable) with a half-life (t1/2) of
3.78E10y
(37.8 billion years).
>>where he
>>gave some real-world measurements. The ratio of oldD/Di ranged from
>>approximately 2/10 to 8/10 -- listing measurements for 176Hf/177Hf of
>>0.282363 to 0.349371 to 0.816178.
>
>Er, no, the ratio of totalD/Di (not oldD/Di) ranged from
>approximately 1/10 to 8/10. ...
right, I meant total D/Di. Of course, that would make hardly any
difference since with a half life of 37.8 billion years, newD would be
just a fraction of total D/Di.
Also, I don't know where you got 1/10. Are you looking at the P/Di
column?
>The single oldD/Di value was 0.282.
you mean the base of 0.282515?
>Of course, if a mineral happens to take up a lot of P when it
>solidifies, after some time its totalD/Di ratio will be significantly
>higher than the world-wide average.
I would think that the only way for there to be as much oldD as the
samples show presently is for there to have been a lot of P in the
original rock. But then that means that after a remelt, there would
still be a lot of P in the melt and all samples would have a
corresponding high quantity of P in them. Unless you're saying that P
at solidification is incorporated unevenly so that some samples of the
rock source would have high quantities of P and other samples have
very small quantities of P?
if this is the case, a test consisting of a mere two samplings of zrc,
biot (only one sample, actually), wr, and xt, is not sufficient to
tell us if we have an accurate picture of oldD/Di ratios, is it?
>And ... it's easier to measure
>accurately if there's (relatively) a lot of P and D in the sample, so
>scientists tend to select minerals for sampling that they expect to
>have a lot of P (and consequently a lot of totalD).
is a sampling of two sufficient to establish a pattern?
>>Those ratios do not seem to
>>indicate what 4.5 billion years worth of accumulated oldD should be in
>>relation to its daughter isotope which has remained the same over
>>time, does it, not when the decay rate for Lu is a half-life of 37.8
>>billion years?
>
>Er, the relation is between the oldD and a non-radiogenic (not derived
>from radioactive decay) isotope, Di. Di is _not_ a daughter isotope.
I thought Di stands for daughter isotope? The way I understood it was
that P stood for parent, D stood for daughter, and Di stood for
daughter isotope -- an isotope that is arranged exactly like the
daughter except for one of its electrons or something like that.
>You can't predict what the ratio of oldD to Di should be from
>radioactive half-lives alone; you need to know something about the
>amount of Di now or previously
Di (now or previously) always remains the same, doesn't it?
>and the amount of P that is or was
>available to decay to oldD.
and that, I take it, is unknown.
>You would expect that younger rocks would have a larger odlD/Di ratio
>than older rocks.
I'm not sure I understand what you mean here. Larger in what sense?
Younger rocks have a wider difference? And if so, in which direction?
>So, if you want to look at oldD/Di ratios and figure out if they're
>reasonable, you'll have to look at several real world data sets that
>measured rocks that came out to be very different ages.
I guess I'll have to do a google for that. Do you have any
suggestions of good search terms I could use to find such data?
Thanks.
----
zoe
zoe_althrop
<rle...@ibmrlevett.uklinux.net> wrote:
snip>
>zoe; did it ever occur to you to harmonise your worldview with the
>data?
isn't that what I'm doing right now? Harmonizing my worldview with
the data? If I come to see sufficient discrepancies in my worldview,
then it would have to change. But one or two unanswered questions
coming from the field of science does not yet begin to undermine a
worldview which is built on a much wider foundation than just science.
----
zoe
zoe_althrop
<host...@sky.fit.qut.edu.au> wrote:
>zoe_althrop <muz...@aol.com> wrote:
>> On Tue, 6 Aug 2002 15:01:18 +0000 (UTC),
>> harshman....@sjm.infi.net (John Harshman) wrote:
>>
>> snip>
>>
>>>Just to clarify: do you now agree that the geologic time scale is correct,
>>>that life is millions of years old, and that the sorting of the fossil
>>>record shows changes in the earth's biota over hundreds of millions of
>>>years?
>>
>> not at all. I agree that it is possible for you to figure out oldD/Di
>> -- not oldD -- given certain assumptions about Di.
>
>Note that D/Di and P/Di are calculated from measurements
>of D, Di and P. I leave it as a mathematical exercise to
>figure out oldD from oldD/Di
then there should never have been the insistence that "We don't know
what oldD is." The given would be that "We know oldD because we have
a way to figure it out."
Why even bother with an isochron? Simply do your calculations for the
ratios, arrive at oldD, and then proceed with the generic method of
subtracting oldD from total D to get newD.
I think, as usual, I'm missing something.
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
>>not at all. I agree that it is possible for you to figure out oldD/Di
>>-- not oldD -- given certain assumptions about Di.
>
>Er, actually, once you've measured Di for a set of samples and
>calculated oldD/Di for all the samples, the value of oldD for each
>sample is easily obtained. Just multiply oldD/Di by the value of Di
>for each sample. the Di's cancel out, leaving oldD.
hmmm...let me think on that one. The claim has been made all along
that it is not possible to know oldD; that only oldD/Di can be known.
So ratios are the means of calculation for the isochron. But I thought
it has been said all along that oldD quantities themselves are not
known.
I know what you're saying, that if you know oldD/Di quantities, then
if you multiply that ratio, you should get quantity of oldD, but then
why the insistence that since oldD is unknown, the isochron has to be
based on ratios only? I would think that the declaration would be
that we know oldD quantities in a rock simply by using ratios.
----
zoe
Jon Fleming
wrote:
>On Fri, 9 Aug 2002 17:32:38 +0000 (UTC), Jon Fleming
What's a base?
The line drawn through the points intercepts the Y-axis at 0.282515.
Three of the points plot so close to this Y-intercept that it looks
like they are right on the Y-axis on a graph.
>>Of course, if a mineral happens to take up a lot of P when it
>>solidifies, after some time its totalD/Di ratio will be significantly
>>higher than the world-wide average.
>
>I would think that the only way for there to be as much oldD as the
>samples show presently is for there to have been a lot of P in the
>original rock. But then that means that after a remelt, there would
>still be a lot of P in the melt and all samples would have a
>corresponding high quantity of P in them.
Nope. In a remelt, the P gets evenly distributed. When the material
solidifies again, various minerals take up different amounts of P
depending on their crystal structure.
> Unless you're saying that P
>at solidification is incorporated unevenly so that some samples of the
>rock source would have high quantities of P and other samples have
>very small quantities of P?
Yep.
>if this is the case, a test consisting of a mere two samplings of zrc,
>biot (only one sample, actually), wr, and xt, is not sufficient to
>tell us if we have an accurate picture of oldD/Di ratios, is it?
There are seven different samples (a fairly typical number ... small
enough so the testing isn't too expensive and large enough to get good
statistical significance). That number is sufficient.
The fact that P is taken up in different amounts when minerals
solidify is _totally_ different from the fact that oldD/Di is the same
for all samples.
You are failing to understand solidification.
Minerals take up different amounts of P, and different amounts of D,
and different amounts of Di, when they solidify. However, minerals
_always_ take up D and Di in the same ratio because D and Di are
chemically identical. (Maybe a mineral takes up only a little D, but
if so it also only takes up a little Di; maybe a mineral takes up lots
of D, but if so it also takes up a lot of Di).
Solidification is a chemical process. What atoms get incorporated in
the various minerals as they solidify to make up the rock depends on
the chemical properties of the atoms and the crystalline structure of
the particular minerals.
Since D and Di have _exactly_ the same chemical properties, when
minerals solidify D and Di are taken up in the same proportion as they
are in the melt (and the proportion in the melt doesn't change). A
particular mineral may not take up much D and Di, another mineral may
take up lots of D and Di, but the _proportion_ of D to Di in all the
minerals is the same.
Since P has very different chemical properties than D or Di, when
minerals solidify they will take up amounts of P that has nothing
particular to do with the amount of D and Di incorporated. A mineral
may take up very little D and Di and P, it may take up very little D
and Di but lots of P, it may take up lots of D and Di but very little
P, or it may take up lots of D and Di and P.
>>And ... it's easier to measure
>>accurately if there's (relatively) a lot of P and D in the sample, so
>>scientists tend to select minerals for sampling that they expect to
>>have a lot of P (and consequently a lot of totalD).
>
>is a sampling of two sufficient to establish a pattern?
Where do you get your "sampling of two"? There are seven samples.
All these samples contribute to the analysis. There are five distinct
points on the plot (three of the points plot so close together that
they can't be distinguished). That is enough to establish a pattern.
>>>Those ratios do not seem to
>>>indicate what 4.5 billion years worth of accumulated oldD should be in
>>>relation to its daughter isotope which has remained the same over
>>>time, does it, not when the decay rate for Lu is a half-life of 37.8
>>>billion years?
>>
>>Er, the relation is between the oldD and a non-radiogenic (not derived
>>from radioactive decay) isotope, Di. Di is _not_ a daughter isotope.
>
>I thought Di stands for daughter isotope?
Not really. Di stands for an isotope that is _not_ a radiogenic
daughter but is chemically the same as the daughter. From the FAQ:
"Isochron dating requires a fourth measurement to be taken, which is
the amount of a different isotope of the same element as the daughter
product of radioactive decay. (For brevity's sake, hereafter I will
refer to the parent isotope as P, the daughter isotope as D, and the
non-radiogenic isotope of the same element as the daughter, as Di"
>The way I understood it was
>that P stood for parent, D stood for daughter, and Di stood for
>daughter isotope -- an isotope that is arranged exactly like the
>daughter except for one of its electrons or something like that.
He didn't say explicitly that Di stands for daughter isotope, but it
probably does. More explicitly, it means "isotope of the daughter
element that is not produced by radioactive decay".
BTW, isotopes differ in the arrangement of their _nuclei_,
specifically the number of neutrons. All isotopes of an element have
the same number of protons in their nuclei, and the same number of
electrons in the same arrangement outside the nucleus. That's why
isotopes are chemically identical; the number and arrangement of the
electrons determines the chemical properties.
>>You can't predict what the ratio of oldD to Di should be from
>>radioactive half-lives alone; you need to know something about the
>>amount of Di now or previously
>
>Di (now or previously) always remains the same, doesn't it?
On average, yes. In detail no; it moves around in molten rock. It
mostly stays the same in solid rock (and we are pretty good at
detecting when it didn't).
>>and the amount of P that is or was
>>available to decay to oldD.
>
>and that, I take it, is unknown.
In detail, yes. The average is known.
>>You would expect that younger rocks would have a larger odlD/Di ratio
>>than older rocks.
>
>I'm not sure I understand what you mean here.
Assume for the moment that there was no D present when the Earth
solidified, but there was some Di. Any mineral that solidified at
that time would have an oldD/Di ratio of zero. A few billion years
later, some P would have decayed to D, so any rock that solidified at
that time would have some oldD and its oldD/Di ratio would be greater
than zero. A few billion years after that, some more P would have
decayed to D and any rock that solidified at that time would have even
more oldD and its oldD/Di ratio would be even more greater than zero.
Of course, there _was_ some D when the Earth solidified, and not all
melts have the average D/Di ratio, but these are details that don't
destroy the overall trend. As time passes and P decays to D. the
average D/Di ratio of the Earth increases, and the chances are that
younger rocks have a higher oldD/Di ratio than older rocks because
there was more D around when the younger rocks solidified.
> Larger in what sense?
Numerically. In the sense that 10 is larger than 5.
>Younger rocks have a wider difference?
Nope. Younger rocks tend to have a larger oldD/Di ratio than older
rocks. They may have wider differences in absolute numbers (I just
don't know), but they should have about the same percentage
differences as older rocks.
> And if so, in which direction?
Huh?
>>So, if you want to look at oldD/Di ratios and figure out if they're
>>reasonable, you'll have to look at several real world data sets that
>>measured rocks that came out to be very different ages.
>
>I guess I'll have to do a google for that. Do you have any
>suggestions of good search terms I could use to find such data?
Well, "isochron" "Rb" and "Sr" might work. You could also look in
Dalrymple's "The Age of the Earth" (see
<http://makeashorterlink.com/?Z30B21C71>; it's a little over ten years
old and doesn't have data form the last decade, but it has a lot of
graphs and references.
>Thanks.
>
>----
>zoe
Eric Rowley
> <host...@sky.fit.qut.edu.au> wrote:
> >zoe_althrop <muz...@aol.com> wrote:
> >> On Tue, 6 Aug 2002 15:01:18 +0000 (UTC), >>
> harshman....@sjm.infi.net (John Harshman) wrote:
<snip>
> >Note that D/Di and P/Di are calculated from measurements of D,
> >Di and P. I leave it as a mathematical exercise to
> >figure out oldD from oldD/Di
> then there should never have been the insistence that "We don't
> know what oldD is."
Was there such an insistence?
There was certainly an insistence that we didn't _need_ to know it.
And we _don't_ know it until after we've done the isochron
calculation, so therefor we don't know it during the calculation.
> The given would be that "We know oldD because
> we have a way to figure it out."
ITYM we can _find out_ what OldD is if we want to because
we have a way to figure it out."
> Why even bother with an isochron?
What's the bother?
Just plug in the values of P, D and Di and get the answer (age)
> Simply do your calculations for
> the ratios, arrive at oldD, and then proceed with the generic
> method of subtracting oldD from total D to get newD.
That's a much more roundabout way of getting the age,
and we aren't really all that interested in OldD.
> I think, as usual, I'm missing something.
Eric
--
<my domain is rixtele>
Jon Fleming
wrote:
>On Fri, 9 Aug 2002 17:34:39 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>snip>
>
>>>not at all. I agree that it is possible for you to figure out oldD/Di
>>>-- not oldD -- given certain assumptions about Di.
>>
>>Er, actually, once you've measured Di for a set of samples and
>>calculated oldD/Di for all the samples, the value of oldD for each
>>sample is easily obtained. Just multiply oldD/Di by the value of Di
>>for each sample. the Di's cancel out, leaving oldD.
>
>hmmm...let me think on that one. The claim has been made all along
>that it is not possible to know oldD; that only oldD/Di can be known.
Not at all. Nobody has made that claim. Many people have claimed
that oldD is not known before applying the isochron method. The
"before applying the isochron method" was always there, explicitly or
implicitly.
>So ratios are the means of calculation for the isochron.
Ratios are the means of calculation for the isochron, but the reason
is that the calculation doesn't work without ratios.
>But I thought
>it has been said all along that oldD quantities themselves are not
>known.
They are not known directly from the measurements. Part of the
process of doing the calculation with ratios is producing the oldD/Di
ratio (which can't be measured directly). Once you have finished the
calculation with the ratios, then you have the information required to
continue to calculate and get the oldD quantities (if you care).
Go back to the data with which you started this thread:
A B C D
Total D/Di 50:4 85:8 120:12 195:24
Present P/Di 10:4 15:8 20:12 25:24
Do you see any oldD quantities or oldD/Di ratios there? Of course you
don't. The _information_ is there, but it's hidden by the form in
which the data is presented (and that's about the only form in which
we can present it to start with, because that's what we measure).
However, _after_ we apply the isochron method we can present the
oldD/Di ratio and, if we wish, the oldD quantities.
We could build a computer program or spreadsheet in which you type in
the measured P, D, and Di quantities and get the oldD and newD
quantities out, without your ever seeing a ratio ... but the ratios
would still be ther in the program, just hidden, because using the
ratios and the isochron method is the only way to get the answer.
>I know what you're saying, that if you know oldD/Di quantities, then
>if you multiply that ratio, you should get quantity of oldD, but then
>why the insistence that since oldD is unknown, the isochron has to be
>based on ratios only?
Because oldD/Di is unknown _before_ _you_ _apply_ _the_ _isochron_
_method_.
> I would think that the declaration would be
>that we know oldD quantities in a rock simply by using ratios.
You could say that ... but people are seldom interested in the oldD
quantities (although sometimes they are), so they seldom say that.
Most people are interested in the age since solidification and how
good the data is.
Eric Rowley
> snip>
> >>not at all. I agree that it is possible for you to figure out
> >>oldD/Di --not oldD --given certain assumptions about Di.
> >Er, actually, once you've measured Di for a set of samples and
> >calculated oldD/Di for all the samples, the value of oldD for
> >each sample is easily obtained. Just multiply oldD/Di by the
> >value of Di for each sample. the Di's cancel out, leaving oldD.
> hmmm...let me think on that one. The claim has been made all
> along that it is not possible to know oldD;
I don't think so, it has been claimed that we _don't_ know OldD,
not that we can't. (Someone may well have said that there is no way
of knowing OldD except by doing the isochron calculations)
> that only oldD/Di can be known. So ratios are the means of
> calculation for the isochron. But I thought it has been said all
> along that oldD quantities themselves are not known.
They aren't!
But they can be calculated and I'm sure that has been said.
Then again maybe not, since it is so obvious that OldD/Di * Di = OldD
nobody may have thought it worth mentioning, specially since the
point of isochron dating is to get the age, not OldD.
> I know what you're saying, that if you know oldD/Di quantities,
> then if you multiply that ratio, you should get quantity of oldD,
> but then why the insistence that since oldD is unknown,
at the start of the calculations OldD _is_ unknown
(so is OldD/Di of course but at least we know it is the same in all
samples)
> the isochron has to be based on ratios only? I would think that the
> declaration would be that we know oldD quantities in a rock
> simply by using ratios.
_After_ we have done the calculations nessesary for the isochron
we can continue and calculate OldD, NewD and OriginalP, (if we want
to).
Jon Fleming
wrote:
>On Fri, 9 Aug 2002 01:25:25 +0000 (UTC), Chris Ho-Stuart
><host...@sky.fit.qut.edu.au> wrote:
>
>>zoe_althrop <muz...@aol.com> wrote:
>>> On Tue, 6 Aug 2002 15:01:18 +0000 (UTC),
>>> harshman....@sjm.infi.net (John Harshman) wrote:
>>>
>>> snip>
>>>
>>>>Just to clarify: do you now agree that the geologic time scale is correct,
>>>>that life is millions of years old, and that the sorting of the fossil
>>>>record shows changes in the earth's biota over hundreds of millions of
>>>>years?
>>>
>>> not at all. I agree that it is possible for you to figure out oldD/Di
>>> -- not oldD -- given certain assumptions about Di.
>>
>>Note that D/Di and P/Di are calculated from measurements
>>of D, Di and P. I leave it as a mathematical exercise to
>>figure out oldD from oldD/Di
>
>then there should never have been the insistence that "We don't know
>what oldD is." The given would be that "We know oldD because we have
>a way to figure it out."
That's what was really meant, explicitly or implicitly. I seem to
recall that this was pretty clear to me.
>Why even bother with an isochron? Simply do your calculations for the
>ratios,
Er. "do your calculations for the ratios" means "use the isochron
method". So you are really saying "Why even bother with an isochron?
Simply use the isochron method ...", which doesn't make much sense.
I think that you are asking "why do a graph?" As we have pointed out
many times, the graph is not an integral part of the method. It is
always done _after_ doing the calculations with the ratios. The
reason _why_ it is done is that humans are very good at recognizing
patterns from a visual presentation. By the time you do the graph,
the numbers have all been calculated.
Do the points lie on a straight line? That question is formally
answered in the calculations, when a number called the "correlation
coefficient" is calculated, and that correlation coefficient tells us
_exactly_ how close the points are to lying on a straight line. But
people still like to see a graph.
Look at the graph of the real-world data at
<http://www.hoxnet.com/se2919609001.jpeg>. You can tell instantly
that the plotted points are damned close to lying on a straight line,
and you don't need any stinkin' correlation coefficient. You know
_in_ _your_ _gut_ that those points lie on a straight line because
you've _seen_ it.
Why did you ask to see a graph of the data with which you started this
thread, even after the answers had all been posted? Because people
like to look at graphs and are good at extracting approximate (not
exact) information form them.
That's what the graph is for.
>arrive at oldD, and then proceed with the generic method of
>subtracting oldD from total D to get newD.
>
>I think, as usual, I'm missing something.
Yup. The "calculations for the ratios" that you know we need produce
the age. That is, by the time we know oldD we also know the age. So,
yes, you could subtract out oldD to get newD and use the generic
method to obtain the age that you already know, but it would be a
waste of time. (We have proved and demonstrated many times that you
get the same answer)
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
zoe copy and pasted:
>>Sample Lu (ppm) Hf (ppm) 176Lu/177Hf 176Hf/177Hf
>>---------------------------------------------------------
>>zrc-1 174.3 15,330 0.001614 0.282363
>>zrc-2 74.22 10,430 0.001010 0.282538
>>biot-2 0.7598 14.41 0.007484 0.282597
>>wr-1 1243 50.61 3.529 0.349371
>>wr-2 1315 73.46 2.564 0.330715
>>xt-1 2948 16.20 28.35 0.816178
>>xt-2 3373 19.82 26.36 0.779948
>>
>>P : 176Lu
>>D : 176Hf
>>Di: 177Hf
>>
>>176Lu beta decays to 176Hf (stable) with a half-life (t1/2) of
>>3.78E10y
>>(37.8 billion years).
snip>
>>>The single oldD/Di value was 0.282.
>>
>>you mean the base of 0.282515?
>
>What's a base?
the zero slope line that represents oldD/Di, or time at
solidification.
snip>
>>I would think that the only way for there to be as much oldD as the
>>samples show presently is for there to have been a lot of P in the
>>original rock. But then that means that after a remelt, there would
>>still be a lot of P in the melt and all samples would have a
>>corresponding high quantity of P in them.
>
>Nope. In a remelt, the P gets evenly distributed.
that's what I mean. If there were a lot of P at the time of formation
of Earth's original rocks, then at remelt, what P is left is evenly
distributed according to the type of mineral that incorporates it.
Therefore, should samples today have high quantities of P in relation
to Di, as reflected by high quantities of oldD/Di?
>When the material
>solidifies again, various minerals take up different amounts of P
>depending on their crystal structure.
>
>> Unless you're saying that P
>>at solidification is incorporated unevenly so that some samples of the
>>rock source would have high quantities of P and other samples have
>>very small quantities of P?
>
>Yep.
>
>>if this is the case, a test consisting of a mere two samplings of zrc,
>>biot (only one sample, actually), wr, and xt, is not sufficient to
>>tell us if we have an accurate picture of oldD/Di ratios, is it?
>
>There are seven different samples (a fairly typical number ... small
>enough so the testing isn't too expensive and large enough to get good
>statistical significance). That number is sufficient.
I meant, the test consists of only two samples of each type of
mineral. Is two sufficient to get an idea of the quantity of P that
is characteristic of that particular type of mineral?
>The fact that P is taken up in different amounts when minerals
>solidify is _totally_ different from the fact that oldD/Di is the same
>for all samples.
I understand that. But if quantity of oldD is fairly high in relation
to Di, and we have not yet made it to even 0.116 percent of the half
life of Lutetium (if that's what Lu stands for), this would indicate a
high quantity of P, right? Either that, or more time has passed for
this mineral than even the age of the universe.
and Jon, using the figures above, would you tell me the original P for
xt-1 and 2? Present P/Di of 28.36 plus total D/Di of 0.816178 =
29.166178, which looks nowhere near what original P looks like on this
diagram. From what I can see, the points don't fit together, going
from the X-axis to the Y-axis. See:
http://www.hoxnet.com/se2919609001.jpeg
why is that? At least, there's not the kind of trajectory for that
sample that would be expected if original P/Di is decaying to D/Di.
>You are failing to understand solidification.
>
>Minerals take up different amounts of P, and different amounts of D,
>and different amounts of Di, when they solidify. However, minerals
>_always_ take up D and Di in the same ratio because D and Di are
>chemically identical. (Maybe a mineral takes up only a little D, but
>if so it also only takes up a little Di; maybe a mineral takes up lots
>of D, but if so it also takes up a lot of Di).
>
>Solidification is a chemical process. What atoms get incorporated in
>the various minerals as they solidify to make up the rock depends on
>the chemical properties of the atoms and the crystalline structure of
>the particular minerals.
>
>Since D and Di have _exactly_ the same chemical properties, when
>minerals solidify D and Di are taken up in the same proportion as they
>are in the melt (and the proportion in the melt doesn't change). A
>particular mineral may not take up much D and Di, another mineral may
>take up lots of D and Di, but the _proportion_ of D to Di in all the
>minerals is the same.
thanks for the explanation. So, does oldD/Di's quantity reflect either
the presence of a LOT of P at original formation of the rock, or the
passing of time much older than 15 billion years?
>Since P has very different chemical properties than D or Di, when
>minerals solidify they will take up amounts of P that has nothing
>particular to do with the amount of D and Di incorporated.
is quantity of oldD a reflection of quantity of originalP that has
decayed up to point of remelt?
>A mineral
>may take up very little D and Di and P, it may take up very little D
>and Di but lots of P, it may take up lots of D and Di but very little
>P, or it may take up lots of D and Di and P.
but relatively speaking, if there were a lot of P at the beginning,
the quantities taken up characteristically by the various types of
minerals would be large compared to if we were down at the
near-exhaustion point of P, in time, right?
>>>And ... it's easier to measure
>>>accurately if there's (relatively) a lot of P and D in the sample, so
>>>scientists tend to select minerals for sampling that they expect to
>>>have a lot of P (and consequently a lot of totalD).
>>
>>is a sampling of two sufficient to establish a pattern?
>
>Where do you get your "sampling of two"?
two of each type -- though I don't know why there is only one of
biotite (is that what "biot" stands for?). Wouldn't you need more
than two samples of a type in order to determine the characteristic
amount of P they would tend to absorb? How widely does P vary in each
sample of this particular mineral type?
>There are seven samples.
>All these samples contribute to the analysis. There are five distinct
>points on the plot (three of the points plot so close together that
>they can't be distinguished). That is enough to establish a pattern.
and this pattern is what?
snip>
>BTW, isotopes differ in the arrangement of their _nuclei_,
>specifically the number of neutrons. All isotopes of an element have
>the same number of protons in their nuclei, and the same number of
>electrons in the same arrangement outside the nucleus. That's why
>isotopes are chemically identical; the number and arrangement of the
>electrons determines the chemical properties.
so is the variation only in the neutrons, not the electrons? And if
the number and arrangement of the electrons change, then the element
is no longer the same?
>>>You can't predict what the ratio of oldD to Di should be from
>>>radioactive half-lives alone; you need to know something about the
>>>amount of Di now or previously
>>
>>Di (now or previously) always remains the same, doesn't it?
>
>On average, yes. In detail no; it moves around in molten rock. It
>mostly stays the same in solid rock (and we are pretty good at
>detecting when it didn't).
I meant, the QUANTITY of Di always remains the same, doesn't it? It
never becomes more, it never becomes less?
snip>
>>>You would expect that younger rocks would have a larger odlD/Di ratio
>>>than older rocks.
>>
>>I'm not sure I understand what you mean here.
>
>Assume for the moment that there was no D present when the Earth
>solidified, but there was some Di. Any mineral that solidified at
>that time would have an oldD/Di ratio of zero. A few billion years
>later, some P would have decayed to D, so any rock that solidified at
>that time would have some oldD and its oldD/Di ratio would be greater
>than zero. A few billion years after that, some more P would have
>decayed to D and any rock that solidified at that time would have even
>more oldD and its oldD/Di ratio would be even more greater than zero.
what would that number look like, as an example? What would a ratio
of zero look like, and what would a ratio greater than zero look like?
Would the first be 0:1 and the second be 1:1, and the third be 2:1?
>Of course, there _was_ some D when the Earth solidified, and not all
>melts have the average D/Di ratio, but these are details that don't
>destroy the overall trend. As time passes and P decays to D. the
>average D/Di ratio of the Earth increases, and the chances are that
>younger rocks have a higher oldD/Di ratio than older rocks because
>there was more D around when the younger rocks solidified.
oh, okay, that makes sense.
>
>> Larger in what sense?
>
>Numerically. In the sense that 10 is larger than 5.
no, I meant, which number should be larger in the ratio if 38.7
billion years have passed, the D or the Di, at the time of
measurement?
>>Younger rocks have a wider difference?
>
>Nope. Younger rocks tend to have a larger oldD/Di ratio than older
>rocks. They may have wider differences in absolute numbers (I just
>don't know), but they should have about the same percentage
>differences as older rocks.
>
>> And if so, in which direction?
>
>Huh?
I meant, is the Di on the lesser side of the ratio now or still on the
greater side. Like maybe there was a time when the ratio was 1:10 but
now it is 10:1.
Please don't begin to get frustrated yet. You must be wondering where
I'm going with all this. Just consider it to be knowledge-gathering.
>>>So, if you want to look at oldD/Di ratios and figure out if they're
>>>reasonable, you'll have to look at several real world data sets that
>>>measured rocks that came out to be very different ages.
>>
>>I guess I'll have to do a google for that. Do you have any
>>suggestions of good search terms I could use to find such data?
>
>Well, "isochron" "Rb" and "Sr" might work. You could also look in
>Dalrymple's "The Age of the Earth" (see
><http://makeashorterlink.com/?Z30B21C71>; it's a little over ten years
>old and doesn't have data form the last decade, but it has a lot of
>graphs and references.
thanks. Looked at the few pages of The Age of the Earth shown on that
site, and haven't decided yet if I'll purchase it or not.
----
zoe
Eric Rowley
<snip>
> that's what I mean. If there were a lot of P at the time of
> formation of Earth's original rocks, then at remelt, what P is
> left is evenly distributed according to the type of mineral that
> incorporates it.
You´re assuming that all the contents of the rock stay together
for all time, rock forms, rock melts, rock solidifies.
That isn't nessesarily how it works, if the rock erodes, P is
water soluble and D isn't then the D might collect in the river
bends while the P is washed out to sea and ends up in limestone
hundreds of miles away. If, 100 million years later both the
river sediment and the limestone are melted and form new rocks
then one rock will have "too much" P and the other "too much" D
> Therefore, should samples today have high
> quantities of P in relation to Di, as reflected by high
> quantities of oldD/Di?
Not nessesarily, the "extra" P could have ended up in other
rocks so you would have to grind up the whole earth and measure
P and D but even that wouldn't help,
there could have been "too much" D when the earth was originaly
formed and the "extra" P could have ended up in the other planets.
Or there could have been "too much" D when the solar system formed,
the supernova/god that created the P may well have created some D
as well so just comparing amounts of P and D does not tell you the
age of anything unless you have some knowledge of the original
amounts!
<snip>
> I meant, the test consists of only two samples of each type of
> mineral. Is two sufficient to get an idea of the quantity of P
> that is characteristic of that particular type of mineral?
Why should we care what "quantity of P that is characteristic of
that particular type of mineral"?
It seems like you are trying to sneak your "accumulated age product"
back into the equation?
What we need for the age calculation are P, D and Di measurements
for at least two (usualy at least 6 are used) samples (with
differing P/Di ratios) and the knowledge that OldD/Di ratios
are the same for all samples from the same melt.
We don't need "characteristic" quantities of anything.
> >The fact that P is taken up in different amounts when minerals
> >solidify is _totally_ different from the fact that oldD/Di is
> >the same for all samples.
> I understand that. But if quantity of oldD is fairly high in
> relation to Di, and we have not yet made it to even 0.116 percent
> of the half life of Lutetium (if that's what Lu stands for), this
> would indicate a high quantity of P, right? Either that, or more
> time has passed for this mineral than even the age of the universe.
Or some of the OldD was there from the begining.
> and Jon, using the figures above, would you tell me the original
> P for xt-1 and 2? Present P/Di of 28.36 plus total D/Di of
> 0.816178 = 29.166178,
What does P/Di + TotalD/Di have to do with anything?
You want P/Di + NewD/Di surely?
You have to subtract OldD/Di
29.166 - .282 = 28.88
(no point in using more decimals then in the number with the
least decimals)
And for xt-2 NewD = .780 - .282 = .498
26.36 - .489 = 25.87
> which looks nowhere near what original P
> looks like on this diagram.
Zoe, the axis on this diagram are not drawn at the same scale,
so the trajectories don't go at a 45 degree angle, they will
be almost vertical.
>From what I can see, the points don't
> fit together, going from the X-axis to the Y-axis. See:
The trajectories shouldn't be going from the X-axis!
They go from the OldD/Di line!
> http://www.hoxnet.com/se2919609001.jpeg
> why is that? At least, there's not the kind of trajectory for
> that sample that would be expected if original P/Di is decaying
> to D/Di.
1 NewD is equilivant to 1 OriginalP, right?
Read the scales on the graph, 1 on the Y-axis is about 3 inches
(on my screen), 1 on the X-axis is about 1/10 of an inch.
<snip>
> thanks for the explanation. So, does oldD/Di's quantity reflect
> either the presence of a LOT of P at original formation of the
> rock, or the passing of time much older than 15 billion years?
No, it reflects the amounts of D and Di present at solidification.
Naturaly the OldD has to come from somewhere, but not nessesarily
from the decay of P, and even for the D that does come from decay
of P, the remaining potion of that P doesn't have to be anywhere
near the D.
> >Since P has very different chemical properties than D or Di,
> >when minerals solidify they will take up amounts of P that has
> >nothing particular to do with the amount of D and Di
> >incorporated.
> is quantity of oldD a reflection of quantity of originalP that
> has decayed up to point of remelt?
No, it's a reflection of quantity of originalP plus quantity of
"originalOldD" and you have to remember that your "original rock"
can be spread over the entire earth.
> >A mineral
> >may take up very little D and Di and P, it may take up very
> >little D and Di but lots of P, it may take up lots of D and Di
> >but very little P, or it may take up lots of D and Di and P.
> but relatively speaking, if there were a lot of P at the
> beginning, the quantities taken up characteristically by the
> various types of minerals would be large compared to if we were
> down at the near-exhaustion point of P, in time, right?
I think it's more like the more P there is in the melt the more
you get of the minerals that contain a lot of P
<snip>
> >There are seven samples.
> >All these samples contribute to the analysis. There are five
> >distinct points on the plot (three of the points plot so close
> >together that they can't be distinguished). That is enough to
> >establish a pattern.
> and this pattern is what?
The isochron line?
> snip>
> >BTW, isotopes differ in the arrangement of their _nuclei_,
> >specifically the number of neutrons. All isotopes of an element
> >have the same number of protons in their nuclei, and the same
> >number of electrons in the same arrangement outside the nucleus.
> >That's why isotopes are chemically identical; the number and
> >arrangement of the electrons determines the chemical properties.
> so is the variation only in the neutrons, not the electrons? And
> if the number and arrangement of the electrons change, then the
> element is no longer the same?
The electrons have nothing to do with what element an atom is.
The number of protons in the nucleus determines the element
and also how many electrons the atom will normaly have,
The number of neutrons in the nucleus determines which isotope
you have, how stable or unstable it is.
Protons and neutrons together determine the atomic weight,
they weigh almost the same and the electron is much, much lighter.
Neutrons are neutral (electricly uncharged), protons have a
positive charge and electrons are negetive, that's why an
uncharged atom has the same number of electrons as protons.
An uncharged atom is what we normaly call an atom, if it has
a charge, more or less electrons, it's called an ion but given
a chance an ion will even out it's charge by altering the number
of electrons. Ions are what make batteries work, they chemicaly
remove electrons from atoms and force them to go though the
electrical circuit to get back to the charged ions and
neutralize them.
The electrons do all the chemical stuff, reacting with other
atoms and so forth.
If you rearange the electrons without changing how many their
are you change the energy stored in the atom, that's how
lasers and flouresent tubes work, the electrons are electricly
excited so they change to a more energetic state and when they
change back again they give of light.
Anyways the electrons are irrelevant to isochron dating.
> >>>You can't predict what the ratio of oldD to Di should be from
> >>>radioactive half-lives alone; you need to know something about
> >>>the amount of Di now or previously
> >>Di (now or previously) always remains the same, doesn't it?
> >On average, yes. In detail no; it moves around in molten rock.
> >It mostly stays the same in solid rock (and we are pretty good
> >at detecting when it didn't).
> I meant, the QUANTITY of Di always remains the same, doesn't it?
> It never becomes more, it never becomes less?
It would probaly be created in supernovas all the time,
but here on earth the quantity should be pretty near constant
since the earths formation.
So the total quantity on earth stays the same, that doesn't mean
that is the same in all rocks.
> snip>
> >>>You would expect that younger rocks would have a larger
> >>>odlD/Di ratio than older rocks.
> >>I'm not sure I understand what you mean here.
> >Assume for the moment that there was no D present when the Earth
> >solidified, but there was some Di. Any mineral that solidified
> >at that time would have an oldD/Di ratio of zero. A few billion
> >years later, some P would have decayed to D, so any rock that
> >solidified at that time would have some oldD and its oldD/Di
> >ratio would be greater than zero. A few billion years after
> >that, some more P would have decayed to D and any rock that
> >solidified at that time would have even more oldD and its
> >oldD/Di ratio would be even more greater than zero.
> what would that number look like, as an example? What would a
> ratio of zero look like, and what would a ratio greater than zero
> look like? Would the first be 0:1 and the second be 1:1, and the
> third be 2:1?
Yes, something like that.
> >Of course, there _was_ some D when the Earth solidified, and not
> >all melts have the average D/Di ratio, but these are details
> >that don't destroy the overall trend. As time passes and P
> >decays to D. the average D/Di ratio of the Earth increases, and
> >the chances are that younger rocks have a higher oldD/Di ratio
> >than older rocks because there was more D around when the
> >younger rocks solidified.
> oh, okay, that makes sense.
> >
> >> Larger in what sense?
> >
> >Numerically. In the sense that 10 is larger than 5.
> no, I meant, which number should be larger in the ratio if 38.7
> billion years have passed, the D or the Di, at the time of
> measurement?
D, since D changes and Di doesn't it would have to be D
(Also if Di _did_ change it would have to get _smaller_
to make D/Di larger)
<snip>
Eric Rowley
ER> From: muz...@aol.com (zoe_althrop)
<snip>
ER> > and Jon, using the figures above, would you tell me the
ER> >original P for xt-1 and 2? Present P/Di of 28.36 plus total
ER> >D/Di of 0.816178 = 29.166178,
ER> What does P/Di + TotalD/Di have to do with anything?
ER> You want P/Di + NewD/Di surely? You have to subtract OldD/Di
ER> 29.166 - .282 = 28.88 (no point in using more decimals then in
ER> the number with the least decimals)
ER> And for xt-2 NewD = .780 - .282 = .498
ER> 26.36 - .489 = 25.87
That should of course have been
26.36 + .489 = 26.85 for OriginalP
Jon Fleming
wrote:
>On Sat, 10 Aug 2002 19:37:44 +0000 (UTC), Jon Fleming
Yes.
>according to the type of mineral that incorporates it.
I'm not sure what you mean here. In a remelt there _are_ no minerals.
Minerals are formed at solidification. At solidification, the P is
_unevenly_ distributed; some mineral take up lots of P, some don't.
>Therefore, should samples today have high quantities of P in relation
>to Di, as reflected by high quantities of oldD/Di?
Some minerals take up lots of P, some don't. Some minerals take up
lots of D, some don't. There's no significant relationship between
the groups.
>>When the material
>>solidifies again, various minerals take up different amounts of P
>>depending on their crystal structure.
>>
>>> Unless you're saying that P
>>>at solidification is incorporated unevenly so that some samples of the
>>>rock source would have high quantities of P and other samples have
>>>very small quantities of P?
>>
>>Yep.
>>
>>>if this is the case, a test consisting of a mere two samplings of zrc,
>>>biot (only one sample, actually), wr, and xt, is not sufficient to
>>>tell us if we have an accurate picture of oldD/Di ratios, is it?
>>
>>There are seven different samples (a fairly typical number ... small
>>enough so the testing isn't too expensive and large enough to get good
>>statistical significance). That number is sufficient.
>
>I meant, the test consists of only two samples of each type of
>mineral. Is two sufficient to get an idea of the quantity of P that
>is characteristic of that particular type of mineral?
When we are analyzing the age, we don't care about the amount of P
that is characteristic of that particular type of mineral. We care
about how much P is in the mineral.
Of course, we like to know how much P is characteristic of a
particular mineral; it helps us select samples for age determination.
But the set of samples that we can use to experimentally determine how
much P is characteristic of a particular mineral is _all_ samples of
that mineral that have _ever_ been gathered and analyzed and reported.
Lots more than two.
When we are analyzing how much P is characteristic of a particular
mineral, we also look at it from a theoretical point of view,
analyzing the crystal structure and its chemical properties.
>>The fact that P is taken up in different amounts when minerals
>>solidify is _totally_ different from the fact that oldD/Di is the same
>>for all samples.
>
>I understand that. But if quantity of oldD is fairly high in relation
>to Di, and we have not yet made it to even 0.116 percent of the half
>life of Lutetium (if that's what Lu stands for), this would indicate a
>high quantity of P, right?
Not necessarily. Nobody ever claimed that oldD is produced _only_ by
radioactive decay; only that it is produced only by radioactive decay
_on_ _the_ _Earth_. It is virtually certain that oldD is produced in
supernovae just like most of the other elements, and was present when
the Earth formed. (The conditions necessary to produce oldD by fusion
are not possible on a solid or even molten planet)
>Either that, or more time has passed for
>this mineral than even the age of the universe.
>
>and Jon, using the figures above, would you tell me the original P for
>xt-1 and 2? Present P/Di of 28.36 plus total D/Di of 0.816178 =
>29.166178, which looks nowhere near what original P looks like on this
>diagram.
Of course. Adding totalD/Di to presentP/Di makes no sense, so of
course the answer makes no sense.
One atom of P decays to one atom of D. Therefore, newD/Di =
decayedP/Di. OriginalP/Di is presentP/Di plus decayedP/Di. Therefore,
originalP/Di can be calculated by:
originalP/Di = presentP/Di + decayedP/Di
But, since decayedP/Di = newD/Di
originalP/Di = presentP/Di + newD/Di
Or, since newD/Di = totalD/Di - oldD/Di
originalP/Di = presentP/Di + totalD/Di - oldD/Di
You added totalD/Di but didn't subtract oldD/Di
So for sample xt-1
originalP/Di = 28.35 + 0.816178 - 0.282515
originalP/Di = 28.88366
For sample xt-2:
originalP/Di = 26.36 + 0.779948 - 0.282515
originalP/Di = 26.85743
>From what I can see, the points don't fit together, going
>from the X-axis to the Y-axis.
Well, of course (as we've said many times before) they don't "go from
the X-axis to the Y-axis". The trajectories _start_ at the point
where the samples would have plotted when they solidified. This is
always on what you are calling the "base", the horizontal line drawn
from the Y-intercept. For example, for sample xt-1, we now know that
when it solidified it had 0.282515 oldD/Di and 28.88366 originalP/Di,
and before it solidified it didn't exist, so its trajectory starts at
P/Di = 28.88366 and D/Di = 0.282515.
then the trajectories extend to the current plotted position of the
point, and can be extended up to the Y-axis.
But a trajectory starting at ten X-axis makes no physical sense unless
oldD is zero (and in this case oldD is not zero).
> See:
>
>http://www.hoxnet.com/se2919609001.jpeg
>
>why is that? At least, there's not the kind of trajectory for that
>sample that would be expected if original P/Di is decaying to D/Di.
You're going to have to be more specific. When I draw the
trajectories, I see the trajectories that I expect (see
<http://fleming-group.com/Isochron/Isochron10.html>).
Are you bothered by the fact that the trajectories are so steep?
That's because the original authors (and I) used very different X and
Y scales so the slope of the line would be evident. It's as if we
"stretched" the plot in the up-down-direction. If we hadn't done
that, if we had used equal X and Y scales, the plotted line would look
horizontal even though it does have a small and measurable slope.
Remember that the graph is _not_ really the analysis; it's just a way
of presenting the data so people can easily comprehend it and evaluate
how good a straight line is formed by the points. You pick scales for
the X and Y axes so they fit in the paper and they demonstrate what
you are trying to demonstrate.
(One of the reasons why all our made-up data sets tended to be one or
two half-lives old is that the plots look nice with equal X and Y
scales and can be plotted in messages).
Even though the trajectories _look_ steep, if you calculate their
slope it is -1 as expected. (You expect the slope to be -1 because
P is decreasing and D is increasing, and the ratio of D increase to P
decrease is -1 because one atom of P decays to one atom of D). For
example, for the trajectory line for sample xt-1, the trajectory
starts at D/Di = oldD/Di and P/Di = originalP/DI, and ends at D/Di =
totalD/Di and P/Di = currentP/Di:
slope = (totalD/Di - oldD/Di)/(currentP/Di - originalP/Di)
slope = (0.816178 - 0.282515)/(28.35 - 28.88366)
slope = 0.5336628/(- 0.5336628)
slope = -1
>>You are failing to understand solidification.
>>
>>Minerals take up different amounts of P, and different amounts of D,
>>and different amounts of Di, when they solidify. However, minerals
>>_always_ take up D and Di in the same ratio because D and Di are
>>chemically identical. (Maybe a mineral takes up only a little D, but
>>if so it also only takes up a little Di; maybe a mineral takes up lots
>>of D, but if so it also takes up a lot of Di).
>>
>>Solidification is a chemical process. What atoms get incorporated in
>>the various minerals as they solidify to make up the rock depends on
>>the chemical properties of the atoms and the crystalline structure of
>>the particular minerals.
>>
>>Since D and Di have _exactly_ the same chemical properties, when
>>minerals solidify D and Di are taken up in the same proportion as they
>>are in the melt (and the proportion in the melt doesn't change). A
>>particular mineral may not take up much D and Di, another mineral may
>>take up lots of D and Di, but the _proportion_ of D to Di in all the
>>minerals is the same.
>
>thanks for the explanation. So, does oldD/Di's quantity reflect either
>the presence of a LOT of P at original formation of the rock,
No, because oldD/Di has nothing to do with the amount of P at rock
formation; only newD/Di has something to do with the amount of P at
rock formation.
>or the
>passing of time much older than 15 billion years?
That's one possibility.
The other possibility is that there is another source of oldD, such as
supernovae.
The other possibility is true; supernovae produce oldD, just like they
produce most of the elements that make up the Earth.
>>Since P has very different chemical properties than D or Di, when
>>minerals solidify they will take up amounts of P that has nothing
>>particular to do with the amount of D and Di incorporated.
>
>is quantity of oldD a reflection of quantity of originalP that has
>decayed up to point of remelt?
The amount of oldD is partly a reflection of the amount of P that
decayed before solidification and partly a reflection of how much D
was created by other processes, such as fusion in supernovae.
>>A mineral
>>may take up very little D and Di and P, it may take up very little D
>>and Di but lots of P, it may take up lots of D and Di but very little
>>P, or it may take up lots of D and Di and P.
>
>but relatively speaking, if there were a lot of P at the beginning,
>the quantities taken up characteristically by the various types of
>minerals would be large compared to if we were down at the
>near-exhaustion point of P, in time, right?
The _quantities_, yes, but the _proportions_ (or ratios) between
different minerals would be about the same.
>>>>And ... it's easier to measure
>>>>accurately if there's (relatively) a lot of P and D in the sample, so
>>>>scientists tend to select minerals for sampling that they expect to
>>>>have a lot of P (and consequently a lot of totalD).
>>>
>>>is a sampling of two sufficient to establish a pattern?
>>
>>Where do you get your "sampling of two"?
>
>two of each type -- though I don't know why there is only one of
>biotite (is that what "biot" stands for?). Wouldn't you need more
>than two samples of a type in order to determine the characteristic
>amount of P they would tend to absorb?
Yes. So what? The point of the study was not to determine the
characteristic amount of P that the minerals absorb. The point of the
study was to measure the age of the rock, and the age of the rock does
not depend on how much P the various minerals characteristically
absorb.
>How widely does P vary in each
>sample of this particular mineral type?
Damned if I know.
>>There are seven samples.
>>All these samples contribute to the analysis. There are five distinct
>>points on the plot (three of the points plot so close together that
>>they can't be distinguished). That is enough to establish a pattern.
>
>and this pattern is what?
A single straight isochron line through all the points, with slope
0.018846 and Y-intercept 0.282515.
>snip>
>
>>BTW, isotopes differ in the arrangement of their _nuclei_,
>>specifically the number of neutrons. All isotopes of an element have
>>the same number of protons in their nuclei, and the same number of
>>electrons in the same arrangement outside the nucleus. That's why
>>isotopes are chemically identical; the number and arrangement of the
>>electrons determines the chemical properties.
>
>so is the variation only in the neutrons, not the electrons? And if
>the number and arrangement of the electrons change, then the element
>is no longer the same?
Um, maybe. There's a lot that could be said about this.
It is the _nucleus_, the number of protons and neutrons, that
determines what the element is.
There are two possible ways that the amount of electrons could change.
If electrons are added or taken away, and the nucleus remains the
same, then the element is the same but it now has an overall
electrical charge. That electrical charge makes it very likely that
it will react with some other atom or atoms or something in a way that
leads to it getting back to the original number of electrons and
having no overall electrical charge.
If electrons are added (or taken away) _and_ the same number of
protons are added (or taken away) then the nucleus changes and you
wind up with a different element that has no overall electrical
charge. That element may or may not be stable; it may decay to yet
another element.
If electrons are added (or taken away) _and_ some _different_ number
of protons are added (or taken away), then you have a different
element that has an overall electrical charge. It will have a strong
tendency to react with other atoms to get to a zero overall electrical
charge, and it may decay into yet another element.
>>>>You can't predict what the ratio of oldD to Di should be from
>>>>radioactive half-lives alone; you need to know something about the
>>>>amount of Di now or previously
>>>
>>>Di (now or previously) always remains the same, doesn't it?
>>
>>On average, yes. In detail no; it moves around in molten rock. It
>>mostly stays the same in solid rock (and we are pretty good at
>>detecting when it didn't).
>
>I meant, the QUANTITY of Di always remains the same, doesn't it? It
>never becomes more, it never becomes less?
Mostly it remains the same in solids. It is possible to Di to be
leached out (but we are pretty good at picking samples in which that
hasn't happened).
In a melt, there can be some variation.
>snip>
>
>>>>You would expect that younger rocks would have a larger odlD/Di ratio
>>>>than older rocks.
>>>
>>>I'm not sure I understand what you mean here.
>>
>>Assume for the moment that there was no D present when the Earth
>>solidified, but there was some Di. Any mineral that solidified at
>>that time would have an oldD/Di ratio of zero. A few billion years
>>later, some P would have decayed to D, so any rock that solidified at
>>that time would have some oldD and its oldD/Di ratio would be greater
>>than zero. A few billion years after that, some more P would have
>>decayed to D and any rock that solidified at that time would have even
>>more oldD and its oldD/Di ratio would be even more greater than zero.
>
>what would that number look like, as an example? What would a ratio
>of zero look like,
Huh? A ratio of zero looks like either the number 0 or the ratio
0:(something not zero). For example, if you measured a rock and found
D = 0 and Di = 1, the D/Di ratio would be 0:1 or 0.
>and what would a ratio greater than zero look like?
>Would the first be 0:1 and the second be 1:1, and the third be 2:1?
Those are possible values, but there are an infinite number of
possible values.
>>Of course, there _was_ some D when the Earth solidified, and not all
>>melts have the average D/Di ratio, but these are details that don't
>>destroy the overall trend. As time passes and P decays to D. the
>>average D/Di ratio of the Earth increases, and the chances are that
>>younger rocks have a higher oldD/Di ratio than older rocks because
>>there was more D around when the younger rocks solidified.
>
>oh, okay, that makes sense.
>
>>
>>> Larger in what sense?
>>
>>Numerically. In the sense that 10 is larger than 5.
>
>no, I meant, which number should be larger in the ratio if 38.7
>billion years have passed, the D or the Di, at the time of
>measurement?
It depends on how much D was created by other processes, such as
supernovae, and the half-life of P, and how much P was created by
supernovae. If the half-life of P is large compared to the age of the
universe, then most of the oldD came from supernovae (or whatever) and
the average oldD/Di ratio doesn't change much over time. If the
half-life of P is much less than the age of the universe, and there
was a lot of P created by supernovae or whatever, then most of the
oldD came from decay of P and the average oldD/Di ratio is increasing
faster over time than in the first case.
>>>Younger rocks have a wider difference?
>>
>>Nope. Younger rocks tend to have a larger oldD/Di ratio than older
>>rocks. They may have wider differences in absolute numbers (I just
>>don't know), but they should have about the same percentage
>>differences as older rocks.
>>
>>> And if so, in which direction?
>>
>>Huh?
>
>I meant, is the Di on the lesser side of the ratio now or still on the
>greater side. Like maybe there was a time when the ratio was 1:10 but
>now it is 10:1.
I don't know. It depends on which particular P, D, and Di you are
talking about.
>Please don't begin to get frustrated yet. You must be wondering where
>I'm going with all this. Just consider it to be knowledge-gathering.
I'm wondering, but I've learned that it's impossible to rationally
predict what you'll come up with.
>>>>So, if you want to look at oldD/Di ratios and figure out if they're
>>>>reasonable, you'll have to look at several real world data sets that
>>>>measured rocks that came out to be very different ages.
>>>
>>>I guess I'll have to do a google for that. Do you have any
>>>suggestions of good search terms I could use to find such data?
>>
>>Well, "isochron" "Rb" and "Sr" might work. You could also look in
>>Dalrymple's "The Age of the Earth" (see
>><http://makeashorterlink.com/?Z30B21C71>; it's a little over ten years
>>old and doesn't have data form the last decade, but it has a lot of
>>graphs and references.
>
>thanks. Looked at the few pages of The Age of the Earth shown on that
>site, and haven't decided yet if I'll purchase it or not.
A good library might have it, especially in these days of easy
inter-library loans.
Chris Ho-Stuart
> On Fri, 9 Aug 2002 01:25:25 +0000 (UTC), Chris Ho-Stuart
> <host...@sky.fit.qut.edu.au> wrote:
>
>>zoe_althrop <muz...@aol.com> wrote:
>>> On Tue, 6 Aug 2002 15:01:18 +0000 (UTC),
>>> harshman....@sjm.infi.net (John Harshman) wrote:
>>>
>>> snip>
>>>
>>>>Just to clarify: do you now agree that the geologic time scale is correct,
>>>>that life is millions of years old, and that the sorting of the fossil
>>>>record shows changes in the earth's biota over hundreds of millions of
>>>>years?
>>>
>>> not at all. I agree that it is possible for you to figure out oldD/Di
>>> -- not oldD -- given certain assumptions about Di.
>>
>>Note that D/Di and P/Di are calculated from measurements
>>of D, Di and P. I leave it as a mathematical exercise to
>>figure out oldD from oldD/Di
>
> then there should never have been the insistence that "We don't know
> what oldD is." The given would be that "We know oldD because we have
> a way to figure it out."
That HAS been the insistence all along.
> Why even bother with an isochron? Simply do your calculations for the
> ratios, arrive at oldD, and then proceed with the generic method of
> subtracting oldD from total D to get newD.
As has been pointed out all along, the thing we want to calculate
is the time since the rock solidified. We do this by simply doing
calculations. The graphs are only ever drawn for the sake of
explaining or presenting the calculations in a simple readable form.
The isochron method is mathematically equivalent to calculating
oldD and then using a generic method. But that is a very much a
round about method which obscures the advantages of the isochron
for a student. Particularly when you go on to study the role of
measurement uncertaininty and error bars, you need to work with
the isochron method directly.
> I think, as usual, I'm missing something.
S'OK. You are learning something very foreign to you in various
ways. Here I repeat some material starting from way back in March.
: You measure P, D and Di. After you have measured them, you
: know these quantities. Everything else is calculated from the
: measured values.
:
: So of course the exact quantities of P, D and Di are known
: for each sample. They are measured. And Di is certainly not
: constant; it will vary from sample to sample as well.
:
: What is constant is the value of oldD/Di for each sample, but
: this is a law of physics, (for a rock which has been closed
: since the melt). This is not a new measurement. It is a relation
: used to calculate the unknown quantity oldD, since it cannot
: be measured directly. We can ignore that calculation and just
: focus on getting the slope using the measured quantities.
:
: Of course, if you don't really understand the physics, people will
: try to explain it in various ways.
:
: For example, you were originally saying that the age of the
: isochron has something to do with the age of the Earth, which
: is incorrect. In an effort to explain *why* it is incorrect,
: people have tried to explain how decay changes quantities and
: ratios over time, and why the isochron measurement has nothing
: at all to do with age of the Earth; only the age of the rock
: itself. In doing this, they have brought up newD and oldD,
: to explain the physical processes going on in a rock with
: radioactive minerals.
:
: That is, the reason for introducing newD and oldD is to show
: that *if* you could actually measure only the new D product
: (and we can't, we can only measure total D) then it would make
: no difference to the isochron slope. Hence we the slope does
: NOT depend on how much D was present when the rock was formed,
: and hence the slope does NOT have any relation to the age of
: the earth -- only the age of the rock.
Let me emphasize that. No-one really cares about oldD and newD.
What we care about is the age of the rock. The variables oldD
and newD are really only introduced for the sake of trying to
explain how the methods work.
Here is another statement of the case.
: > so having admitted that we do NOT know newD, how do you arrive at the
: > figures for newD?
:
: By calculating it.
What you are missing, all the time, is that there are some
very simple mathematical statements which capture the
assumptions of being cogenetic, and closed over time.
Everything else follows from those assumption, like 2 follows
from 1 + 1.
Let me try and explain, one more time. Here, I will derive
the formulae for getting age from isochron measurements, and
without any reference to oldD or newD. Instead, I will use
a variable I call Q, which refers to the amount of "P" in
the rock at the time it was formed.
Variable Definition
-------------------
The calculations will involve certain numeric quantities, which
for convenience I label as follows.
Dx Measured D in sample x
Dix Measured Di in sample x
Px Measured P in sample x
Qx The amount of P that was in the sample at the
the time the rock solidified. (Unknown)
t The amount of time since the rock solidified. (Unknown)
H Measured property of element "P". (Half Life)
The two variables marked as "unknown" are quantities which cannot be
measured directly. They may, however, be calculated from other
quantities, as long as we know some relationships between them.
Closed world assumption
-----------------------
The closed world assumption is an assumption we make about the
rock. It is not proven; though we can get evidence that it holds
for a rock. This has been discussed previously; I will leave it
for now.
The point to note is that if the assumption is not true,
then the relations given below may not be true either.
The assumption is that the atoms in any sample are the
same atoms that were present in that sample when the rock
solidified. Some atoms may have decayed (changed from P into
D), but they do not move in and out of the sample.
Specifically:
The P in each sample is what is left from decay of the P that was
present at time of solidification, and nothing else.
The Di in each sample is the same as what was there when the rock
solidified.
The D in each sample is the same as what was there when the rock
solidified, plus what was added by decay of P since solidification.
Radioactive decay law
---------------------
This is a relationship which is based on knowledge of physics. It
is assumed to hold now, and in the past, over all the time since
the rock solidified, and in the same way for all samples. If this
assumption is not true, then the relations given may not be true
either. The assumption is tested in many ways; and the isochron
method builds upon that knowledge. It is not merely assumed by
faith.
The law can be expressed as a relation between Px, Qx, t and H;
this holds for any sample x.
Px = Qx * 2^(-t/H)
Cogenetic assumption
--------------------
This assumption is that when the rock solidified, all samples
originally had the same ratio of D to Di. This also follows
from basic physics as studied now, for a rock formed from a
single melt.
D and Di are isotopes of the same element, so they behave
the same way in chemical and other non-nuclear reactions. The
assumption might be violated if different parts of the rock
formed at different times, from a different mix of source
material
By the closed world assumption, the amount of "P" which has
decayed since solidification is (Q-P), and so the amount of
D originally in the sample at solidification is D-(Q-P)
I will write the equation "a/b = c/d" as "a*d = b*c". This
is perfectly valid, and avoids problems with divisions.
(Dx-Qx+Px)*Diy = (Dy-Qy+Py)*Dix for any two x and y samples
That is it. That is all there is to it.
Everything else is just basic maths.
The Age calculation
-------------------
Given any two samples x and y, we can calculate an age.
Px = Qx * 2^(-t/H)
Py = Qy * 2^(-t/H)
(Dx-Qx+Px)*Diy = (Dy-Qy+Py)*Dix
Three equations. Three unknowns. Solve.
Remember;
P, D and Di are just measured numbers; and we have those numbers
for two samples, x and y.
Q is not measured. It refers to the amount of "P" which was
in the rock at the time of solidification. We let Qx refer to
the amount of P originally in sample "x".
t is the time since solidification; the age of the rock.
H is the half life.
The first two equations given above are just the radioactive decay
law, applied to both samples.
The second two equations are roughly the same as oldD1/Di1 = oldD2/Di2;
except that I have chosen to use a different symbol "Q" to refer to a
different quantity -- the amount of "P" in the sample at the time of
solidification.
By the closed world assumption, the amount of "D" that was in the sample
at the time of solidification is the amount of "D" now, minus the
amount which has been added since solidification by decay of "P".
That is, if you wanted to refer to an original_D value, it would
be D - (Q-P). But note; I am not using any such variable.
Solving equations
-----------------
Let r = 2^(t/H)
Qx = r*Px
Qy = r*Py
Subtitute
(Dx-(r-1)*Px)*Diy = (Dy-(r-1)*Py)*Dix
Dx*Diy - (r-1)*Px*Diy = Dy*Dix - (r-1)*Py*Dix
(r-1)*Py*Dix - (r-1)*Px*Diy = Dy*Dix - Dx*Diy
(r-1)*(Py*Dix - Px*Diy) = Dy*Dix - Dx*Diy
r-1 = (Dy*Dix - Dx*Diy) / (Py*Dix - Px*Diy)
r = 1 + (Dy/Diy - Dx/Dix) / (Py/Diy - Px/Dix)
t = H*Log2(1 + (Dy/Diy - Dx/Dix) / (Py/Diy - Px/Dix))
A caution for those unfamiliar with this kind of analysis:
Variable definitions just establish a vocabulary. It is not
sensible to "interpret" t as "Age of the Earth". If we want
to consider the age of the earth, then we would introduce a
different symbol for that different quantity, and perhaps give
some mathematic relations involving it. I do not know of any
such relations.
The definitions used here reserve the symbol "t" as shorthand
for "The amount of time since the rock solidified".
The age of the rock, since solidification, is just
t = H*Log2(1 + (Dy/Diy - Dx/Dix) / (Py/Diy - Px/Dix))
If you accept the closed world assumption, and the cogentic
assumption, and the radioactive decay laws, then this age
since solidification follows by basic, simple, mathematics.
There is no place for reinterpretation; to disagree is just
to be bad at maths; nothing else.
Is the age defined?
-------------------
Generally, we take three or more samples. This means we can
calculate many ages, by using different pairs of samples. There
are small measurement errors when obtaining P, D and Di; I'll
ignore this for the time being.
If we get get different ages, by using different pairs,
then the starting assumptions must have been incorrect.
There is no other alternative. If the equations were correct,
then the ages would be the same for all pairs of samples.
This is a disproof of the starting assumptions (closed world
and cogenetic assumptions).
If all ages are the same, this is strong evidence that our
starting assumptions were correct, and that means we have
calculated the age of the rock since solidification.
What about the isochron?
------------------------
The isochron is just a way of plotting points to help explain
the calculations. The equations involve a quantity which
happens to be the slope on a certain plot of the data. This
also, is basic, trivial, maths.
Cheers -- Chris
From hostuart Wed May 15 16:59:43 2002
zoe_althrop
Rowley) wrote:
>From: muz...@aol.com (zoe_althrop)
><snip>
>
>> that's what I mean. If there were a lot of P at the time of
>> formation of Earth's original rocks, then at remelt, what P is
>> left is evenly distributed according to the type of mineral that
>> incorporates it.
>
>You´re assuming that all the contents of the rock stay together
>for all time, rock forms, rock melts, rock solidifies.
>That isn't nessesarily how it works, if the rock erodes, P is
>water soluble and D isn't then the D might collect in the river
>bends while the P is washed out to sea and ends up in limestone
>hundreds of miles away. If, 100 million years later both the
>river sediment and the limestone are melted and form new rocks
>then one rock will have "too much" P and the other "too much" D
I'm talking about those rocks that you consider to be closed. It
sounds as if you are saying that rocks cannot be reliably dated
because of the conditions that you mention above. How can we be
certain that such conditions do not apply to any and all rocks dated?
You're making a case for me, you know.
However, the particular rocks I'm interested in are those rock layers
that are associated with fossils -- layers of igneous rock that have
not eroded but remain as a layer between or over layers of
fossil-bearing sedimentary rock. Since fossil-bearing layers are
dated by the age of the igneous rock that intrudes or forms a layer
above the fossils, such layers had better not leach or be
contaminated, right?
>> Therefore, should samples today have high
>> quantities of P in relation to Di, as reflected by high
>> quantities of oldD/Di?
>
>Not nessesarily, the "extra" P could have ended up in other
>rocks so you would have to grind up the whole earth and measure
>P and D but even that wouldn't help,
nothing so extreme. I'm only interested in those layers of rock that
have not eroded and joined the rivers or oceans.
>there could have been "too much" D when the earth was originaly
>formed and the "extra" P could have ended up in the other planets.
>Or there could have been "too much" D when the solar system formed,
>the supernova/god that created the P may well have created some D
>as well so just comparing amounts of P and D does not tell you the
>age of anything unless you have some knowledge of the original
>amounts!
it's speculation at this point, right? And how do you determine that
supernovas create D from the start? Are you saying that D does not
necessarily have to be a result of parent decay? And if so, how so?
><snip>
>
>> I meant, the test consists of only two samples of each type of
>> mineral. Is two sufficient to get an idea of the quantity of P
>> that is characteristic of that particular type of mineral?
>
>Why should we care what "quantity of P that is characteristic of
>that particular type of mineral"?
we should care, if for no other reason than that this may be
information that comes in useful at some later date. Knowledge is
never wasted.
>It seems like you are trying to sneak your "accumulated age product"
>back into the equation?
sneak? I'm openly investigating such a possibility as accumulated age
product being mistaken for newD. For now, I'm looking at the actual
premises or assumptions upon which the isochron is based, not whether
oldD/Di can be determined if the starting assumption is correct.
>What we need for the age calculation are P, D and Di measurements
>for at least two (usualy at least 6 are used) samples (with
>differing P/Di ratios) and the knowledge that OldD/Di ratios
>are the same for all samples from the same melt.
>We don't need "characteristic" quantities of anything.
not for the isochron, maybe, but for general knowledge, yes. It could
be quite useful.
>> >The fact that P is taken up in different amounts when minerals
>> >solidify is _totally_ different from the fact that oldD/Di is
>> >the same for all samples.
>
>> I understand that. But if quantity of oldD is fairly high in
>> relation to Di, and we have not yet made it to even 0.116 percent
>> of the half life of Lutetium (if that's what Lu stands for), this
>> would indicate a high quantity of P, right? Either that, or more
>> time has passed for this mineral than even the age of the universe.
>
>Or some of the OldD was there from the begining.
is there any evidence that D can form outside of parent decay? If
not, then there is no basis for claiming that oldD was there from the
beginning, is there?
>> and Jon, using the figures above, would you tell me the original
>> P for xt-1 and 2? Present P/Di of 28.36 plus total D/Di of
>> 0.816178 = 29.166178,
>
>What does P/Di + TotalD/Di have to do with anything?
>You want P/Di + NewD/Di surely?
yes. I misspoke.
>You have to subtract OldD/Di
>29.166 - .282 = 28.88
>(no point in using more decimals then in the number with the
>least decimals)
>
>And for xt-2 NewD = .780 - .282 = .498
>26.36 - .489 = 25.87
>
>> which looks nowhere near what original P
>> looks like on this diagram.
>
>Zoe, the axis on this diagram are not drawn at the same scale,
>so the trajectories don't go at a 45 degree angle, they will
>be almost vertical.
that explains it. Okay.
>>From what I can see, the points don't
>> fit together, going from the X-axis to the Y-axis. See:
>
>The trajectories shouldn't be going from the X-axis!
>They go from the OldD/Di line!
I was going from the oldD/Di line, but couldn't see the end of the
trajectory. You've answered my question, though.
snip>
>The electrons have nothing to do with what element an atom is.
>The number of protons in the nucleus determines the element
>and also how many electrons the atom will normaly have,
>The number of neutrons in the nucleus determines which isotope
>you have, how stable or unstable it is.
>Protons and neutrons together determine the atomic weight,
>they weigh almost the same and the electron is much, much lighter.
>Neutrons are neutral (electricly uncharged), protons have a
>positive charge and electrons are negetive, that's why an
>uncharged atom has the same number of electrons as protons.
>An uncharged atom is what we normaly call an atom, if it has
>a charge, more or less electrons, it's called an ion but given
>a chance an ion will even out it's charge by altering the number
>of electrons. Ions are what make batteries work, they chemicaly
>remove electrons from atoms and force them to go though the
>electrical circuit to get back to the charged ions and
>neutralize them.
>
>The electrons do all the chemical stuff, reacting with other
>atoms and so forth.
>If you rearange the electrons without changing how many their
>are you change the energy stored in the atom, that's how
>lasers and flouresent tubes work, the electrons are electricly
>excited so they change to a more energetic state and when they
>change back again they give of light.
so I was correct, then, when I said that the electron is what made the
difference in the element, causing isotope formation?
>Anyways the electrons are irrelevant to isochron dating.
just information gathering, that's all.
>> >>>You can't predict what the ratio of oldD to Di should be from
>> >>>radioactive half-lives alone; you need to know something about
>> >>>the amount of Di now or previously
>
>> >>Di (now or previously) always remains the same, doesn't it?
>
>> >On average, yes. In detail no; it moves around in molten rock.
>> >It mostly stays the same in solid rock (and we are pretty good
>> >at detecting when it didn't).
>
>> I meant, the QUANTITY of Di always remains the same, doesn't it?
>> It never becomes more, it never becomes less?
>
>It would probaly be created in supernovas all the time,
>but here on earth the quantity should be pretty near constant
>since the earths formation.
what evidence do you have, Eric, that Di is probably created in
supernovas? Do you consider that Earth was once a supernova, and
therefore...? Even so, the Earth is supposedly done with being a
supernova now, so we can rely on the constancy of Di?
>So the total quantity on earth stays the same, that doesn't mean
>that is the same in all rocks.
but for a cogenetic source, which is what I'm talking about, the Di
stays the same, right?
snip>
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
sip>
>>>Nope. In a remelt, the P gets evenly distributed.
>>
>>that's what I mean. If there were a lot of P at the time of formation
>>of Earth's original rocks, then at remelt, what P is left is evenly
>>distributed
>
>Yes.
>
>>according to the type of mineral that incorporates it.
>
>I'm not sure what you mean here. In a remelt there _are_ no minerals.
>Minerals are formed at solidification. At solidification, the P is
>_unevenly_ distributed; some mineral take up lots of P, some don't.
but if a comparison were made among minerals of the same type, they
should all carry an equivalent amount of P?
>
>>Therefore, should samples today have high quantities of P in relation
>>to Di, as reflected by high quantities of oldD/Di?
>
>Some minerals take up lots of P, some don't. Some minerals take up
>lots of D, some don't. There's no significant relationship between
>the groups.
I was asking about a relationship WITHIN the groups.
snip>
>>>There are seven different samples (a fairly typical number ... small
>>>enough so the testing isn't too expensive and large enough to get good
>>>statistical significance). That number is sufficient.
>>
>>I meant, the test consists of only two samples of each type of
>>mineral. Is two sufficient to get an idea of the quantity of P that
>>is characteristic of that particular type of mineral?
>
>When we are analyzing the age, we don't care about the amount of P
>that is characteristic of that particular type of mineral. We care
>about how much P is in the mineral.
I'm not trying to analyze age since solidification, at the moment.
I'm just curious about OldD/Di and its behavior. For instance, if a
melt contains exactly 100 atoms of Di and 210 atoms of D, how do these
100 atoms of Di and 210 atoms of D form same ratios for each mineral?
Would there be some oldD left unattached? What happens to those atoms?
>Of course, we like to know how much P is characteristic of a
>particular mineral; it helps us select samples for age determination.
>But the set of samples that we can use to experimentally determine how
>much P is characteristic of a particular mineral is _all_ samples of
>that mineral that have _ever_ been gathered and analyzed and reported.
>Lots more than two.
>
>When we are analyzing how much P is characteristic of a particular
>mineral, we also look at it from a theoretical point of view,
>analyzing the crystal structure and its chemical properties.
how does a theoretical point of view tell you how much real-world P is
really absorbed by a particular solidifying crystal?
>>>The fact that P is taken up in different amounts when minerals
>>>solidify is _totally_ different from the fact that oldD/Di is the same
>>>for all samples.
>>
>>I understand that. But if quantity of oldD is fairly high in relation
>>to Di, and we have not yet made it to even 0.116 percent of the half
>>life of Lutetium (if that's what Lu stands for), this would indicate a
>>high quantity of P, right?
>
>Not necessarily. Nobody ever claimed that oldD is produced _only_ by
>radioactive decay; only that it is produced only by radioactive decay
>_on_ _the_ _Earth_.
and that is all we've got to work with, right? Anything more is pure
speculation?
>It is virtually certain that oldD is produced in
>supernovae just like most of the other elements, and was present when
>the Earth formed.
have supernovae been sampled for original D?
> (The conditions necessary to produce oldD by fusion
>are not possible on a solid or even molten planet)
>
>>Either that, or more time has passed for
>>this mineral than even the age of the universe.
>>
>>and Jon, using the figures above, would you tell me the original P for
>>xt-1 and 2? Present P/Di of 28.36 plus total D/Di of 0.816178 =
>>29.166178, which looks nowhere near what original P looks like on this
>>diagram.
>
>Of course. Adding totalD/Di to presentP/Di makes no sense, so of
>course the answer makes no sense.
no, I really meant newD/Di plus present P/Di. Also, I didn't notice
that the Y-axis was on a different scale than the X-axis.
>One atom of P decays to one atom of D. Therefore, newD/Di =
>decayedP/Di. OriginalP/Di is presentP/Di plus decayedP/Di. Therefore,
>originalP/Di can be calculated by:
>
> originalP/Di = presentP/Di + decayedP/Di
>
>But, since decayedP/Di = newD/Di
>
> originalP/Di = presentP/Di + newD/Di
>
>Or, since newD/Di = totalD/Di - oldD/Di
>
> originalP/Di = presentP/Di + totalD/Di - oldD/Di
>
>You added totalD/Di but didn't subtract oldD/Di
>
>So for sample xt-1
>
> originalP/Di = 28.35 + 0.816178 - 0.282515
>
> originalP/Di = 28.88366
>
>For sample xt-2:
>
> originalP/Di = 26.36 + 0.779948 - 0.282515
>
> originalP/Di = 26.85743
okay.
okay.
snip>
>>thanks for the explanation. So, does oldD/Di's quantity reflect either
>>the presence of a LOT of P at original formation of the rock,
>
>No, because oldD/Di has nothing to do with the amount of P at rock
>formation; only newD/Di has something to do with the amount of P at
>rock formation.
I meant that if there is a lot of oldD/Di in the rock, is this some
kind of indication of quantity of pre-melt P? And if so, can quantity
of OldD be related to quantity of pre-melt P, especially if the rock
gives no sign of having remelted?
>>or the
>>passing of time much older than 15 billion years?
>
>That's one possibility.
>
>The other possibility is that there is another source of oldD, such as
>supernovae.
what evidence do you have that supernovae produce D outside of decay
from P?
>
>The other possibility is true; supernovae produce oldD, just like they
>produce most of the elements that make up the Earth.
how does one go about measuring the products of a supernova?
>>>Since P has very different chemical properties than D or Di, when
>>>minerals solidify they will take up amounts of P that has nothing
>>>particular to do with the amount of D and Di incorporated.
>>
>>is quantity of oldD a reflection of quantity of originalP that has
>>decayed up to point of remelt?
>
>The amount of oldD is partly a reflection of the amount of P that
>decayed before solidification and partly a reflection of how much D
>was created by other processes, such as fusion in supernovae.
that supernova again. How do you measure for D created in a supernova
fusion?
>
>>>A mineral
>>>may take up very little D and Di and P, it may take up very little D
>>>and Di but lots of P, it may take up lots of D and Di but very little
>>>P, or it may take up lots of D and Di and P.
>>
>>but relatively speaking, if there were a lot of P at the beginning,
>>the quantities taken up characteristically by the various types of
>>minerals would be large compared to if we were down at the
>>near-exhaustion point of P, in time, right?
>
>The _quantities_, yes, but the _proportions_ (or ratios) between
>different minerals would be about the same.
I'm not comparing proportions between minerals. A single mineral will
do, if there are sufficient samples to detect a trend.
>>>>>And ... it's easier to measure
>>>>>accurately if there's (relatively) a lot of P and D in the sample, so
>>>>>scientists tend to select minerals for sampling that they expect to
>>>>>have a lot of P (and consequently a lot of totalD).
>>>>
>>>>is a sampling of two sufficient to establish a pattern?
>>>
>>>Where do you get your "sampling of two"?
>>
>>two of each type -- though I don't know why there is only one of
>>biotite (is that what "biot" stands for?). Wouldn't you need more
>>than two samples of a type in order to determine the characteristic
>>amount of P they would tend to absorb?
>
>Yes. So what? The point of the study was not to determine the
>characteristic amount of P that the minerals absorb. The point of the
>study was to measure the age of the rock, and the age of the rock does
>not depend on how much P the various minerals characteristically
>absorb.
I'm just wondering about premelt conditions.
snip more good stuff>
>>no, I meant, which number should be larger in the ratio if 38.7
>>billion years have passed, the D or the Di, at the time of
>>measurement?
>
>It depends on how much D was created by other processes, such as
>supernovae, and the half-life of P, and how much P was created by
>supernovae.
so is there evidence that new quantities of D and P are being formed
out there in the universe? What is this evidence?
But even so, there is no new input here on Earth, right? And that is
the only place we're concerned with right now.
>If the half-life of P is large compared to the age of the
>universe, then most of the oldD came from supernovae (or whatever) and
>the average oldD/Di ratio doesn't change much over time. If the
>half-life of P is much less than the age of the universe, and there
>was a lot of P created by supernovae or whatever, then most of the
>oldD came from decay of P and the average oldD/Di ratio is increasing
>faster over time than in the first case.
is there a way to turn the "ifs" into "whens"?
snip>
----
zoe
zoe_althrop
<host...@sky.fit.qut.edu.au> wrote:
snip>
then I must be "no-one," then. I do care about oldD, and also P
before solidification. If there is sufficient oldD in relation to
premelt P to indicate an age much older than the age of the earth,
then this may mean that something is wrong with our understanding of P
to D decay, which could also lead to questions about the formation of
rock in same ratios of D to Di.
>What we care about is the age of the rock. The variables oldD
>and newD are really only introduced for the sake of trying to
>explain how the methods work.
>
>Here is another statement of the case.
>
>: > so having admitted that we do NOT know newD, how do you arrive at the
>: > figures for newD?
>:
>: By calculating it.
then don't say you don't know newD. If you can calculate it, then you
do indeed know newD.
you have mentioned two starting assumptions that, if not accurate,
would lead to inaccurate dating. But there are two more assumptions
that, if inaccurate, could still possibly produce an isochron, but an
incorrect isochron, and those assumptions are that there has been no
partial remelting of crystals, and that ALL oldD has been successfully
tied to Di, with none left over.
Should these assumptions be investigated?
----
zoe
Jon Fleming
wrote:
>On Sun, 11 Aug 2002 13:21:00 +0000 (UTC), e...@FAKEADRESS.com (Eric
>Rowley) wrote:
>
>>From: muz...@aol.com (zoe_althrop)
>
>><snip>
>>
>>> that's what I mean. If there were a lot of P at the time of
>>> formation of Earth's original rocks, then at remelt, what P is
>>> left is evenly distributed according to the type of mineral that
>>> incorporates it.
>>
>>You´re assuming that all the contents of the rock stay together
>>for all time, rock forms, rock melts, rock solidifies.
>>That isn't nessesarily how it works, if the rock erodes, P is
>>water soluble and D isn't then the D might collect in the river
>>bends while the P is washed out to sea and ends up in limestone
>>hundreds of miles away. If, 100 million years later both the
>>river sediment and the limestone are melted and form new rocks
>>then one rock will have "too much" P and the other "too much" D
>
>I'm talking about those rocks that you consider to be closed. It
>sounds as if you are saying that rocks cannot be reliably dated
>because of the conditions that you mention above. How can we be
>certain that such conditions do not apply to any and all rocks dated?
By examining and selecting the samples. Rocks that have experienced
leaching are usually identifiable.
If we make a mistake, and the samples we have picked have experienced
leaching, then the plotted points will almost certainly not lie on one
straight line.
So there are two safeguards against the problem.
All in all, there may be one or two results in the literature that are
wrong because the investigators didn't detect leaching and the
leaching just happened to remove just the right amount of one of the
elements from the samples (fantastically improbable) and the points
wound up plotted on a straight line. But most published results are
for closed systems.
>it's speculation at this point, right? And how do you determine that
>supernovas create D from the start?
The current scientific theory, well investigated and supported, is
that _all_ elements heavier than iron are created in supernovae, and
in a few other rare conditions. See
<http://www.aoc.nrao.edu/~kdyer/supernova.html> and
<http://imagine.gsfc.nasa.gov/docs/ask_astro/answers/010125a.html>.
> Are you saying that D does not
>necessarily have to be a result of parent decay?
Yes, but ...
D is only created by parent decay _on_ _the_ _Earth_
>And if so, how so?
Nuclear fusion in conditions of temperature and pressure so high that
people and planets and solids and liquids cannot exist.
<snip>
..
>>We don't need "characteristic" quantities of anything.
>
>not for the isochron, maybe, but for general knowledge, yes. It could
>be quite useful.
Agreed.
>>> >The fact that P is taken up in different amounts when minerals
>>> >solidify is _totally_ different from the fact that oldD/Di is
>>> >the same for all samples.
>>
>>> I understand that. But if quantity of oldD is fairly high in
>>> relation to Di, and we have not yet made it to even 0.116 percent
>>> of the half life of Lutetium (if that's what Lu stands for), this
>>> would indicate a high quantity of P, right? Either that, or more
>>> time has passed for this mineral than even the age of the universe.
>>
>>Or some of the OldD was there from the begining.
>
>is there any evidence that D can form outside of parent decay?
Yes. There is lots of evidence that D can be and is formed in
supernovae. See <http://zebu.uoregon.edu/textbook/energygen.html> and
<http://helios.gsfc.nasa.gov/nucleo.html> for starters. The key word
is "nucleosynthesis".
<snip>
No, you were 100% wrong. As he said, "The electrons have nothing to
do with what element an atom is." He went on to say that "The number
of protons in the nucleus determines the element and also how many
electrons the atom will normally have. The number of neutrons in the
nucleus determines which isotope you have,"
The electron is _not_ what made the difference in the element.
<snip>
>what evidence do you have, Eric, that Di is probably created in
>supernovas?
There's lots. That's a whole different subject in itself.
>Do you consider that Earth was once a supernova, and
>therefore...?
No. Supernovae are exploding really big stars (see
<http://www.rog.nmm.ac.uk/leaflets/supernovae/supernovae.html> and
<http://heasarc.gsfc.nasa.gov/docs/snr.html> and
<http://rsd-www.nrl.navy.mil/7212/montes/sne.html>. (Novae are
exploding medium-size stars). The Earth is made mostly from elements
that were formed in supernovae, then became part of a galactic gas,
then condensed. It's not reasonable to say that "the Earth was once a
supernova ...", that is saying that the Earth once exploded (and the
entire solar system isn' enough mass to make a supernova).
>Even so, the Earth is supposedly done with being a
>supernova now, so we can rely on the constancy of Di?
(ROFL) Zoe, the entire Earth would vaporize long before it got hot
enough to create D or P or Di. You need a tremendous temperature,
_and_ you need a tremendous mass (the entire solar system isn't
enough) to get enough gravity to force the nuclei together at those
temperatures (by gravity, mostly; in supernovae the shock wave of the
explosion plays a part too).
If D is formed on Earth by something other than radioactive decay, or
if P or D are created on Earth, we won't be around to see it.
<snip>
Chris Ho-Stuart
> On Mon, 12 Aug 2002 01:53:40 +0000 (UTC), Chris Ho-Stuart
> <host...@sky.fit.qut.edu.au> wrote:
>>Let me emphasize that. No-one really cares about oldD and newD.
>
> then I must be "no-one," then. I do care about oldD, and also P
> before solidification. If there is sufficient oldD in relation to
> premelt P to indicate an age much older than the age of the earth,
> then this may mean that something is wrong with our understanding of P
> to D decay, which could also lead to questions about the formation of
> rock in same ratios of D to Di.
The relative amounts of oldD and P do not have any clear
association with the age of the Earth. You keep saying this,
but it is wrong.
>>What we care about is the age of the rock. The variables oldD
>>and newD are really only introduced for the sake of trying to
>>explain how the methods work.
>>
>>Here is another statement of the case.
>>
>>: > so having admitted that we do NOT know newD, how do you arrive at the
>>: > figures for newD?
>>:
>>: By calculating it.
>
> then don't say you don't know newD. If you can calculate it, then you
> do indeed know newD.
Shrug. If you refuse to read for comprehension, that is your
look out. The distinction between values measured and values
inferred is obvious. The calculation relies on certain basic
assumptions
-- closed world for the rock since solidification
-- one source (rock does not incorporate pieces solidified
at very different times.
-- physics works (constant decay rates)
These assumptions can be violated, in some cases (the first two,
anyway). The isochron also gives ways to detect such violations.
[snip algebra from my previous post, retain statement of assumptions]
[snip some more stuff]
> you have mentioned two starting assumptions that, if not accurate,
> would lead to inaccurate dating. But there are two more assumptions
> that, if inaccurate, could still possibly produce an isochron, but an
> incorrect isochron, and those assumptions are that there has been no
> partial remelting of crystals, and that ALL oldD has been successfully
> tied to Di, with none left over.
>
> Should these assumptions be investigated?
No. Your assumptions are either already covered in the assumptions
I have given, or else they are irrelevant.
The assumptions I give above are all you need. Really. This is
also basic, like 1 + 1 = 2. Read them carefully. I have stated
them very precisely.
Here it is a very concise restatement:
All samples have the same atoms they had at solidification
time t. All samples had the same D/Di ratio at time t.
Cheers -- Chris
Jon Fleming
wrote:
>>
>>Let me emphasize that. No-one really cares about oldD and newD.
>
>then I must be "no-one," then. I do care about oldD, and also P
>before solidification. If there is sufficient oldD in relation to
>premelt P to indicate an age much older than the age of the earth,
>then this may mean that something is wrong with our understanding of P
>to D decay, which could also lead to questions about the formation of
>rock in same ratios of D to Di.
I don't agree with Chris. What he should have said is "No-one who is
measuring the age of rocks with the isochron method cares about oldD
and newD".
There are geologists who investigate oldD and any consequences
thereof.
However, the amount of oldD in relation to P does not indicate the age
of the Earth, since when the Earth was formed its constituents
included D that was not created by radioactive decay of P. You are
looking down a blind alley.
Jon Fleming
wrote:
>On Sun, 11 Aug 2002 15:06:50 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>sip>
>
>>>>Nope. In a remelt, the P gets evenly distributed.
>>>
>>>that's what I mean. If there were a lot of P at the time of formation
>>>of Earth's original rocks, then at remelt, what P is left is evenly
>>>distributed
>>
>>Yes.
>>
>>>according to the type of mineral that incorporates it.
>>
>>I'm not sure what you mean here. In a remelt there _are_ no minerals.
>>Minerals are formed at solidification. At solidification, the P is
>>_unevenly_ distributed; some mineral take up lots of P, some don't.
>
>but if a comparison were made among minerals of the same type, they
>should all carry an equivalent amount of P?
Probably not. The amount of P taken up in the mineral depends on the
chemistry and crystal structure of the mineral _and_ the amount of P
in the melt. The amount of P in the melt will vary from one melt to
another.
<snip>
>>When we are analyzing the age, we don't care about the amount of P
>>that is characteristic of that particular type of mineral. We care
>>about how much P is in the mineral.
>
>I'm not trying to analyze age since solidification, at the moment.
>I'm just curious about OldD/Di and its behavior. For instance, if a
>melt contains exactly 100 atoms of Di and 210 atoms of D, how do these
>100 atoms of Di and 210 atoms of D form same ratios for each mineral?
It's difficult to explain, or illustrate with an example, what happens
step-by-step as a mineral solidifies without getting into some pretty
complex probability concepts that you need years of work to comprehend
(and I'd have to study up some before doing the numbers).
However, the step-by-step scenario doesn't matter. No matter how D and
Di are incorporated into the each mineral that solidifies as the melt
cools, they will be incorporated in that mineral in the same
proportion that they are in the melt. This follows immediately from
the fact that the solidification process cannot distinguish between D
and Di.
When an atom moves from the melt to a solid mineral, it is either an
atom of that mineral's normal constituents or it is an atom of D or it
is an atom of Di. We don't care about the case where it is an atom of
the mineral's normal constituent. In the case where it is an atom of
D or an atom of Di, the probability that it is an atom of D is the
same as the number of atoms of D divided by the total number of atoms
of D and Di, and similarly for Di. (In your example, the probability
that it is an atom of D is 210/(100+210) = 210/310 = 67.74% and the
probability that it is an atom of Di is100/310 = 32.36%).
As the mineral solidifies (incorporating trillions of trillions of
trillions of atoms in a real mineral) this "coin flip" of selecting an
atom of D or Di randomly is repeated trillions of trillions of
trillions of times. The ratio of D to Di in the mineral will wind up
in exactly (or indistinguishably close to) the same proportion as the
ratio was in the melt, just as if you toss a fair coin trillions of
trillions of trillions of times you will wind up with 50% heads and
50% tails.
So what happens to the melt? As long as atoms of D and Di are removed
from the melt (into minerals) in the same proportion as they are in
the melt, the ratio of D to Di in the melt will not change. A few
examples:
Suppose we have 210 atoms of D and 100 atoms of Di, or a ratio of 2.1.
Say 21 atoms of D go into a mineral, and therefore 10 atoms of Di also
go into that mineral (so the proportion of D to Di in the mineral is
21/10 or 2.1). There are 210 - 21 = 189 atoms of D left in the melt
and 100 - 10 = 90 atoms of Di left in the melt. The D/Di ratio in the
melt is 189/90 = 2.1, the same as before.
Or suppose instead that 42 atoms of D go into the mineral, and
therefore 20 atoms of Di go into the mineral. The D/Di ratio of the
mineral is 2.1. There are 168 atoms of D left in the melt, and 80
atoms of Di. 168/80 = 2.1. There are 42 atoms of D in the mineral
and 20 atoms of Di in the mineral. 42/20 = 2.1.
It's easy to prove with simple algebra that this happens no matter
_how_ _many_ atoms of D and Di go into each mineral, as long as they
go into the minerals with probabilities that depend on their
proportion in the melt. Since solidification cannot distinguish
between D and Di, they go into the minerals with probabilities that
depend on their proportion in the melt, and the ratio of D to Di is
the same in the melt and all minerals solidified from that melt.
>Would there be some oldD left unattached?
There might be oldD left that is not incorporated into other minerals.
It depends on how easily D is taken up by the last few minerals that
solidify.
>What happens to those atoms?
They solidify into solid D, tiny grains of D and Di intermixed in the
ratio they were in the melt and in all other minerals that solidified
from that melt, nestled in between the other minerals.
>>Of course, we like to know how much P is characteristic of a
>>particular mineral; it helps us select samples for age determination.
>>But the set of samples that we can use to experimentally determine how
>>much P is characteristic of a particular mineral is _all_ samples of
>>that mineral that have _ever_ been gathered and analyzed and reported.
>>Lots more than two.
>>
>>When we are analyzing how much P is characteristic of a particular
>>mineral, we also look at it from a theoretical point of view,
>>analyzing the crystal structure and its chemical properties.
>
>how does a theoretical point of view tell you how much real-world P is
>really absorbed by a particular solidifying crystal?
We have mathematical models of crystal formation and chemistry that
predict very well how this happens.
>>>>The fact that P is taken up in different amounts when minerals
>>>>solidify is _totally_ different from the fact that oldD/Di is the same
>>>>for all samples.
>>>
>>>I understand that. But if quantity of oldD is fairly high in relation
>>>to Di, and we have not yet made it to even 0.116 percent of the half
>>>life of Lutetium (if that's what Lu stands for), this would indicate a
>>>high quantity of P, right?
>>
>>Not necessarily. Nobody ever claimed that oldD is produced _only_ by
>>radioactive decay; only that it is produced only by radioactive decay
>>_on_ _the_ _Earth_.
>
>and that is all we've got to work with, right? Anything more is pure
>speculation?
Well, it's not speculation that the only minerals that we are going to
use to measure the age of the Earth or things on the Earth are
minerals on the Earth (or other solid bodies in the Solar System).
Therefore it's not speculation to say the that any newD that we
calculate (or measure, in the generic method) in such minerals is
produced by decay of P
>>It is virtually certain that oldD is produced in
>>supernovae just like most of the other elements, and was present when
>>the Earth formed.
>
>have supernovae been sampled for original D?
Yes. Each element gives off a distinctive pattern of light when it is
hot and absorbs a distinctive pattern of light when it is cold. By
analyzing the content of light from supernovae, we can detect what
elements are there. References below.
(Interesting side factoid: helium was first discovered when
scientists used this method on the Sun, and was only discovered on
Earth afterwards. The name "helium" comes from "helios", the Greek
for "Sun". The method works. See
<http://antwrp.gsfc.nasa.gov/apod/ap010120.html>.)
..<snip>
>>>thanks for the explanation. So, does oldD/Di's quantity reflect either
>>>the presence of a LOT of P at original formation of the rock,
>>
>>No, because oldD/Di has nothing to do with the amount of P at rock
>>formation; only newD/Di has something to do with the amount of P at
>>rock formation.
>
>I meant that if there is a lot of oldD/Di in the rock, is this some
>kind of indication of quantity of pre-melt P?
It is some kind of indication, but not a very useful one. It's too
"fuzzy" because there are several other factors that affect the amount
of oldD/Di in the rock.
> And if so, can quantity
>of OldD be related to quantity of pre-melt P, especially if the rock
>gives no sign of having remelted?
Not really.
>>>or the
>>>passing of time much older than 15 billion years?
>>
>>That's one possibility.
>>
>>The other possibility is that there is another source of oldD, such as
>>supernovae.
>
>what evidence do you have that supernovae produce D outside of decay
>from P?
Detection of D in supernovae.
Theoretical physics, which predicts that D can only be produced by
means other than radioactive decay under conditions such as found in
supernovae, and further predicts that D _must_ be produced in such
conditions.
>>The other possibility is true; supernovae produce oldD, just like they
>>produce most of the elements that make up the Earth.
>
>how does one go about measuring the products of a supernova?
By analyzing the light (including many wavelengths that we don't see
directly). See
<http://imagine.gsfc.nasa.gov/docs/teachers/lessons/supernova/supernova_chemistry.html>,
<http://cosmology.berkeley.edu/Education/DEMOS/Desk_Top_Stars/LS.html>,
and
<http://imagine.gsfc.nasa.gov/docs/teachers/lessons/xray_spectra/spectra_cover.html>.
>>>>Since P has very different chemical properties than D or Di, when
>>>>minerals solidify they will take up amounts of P that has nothing
>>>>particular to do with the amount of D and Di incorporated.
>>>
>>>is quantity of oldD a reflection of quantity of originalP that has
>>>decayed up to point of remelt?
>>
>>The amount of oldD is partly a reflection of the amount of P that
>>decayed before solidification and partly a reflection of how much D
>>was created by other processes, such as fusion in supernovae.
>
>that supernova again. How do you measure for D created in a supernova
>fusion?
No problem.
<snip>
>>If the half-life of P is large compared to the age of the
>>universe, then most of the oldD came from supernovae (or whatever) and
>>the average oldD/Di ratio doesn't change much over time. If the
>>half-life of P is much less than the age of the universe, and there
>>was a lot of P created by supernovae or whatever, then most of the
>>oldD came from decay of P and the average oldD/Di ratio is increasing
>>faster over time than in the first case.
>
>is there a way to turn the "ifs" into "whens"?
Yes. Chose a particular P-D-Di system and do some research.
Jon Fleming
wrote:
>you have mentioned two starting assumptions that, if not accurate,
>would lead to inaccurate dating. But there are two more assumptions
>that, if inaccurate, could still possibly produce an isochron, but an
>incorrect isochron, and those assumptions are that there has been no
>partial remelting of crystals, and that ALL oldD has been successfully
>tied to Di, with none left over.
>
>Should these assumptions be investigated?
There's not much reason to, because they have been investigated
already. Over and over again, from all sorts of different points of
view and with all sorts of hypotheses. (You are falling into the old
creationist pattern of _assuming_ that scientists don't question and
re-question and re-re-question their methods and results. They do.)
However, ther are still people out there that are investigating these
assumptions, because scientists _never_ stop questioning.
Partial remelting is easily detected when selecting samples. If a
sample was partially remelted and we miss that fact, it is virtually
certain that the partial remelt will destroy the straight-line
relationship on the plot. That is, there are two levels of
safeguards, each with a very high probability of detecting the
problem. The overall probability of missing the problem is very low.
There may be one or two published results where samples were partly
remelted _and_ were partially remelted in just the right way to
maintain the straght-line relationship _and_ the investigator did not
detect the partial remelt when selecting the samples, but it is
certain that the vast majority of published results are not from
partially remelted samples.
The "tying" of oldD to Di has been thoroughly investigated,
theoretically and experimentally. The hypothesis that they are not
"tied" is a dry hole. If you want to learn enough nuclear physics and
chemistry to propose a new line of inquiry, go for it.
Derek Stevenson
news:8ahklusid1n9hiab6...@4ax.com...
> On Tue, 13 Aug 2002 23:55:14 +0000 (UTC), muz...@aol.com (zoe_althrop)
> wrote:
> >On Sun, 11 Aug 2002 15:06:50 +0000 (UTC), Jon Fleming
> ><jo...@fleming-nospam.com> wrote:
> >>I'm not sure what you mean here. In a remelt there _are_ no minerals.
> >>Minerals are formed at solidification. At solidification, the P is
> >>_unevenly_ distributed; some mineral take up lots of P, some don't.
> >
> >but if a comparison were made among minerals of the same type, they
> >should all carry an equivalent amount of P?
>
> Probably not. The amount of P taken up in the mineral depends on the
> chemistry and crystal structure of the mineral _and_ the amount of P
> in the melt. The amount of P in the melt will vary from one melt to
> another.
I'm not sure you understood Zoe's question. I suspect that by "minerals of
the same type", she actually means "the same mineral", and by "an
equivalent amount of P", she actually means "the same proportion of P". In
other words, she's asking whether all crystals of, say, shazamite should
contain the same ratio of P to all other elements.
And the answer to that, I think, should be "yes". By definition. All
crystals of shazamite should contain one atom of P, two atoms of carbon,
three of silicon and one of oxygen, because that's what *makes* it
shazamite. If its composition was two atoms of P, three of carbon, no
silicon and two oxygen, it wouldn't be shazamite, it would be sivanite.
Unless I'm as confused as Zoe here ...
[snip]
> However, the step-by-step scenario doesn't matter. No matter how D and
> Di are incorporated into the each mineral that solidifies as the melt
> cools, they will be incorporated in that mineral in the same
> proportion that they are in the melt. This follows immediately from
> the fact that the solidification process cannot distinguish between D
> and Di.
>
> When an atom moves from the melt to a solid mineral, it is either an
> atom of that mineral's normal constituents or it is an atom of D or it
> is an atom of Di. We don't care about the case where it is an atom of
> the mineral's normal constituent.
Okay, now *I'm* confused. Why this distinction between D, Di and "normal
constituents"? Aren't D and Di "normal constituents" of the minerals
they're incorporated into?
The best way I can describe the way I understand minerals form in a cooling
melt is with a (somewhat bizarre) sports analogy. Let's say I'm in charge
of creating a sports team to compete on my country's behalf in the
Olympics. It hasn't yet been decided *which* sport my team is going to
compete in -- that will be decided by the collective skills of the people I
manage to recruit for the team. (Well, I said it was a bizarre analogy!)
So, I have a large pool of kickers, catchers, throwers, runners, hitters
and swimmers to choose from. All of their names are thrown into a hat, and
I have to choose nine of them, without knowing who they are.
If I pull out the names of a catcher, a thrower, four hitters, three
runners, no kickers and no swimmers, I can field a baseball team. If I get
two catchers, two throwers and five swimmers, and no kickers, runners or
hitters, I have a water polo team. What I actually end up with is five
kickers and four runners, which means that I have a football (American or
Canadian football) team.
As it happens, there are two kinds of kickers -- kickers who learned their
skills playing football, and kickers who learned their skills playing
soccer. As I recruit my team, I'm not aware of this distinction: the
information doesn't appears on the players' resumes, and it makes no
difference to me anyway. Since the ratio of football-kickers to
soccer-kickers in the talent pool is 4:1, the odds are good that the same
ratio will apply to the five kickers on my football team.
But in what way are the kickers not "normal constituents" of a football
team?
Now, just to clarify my own understanding, let's continue the analogy.
Runners who play football, it turns out, eventually strengthen their legs
to the point where they become kickers. The amount of time this takes
varies depending on the individual, but if you take a group of four runners
and have them play football, a year later you'll have two runners and two
kickers. So if my football team plays together for a year, at the end of
that time I'll have gone from five kickers and four runners to seven
kickers and two runners.
A journalist writing a profile of the team wants to figure out how long
we've been together. Since we went in for football, he knows that we must
have started out with five kickers and four runners. He knows that we know
have seven kickers and two runners. He knows that of the seven kickers, six
are football-kickers and one is a soccer-kicker (because football kickers
always get a tattoo of a football on their arms, while soccer-kickers
always get one of a soccer ball). He knows that in one year, two out of
every four runners who play football become football-kickers. Therefore,
since I had to have started out with five kickers, two of my seven kickers
must be former runners and that the team has been together a year.
Am I on the right track here? And Zoe, is this making any sense to you? (If
you can't figure out how the elements of this analogy correspond to
isochron dating, just ask. Note that this analogy applies to figuring out
the history of a *single sample*, not to the multiple samples required to
plot an isochron.)
[snip]
Derek Stevenson
news:3d599674...@news-server.cfl.rr.com...
> On Sun, 11 Aug 2002 13:21:00 +0000 (UTC), e...@FAKEADRESS.com (Eric
> Rowley) wrote:
[snip]
> >there could have been "too much" D when the earth was originaly
> >formed and the "extra" P could have ended up in the other planets.
> >Or there could have been "too much" D when the solar system formed,
> >the supernova/god that created the P may well have created some D
> >as well so just comparing amounts of P and D does not tell you the
> >age of anything unless you have some knowledge of the original
> >amounts!
>
> it's speculation at this point, right? And how do you determine that
> supernovas create D from the start? Are you saying that D does not
> necessarily have to be a result of parent decay? And if so, how so?
[snip]
> >It seems like you are trying to sneak your "accumulated age product"
> >back into the equation?
>
> sneak? I'm openly investigating such a possibility as accumulated age
> product being mistaken for newD. For now, I'm looking at the actual
> premises or assumptions upon which the isochron is based, not whether
> oldD/Di can be determined if the starting assumption is correct.
[snip]
> >Or some of the OldD was there from the begining.
>
> is there any evidence that D can form outside of parent decay? If
> not, then there is no basis for claiming that oldD was there from the
> beginning, is there?
[snip]
> so I was correct, then, when I said that the electron is what made the
> difference in the element, causing isotope formation?
[snip]
> >It would probaly be created in supernovas all the time,
> >but here on earth the quantity should be pretty near constant
> >since the earths formation.
>
> what evidence do you have, Eric, that Di is probably created in
> supernovas? Do you consider that Earth was once a supernova, and
> therefore...? Even so, the Earth is supposedly done with being a
> supernova now, so we can rely on the constancy of Di?
[snip]
In _The Restaurant at the End of the Universe_, Douglas Adams has this to
say about humans' lack of awareness that the Earth was actually a
supercomputer built to determine the Ultimate Question to the Answer of
Life, the Universe and Everything:
"[T]his is very odd, because without that fairly simple and obvious piece
of knowledge, nothing that ever happened on the Earth could possibly make
the slightest bit of sense."
Over the last couple of days, it's become clear that Zoe is lacking quite a
few pieces of knowledge which, unfortunately, are neither simple nor
obvious. Things like where elements come from, what isotopes are, what
distinguishes radiogenic from non-radiogenic isotopes, the basics of
mineral crystal formation, the basics of radioactive decay, and so on.
Lacking those pieces of knowledge, it's not surprising that, to her,
isochronology can't possibly make the slightest bit of sense.
And, without that (in hindsight) fairly simple and obvious piece of
knowledge (i.e., that Zoe's missing a huge amount of background), nothing
that's happened in this interminable discussion has made the slightest bit
of sense to any of the participants. (Well, many of the explanations
offered have made sense, but that's only added to the general confusion
because they've been so clear that her failure to understand them has been
completely baffling.)
There have been moments along the way when I wondered just how much Zoe
actually understood, and whether it wouldn't be more productive to stop the
discussion completely and go back to the very, *very* basics -- mineral
formation, radioactive decay and the like -- and then very gradually bring
her up to speed. I can't have been the only one to think that, can I? So
why didn't any of us ever say that?
This is your fault too, Zoe. As much as you might claim to be a simple,
unsophisticated country girl, you have a degree of arrogance and
self-satisfaction that's sadly misplaced in this sort of discussion. You
seem to assume right off the bat that whatever you *think* you know about a
subject is:
a) accurate and
b) so detailed as to constitute a sufficient understanding.
This is rarely the case, and it almost always requires a monumental
struggle to get you to see it. You're like the would-be flying sheep in the
Monty Python episode, of whom the farmer says, "Once they gets an oidear in
their 'eads, there's no shiftin' it."
You should rely a little less on your own sense of satisfaction with your
comprehension of an idea (I've never yet seen you announce your
satisfaction with an idea that was *right*), and put a little more trust in
the consensus of the group as to whether you've understood. Ask more
questions, even if they seem very simple and basic. *Especially* if they
seem very simple and basic, because this is where you tend to go wrong.
Above all, don't -- as I keep trying to remind you -- assume that just
because an idea makes you feel good, that it's right. Or vice versa.
Jon Fleming
<derekste...@yahoo.com> wrote:
>"Jon Fleming" <jo...@fleming-nospam.com> wrote in message
>news:8ahklusid1n9hiab6...@4ax.com...
>> On Tue, 13 Aug 2002 23:55:14 +0000 (UTC), muz...@aol.com (zoe_althrop)
>> wrote:
>> >On Sun, 11 Aug 2002 15:06:50 +0000 (UTC), Jon Fleming
>> ><jo...@fleming-nospam.com> wrote:
>
>> >>I'm not sure what you mean here. In a remelt there _are_ no minerals.
>> >>Minerals are formed at solidification. At solidification, the P is
>> >>_unevenly_ distributed; some mineral take up lots of P, some don't.
>> >
>> >but if a comparison were made among minerals of the same type, they
>> >should all carry an equivalent amount of P?
>>
>> Probably not. The amount of P taken up in the mineral depends on the
>> chemistry and crystal structure of the mineral _and_ the amount of P
>> in the melt. The amount of P in the melt will vary from one melt to
>> another.
>
>I'm not sure you understood Zoe's question.
That's a distinct possibility.
>I suspect that by "minerals of
>the same type", she actually means "the same mineral", and by "an
>equivalent amount of P", she actually means "the same proportion of P". In
>other words, she's asking whether all crystals of, say, shazamite should
>contain the same ratio of P to all other elements.
Ah. That certainly is a valid question, whether or not it's the
question she meant to ask.
The answer is "maybe". The atoms in which we are interested may or
may not be a part of the chemical formula of the mineral. It is
certainly reasonably common for them not to be in the chemical
formula. They are often incorporated into the mineral by substituting
for a sufficiently similar atom that _is_ in the chemical formula, or
by physical entrapment as the mineral crystal grows around them.
Without stating so, I was addressing the last two possibilities
(perhaps because I work with semiconductors, and they have substitute
atoms.)
If the atoms in which we are interested _are_ part of the chemical
formula of the mineral then, of course, then all samples of that
mineral will contain the same proportion of, for example, P.
I just don't know enough to say anything about the relative
frequencies of incorporation by being part of the chemical formula and
incorporation by other means. I know that the latter happens; e.g.
zirconia (ZrSiO4, see <http://webmineral.com/data/Zircon.shtml>) which
is widely used for U-Pb concordia-discordia dating (because
concordia-discordia dating requires zero or essentially zero oldD, and
zircons strongly reject lead but accept uranium at solidification and
hold lead created in the solid by decay of uranium pretty well).
>And the answer to that, I think, should be "yes". By definition. All
>crystals of shazamite should contain one atom of P, two atoms of carbon,
>three of silicon and one of oxygen, because that's what *makes* it
>shazamite. If its composition was two atoms of P, three of carbon, no
>silicon and two oxygen, it wouldn't be shazamite, it would be sivanite.
>
>Unless I'm as confused as Zoe here ...
You're discounting the significant possibilities I discussed above.
>> However, the step-by-step scenario doesn't matter. No matter how D and
>> Di are incorporated into the each mineral that solidifies as the melt
>> cools, they will be incorporated in that mineral in the same
>> proportion that they are in the melt. This follows immediately from
>> the fact that the solidification process cannot distinguish between D
>> and Di.
>>
>> When an atom moves from the melt to a solid mineral, it is either an
>> atom of that mineral's normal constituents or it is an atom of D or it
>> is an atom of Di. We don't care about the case where it is an atom of
>> the mineral's normal constituent.
>
>Okay, now *I'm* confused. Why this distinction between D, Di and "normal
>constituents"? Aren't D and Di "normal constituents" of the minerals
>they're incorporated into?
Maybe, maybe not. If you want to consider the case in which D and Di
_are_ normal constituents of the mineral,then just consider the number
of atoms of "normal constituents" to be zero in my discussion.
I think (without any data) that it is pretty common for P, D, and Di
_not_ to be "normal constituents" of any noticeable number of
minerals.
>The best way I can describe the way I understand minerals form in a cooling
>melt is with a (somewhat bizarre) sports analogy. Let's say I'm in charge
>of creating a sports team to compete on my country's behalf in the
>Olympics. It hasn't yet been decided *which* sport my team is going to
>compete in -- that will be decided by the collective skills of the people I
>manage to recruit for the team. (Well, I said it was a bizarre analogy!)
Sure is so far! <grin>
>So, I have a large pool of kickers, catchers, throwers, runners, hitters
>and swimmers to choose from. All of their names are thrown into a hat, and
>I have to choose nine of them, without knowing who they are.
>
>If I pull out the names of a catcher, a thrower, four hitters, three
>runners, no kickers and no swimmers, I can field a baseball team. If I get
>two catchers, two throwers and five swimmers, and no kickers, runners or
>hitters, I have a water polo team. What I actually end up with is five
>kickers and four runners, which means that I have a football (American or
>Canadian football) team.
This is getting a bit far afield from mineral solidification ... in
mineral solidification the environment (temperature, pressure) is
significant.
Somewhat, but it's not an analogy for isochron dating; it's an analogy
for generic dating, since you don't have an analog for Di.
Eric
> On Sun, 11 Aug 2002 13:21:00 +0000 (UTC), e...@FAKEADRESS.com
> (Eric Rowley) wrote:
> >From: muz...@aol.com (zoe_althrop)
<snip>
> >You愉e assuming that all the contents of the rock stay together
> >for all time, rock forms, rock melts, rock solidifies. That
> >isn't nessesarily how it works, if the rock erodes, P is water
> >soluble and D isn't then the D might collect in the river bends
> >while the P is washed out to sea and ends up in limestone
> >hundreds of miles away. If, 100 million years later both the
> >river sediment and the limestone are melted and form new rocks
> >then one rock will have "too much" P and the other "too much" D
> I'm talking about those rocks that you consider to be closed.
They are considered closed after solidification,
not from the begining of time to the last solidification.
> It
> sounds as if you are saying that rocks cannot be reliably dated
> because of the conditions that you mention above.
No, I惴 saying you can愒 make assumtions about how much P and D
they "should" have had at solidification.
>How can we be
> certain that such conditions do not apply to any and all rocks
> dated?
>You're making a case for me, you know.
No I惴 not!
You say that the amounts of P and D in the melt should be dependent
on the age of the earth at that time, I giva a list of reasons why
that isnæ„’ nessesarily so, how is that making your case for you?
> However, the particular rocks I'm interested in are those rock
> layers that are associated with fossils --layers of igneous rock
> that have not eroded but remain as a layer between or over layers
> of fossil-bearing sedimentary rock. Since fossil-bearing layers
> are dated by the age of the igneous rock that intrudes or forms a
> layer above the fossils, such layers had better not leach or be
> contaminated, right?
Right.
And as far as we know they donæ„’.
And if they did they would be very very unlikely to give a straight
line on an isocron graph.
> >> Therefore, should samples today have high quantities of P
> >> in relation to Di, as reflected by high quantities of oldD/Di?
> >Not nessesarily, the "extra" P could have ended up in other
> >rocks so you would have to grind up the whole earth and measure
> >P and D but even that wouldn't help,
> nothing so extreme. I'm only interested in those layers of rock
> that have not eroded and joined the rivers or oceans.
Zoe, there are very few rocks left over from the time the earth
formed and they arenæ„’ likely to be found between fossil layers.
All other rocks have been recycled one way or another, and if you
donæ„’ know where the rocks that went into a melt came from there
is no way to know how much OriginalP and OldD should come out of
it.
> >there could have been "too much" D when the earth was originaly
> >formed and the "extra" P could have ended up in the other
> >planets. Or there could have been "too much" D when the solar
> >system formed, the supernova/god that created the P may well
> >have created some D as well so just comparing amounts of P and D
> >does not tell you the age of anything unless you have some
> >knowledge of the original amounts!
> it's speculation at this point, right? And how do you determine
> that supernovas create D from the start?
A: See what elements are in stars and in the material left after
an exploding supernova, and yes it is possible to tell what elements
there are in stars and out in space millions of miles from here
(look up spectrometry)
B: Determine what conditions are likely in stars before and during
a supernova explosion, experiment to see how atoms behave under
various conditions and do a lot of calculations.
> Are you saying that D
> does not necessarily have to be a result of parent decay?
Yes.
However any changes to the amount of D in rocks here on earth
are a result of "parent decay"
> And if so, how so?
It means that OldD isnæ„’ a reliable indicator of the earths age.
<snip>
> >Why should we care what "quantity of P that is characteristic of
> >that particular type of mineral"?
> we should care, if for no other reason than that this may be
> information that comes in useful at some later date. Knowledge is
> never wasted.
True, I was thinking in the narrower terms of what is useful for
calculating the age of a rock using the isochron method.
> >It seems like you are trying to sneak your "accumulated age
> >product" back into the equation?
> sneak? I'm openly investigating such a possibility as accumulated
> age product being mistaken for newD. For now, I'm looking at the
> actual premises or assumptions upon which the isochron is based,
> not whether oldD/Di can be determined if the starting assumption
> is correct.
OK, but the relative amounts of OriginalP and OldD are not among
those starting assumptions.
> >What we need for the age calculation are P, D and Di
> >measurements for at least two (usualy at least 6 are used)
> >samples (with differing P/Di ratios) and the knowledge that
> >OldD/Di ratios are the same for all samples from the same melt.
> >We don't need "characteristic" quantities of anything.
> not for the isochron, maybe, but for general knowledge, yes. It
> could be quite useful.
True, but if we are still discusing the isochron method it would
be simpler to stick to whatæ„€ relevant to the isochron method.
> >> >The fact that P is taken up in different amounts when
> minerals >> >solidify is _totally_ different from the fact that
> oldD/Di is >> >the same for all samples. >
> >> I understand that. But if quantity of oldD is fairly high in
> >> relation to Di, and we have not yet made it to even 0.116
> percent >> of the half life of Lutetium (if that's what Lu stands
> for), this >> would indicate a high quantity of P, right? Either
> that, or more >> time has passed for this mineral than even the
> age of the universe. >
> >Or some of the OldD was there from the begining.
> is there any evidence that D can form outside of parent decay?
Yes, but not on earth.
> If not, then there is no basis for claiming that oldD was there
> from the beginning, is there?
Not from the begining of the universe anyways but thatæ„€ not
what you mean is it?
As soon as there was any P it would have started decaying to D
and that D doesnæ„’ have to stay in the same place as the nondecayed
part of the P that partly decayed to produce the D unless itæ„€
in solid rock.
<snip>
> so I was correct, then, when I said that the electron is what
> made the difference in the element, causing isotope formation?
No where did you get that idea from?
The number of Protons determines the element.
The number of Neutrons determines the isotope.
The number of Electrons is normaly the same as
the number of Protons, any change in the number of
Electrons is basically temperary.
<snip>
> >> I meant, the QUANTITY of Di always remains the same, doesn't
> >>it? It never becomes more, it never becomes less?
> >It would probaly be created in supernovas all the time, but
> >here on earth the quantity should be pretty near constant
> >since the earths formation.
> what evidence do you have, Eric, that Di is probably created in
> supernovas?
Me personally? Nothing.
But I扉e read a bit about it and it seems to me that the physicists
and cosmologists have a pretty good idea of what goes on in stars
and nova based on measurments of whats in the stars, whats in space
presumably thrown out of supernova, measurements of how atoms (ions)
react when they hit each other at different speeds and theoretical
calculations.
>Do you consider that Earth was once a supernova, and therefore...?
The material that makes up the solar system (including the earth)
is thought to have come partly/largely?/mainly? from a (or several?)
supernova.
>Even so, the Earth is supposedly done with being a
> supernova now, so we can rely on the constancy of Di?
Depends, we can rely on the amount of Di staying the same in any
particular rock, but not on different rocks all having the same
amounts.
<snip>
Eric
<My domain is rixtele>
Eric
<snip>
> I'm not trying to analyze age since solidification, at the
> moment. I'm just curious about OldD/Di and its behavior. For
> instance, if a melt contains exactly 100 atoms of Di and 210
> atoms of D,
It would either be to small to see or have so little D and Di
that it would be very hard to measure.
We扉e been using numbers like this in examples because they are
easy to imagine and count with but in actual fact any real
sample would have contain billions of atoms.
1 gram of rock contains more than 1.000.000.000.000.000.000.000
atoms, naturaly not all of those would be D or Di but a lot
more than a couple hundred.
> how do these 100 atoms of Di and 210 atoms of D form
> same ratios for each mineral?
Just by being mixed and there not being any way for them to get
"unmixed" i nature.
> Would there be some oldD left
> unattached? What happens to those atoms?
They donæ„’ attach to each other, imagine taking 10 tons of sand,
dye 6 tons green and 4 tons red, mix it all together and dump
it in a river. Some of the sand will be added to sandbanks in
the river, some will get washed out to sea and form part of a
beach. Get a bucket of sand from any sandbank or the beach and
you should find that the ratio of green to red sand is 3:2 that
is how the D/Di ratio stays constant too.
<snip>
> so is there evidence that new quantities of D and P are being
> formed out there in the universe? What is this evidence?
Read a book on cosmology.
> But even so, there is no new input here on Earth, right?
Right.
(But there is old input from the time when the earth was formed)
> And that is the only place we're concerned with right now.
Yes, so where is this going?
> >If the half-life of P is large compared to the age of the
> >universe, then most of the oldD came from supernovae (or
> >whatever) and the average oldD/Di ratio doesn't change much over
> >time. If the half-life of P is much less than the age of the
> >universe, and there was a lot of P created by supernovae or
> >whatever, then most of the oldD came from decay of P and the
> >average oldD/Di ratio is increasing
> >faster over time than in the first case.
> is there a way to turn the "ifs" into "whens"?
I would just replace the "ifs" with "whens", how does that
change the meaning in your view?
As far as I can see it just depends on which element we call P.
Jon Fleming
<derekste...@yahoo.com> wrote:
<snip>
>There have been moments along the way when I wondered just how much Zoe
>actually understood, and whether it wouldn't be more productive to stop the
>discussion completely and go back to the very, *very* basics -- mineral
>formation, radioactive decay and the like -- and then very gradually bring
>her up to speed. I can't have been the only one to think that, can I?
Nope, you're not the only one.
>So
>why didn't any of us ever say that?
I either did or intimated it, but I don't know in which messages.
It's a tad tricky finding things in these threads.
Of course, going back to the basics and trying to teach her al this
stuff on line is impractical.
>This is your fault too, Zoe. As much as you might claim to be a simple,
>unsophisticated country girl, you have a degree of arrogance and
>self-satisfaction that's sadly misplaced in this sort of discussion. You
>seem to assume right off the bat that whatever you *think* you know about a
>subject is:
>a) accurate and
>b) so detailed as to constitute a sufficient understanding.
>This is rarely the case, and it almost always requires a monumental
>struggle to get you to see it. You're like the would-be flying sheep in the
>Monty Python episode, of whom the farmer says, "Once they gets an oidear in
>their 'eads, there's no shiftin' it."
>
>You should rely a little less on your own sense of satisfaction with your
>comprehension of an idea (I've never yet seen you announce your
>satisfaction with an idea that was *right*), and put a little more trust in
>the consensus of the group as to whether you've understood.
I assume you don't mean she should put more trust in the consensus of
the group as to what is correct ... she's not going to do that. I
think I understand your point.
>Ask more
>questions, even if they seem very simple and basic. *Especially* if they
>seem very simple and basic, because this is where you tend to go wrong.
Second the motion.
Derek Stevenson
> On Wed, 14 Aug 2002 16:20:04 +0000 (UTC), "Derek Stevenson"
> <derekste...@yahoo.com> wrote:
> >"Jon Fleming" <jo...@fleming-nospam.com> wrote in message
> >news:8ahklusid1n9hiab6...@4ax.com...
> >> Probably not. The amount of P taken up in the mineral depends on the
> >> chemistry and crystal structure of the mineral _and_ the amount of P
> >> in the melt. The amount of P in the melt will vary from one melt to
> >> another.
> >
> >I'm not sure you understood Zoe's question.
>
> That's a distinct possibility.
>
> >I suspect that by "minerals of
> >the same type", she actually means "the same mineral", and by "an
> >equivalent amount of P", she actually means "the same proportion of P".
In
> >other words, she's asking whether all crystals of, say, shazamite should
> >contain the same ratio of P to all other elements.
>
> Ah. That certainly is a valid question, whether or not it's the
> question she meant to ask.
>
> The answer is "maybe". The atoms in which we are interested may or
> may not be a part of the chemical formula of the mineral. It is
> certainly reasonably common for them not to be in the chemical
> formula. They are often incorporated into the mineral by substituting
> for a sufficiently similar atom that _is_ in the chemical formula, or
> by physical entrapment as the mineral crystal grows around them.
> Without stating so, I was addressing the last two possibilities
> (perhaps because I work with semiconductors, and they have substitute
> atoms.)
Ah. That makes it clearer.
> If the atoms in which we are interested _are_ part of the chemical
> formula of the mineral then, of course, then all samples of that
> mineral will contain the same proportion of, for example, P.
>
> I just don't know enough to say anything about the relative
> frequencies of incorporation by being part of the chemical formula and
> incorporation by other means. I know that the latter happens; e.g.
> zirconia (ZrSiO4, see <http://webmineral.com/data/Zircon.shtml>) which
> is widely used for U-Pb concordia-discordia dating (because
> concordia-discordia dating requires zero or essentially zero oldD, and
> zircons strongly reject lead but accept uranium at solidification and
> hold lead created in the solid by decay of uranium pretty well).
>
> >And the answer to that, I think, should be "yes". By definition. All
> >crystals of shazamite should contain one atom of P, two atoms of carbon,
> >three of silicon and one of oxygen, because that's what *makes* it
> >shazamite. If its composition was two atoms of P, three of carbon, no
> >silicon and two oxygen, it wouldn't be shazamite, it would be sivanite.
> >
> >Unless I'm as confused as Zoe here ...
>
> You're discounting the significant possibilities I discussed above.
Because I wasn't aware of them. I appreciate the correction.
Ah, well. Every analogy has to break down at some point.
> >As it happens, there are two kinds of kickers -- kickers who learned
their
> >skills playing football, and kickers who learned their skills playing
> >soccer. As I recruit my team, I'm not aware of this distinction: the
> >information doesn't appears on the players' resumes, and it makes no
> >difference to me anyway. Since the ratio of football-kickers to
> >soccer-kickers in the talent pool is 4:1, the odds are good that the
same
> >ratio will apply to the five kickers on my football team.
> >
> >But in what way are the kickers not "normal constituents" of a football
> >team?
You've explained that above. Thanks.
I intended the kickers who learned their trade playing soccer to serve as
the analog for Di: they originate in a different manner from the
football-kickers (football-kickers representing D, be it old or new), but
are indistinguishable from them for purposes of forming the team. They can
be identified as a different breed in a subsequent analysis, however.
Derek Stevenson
> >There have been moments along the way when I wondered just how much Zoe
> >actually understood, and whether it wouldn't be more productive to stop
the
> >discussion completely and go back to the very, *very* basics -- mineral
> >formation, radioactive decay and the like -- and then very gradually
bring
> >her up to speed. I can't have been the only one to think that, can I?
>
> Nope, you're not the only one.
>
> >So
> >why didn't any of us ever say that?
>
> I either did or intimated it, but I don't know in which messages.
> It's a tad tricky finding things in these threads.
>
> Of course, going back to the basics and trying to teach her al this
> stuff on line is impractical.
More so than spending -- what is it now, five months -- embroiled in an
argument which we know realize she never understood?
> >This is your fault too, Zoe. As much as you might claim to be a simple,
> >unsophisticated country girl, you have a degree of arrogance and
> >self-satisfaction that's sadly misplaced in this sort of discussion. You
> >seem to assume right off the bat that whatever you *think* you know
about a
> >subject is:
> >a) accurate and
> >b) so detailed as to constitute a sufficient understanding.
> >This is rarely the case, and it almost always requires a monumental
> >struggle to get you to see it. You're like the would-be flying sheep in
the
> >Monty Python episode, of whom the farmer says, "Once they gets an oidear
in
> >their 'eads, there's no shiftin' it."
> >
> >You should rely a little less on your own sense of satisfaction with
your
> >comprehension of an idea (I've never yet seen you announce your
> >satisfaction with an idea that was *right*), and put a little more trust
in
> >the consensus of the group as to whether you've understood.
>
> I assume you don't mean she should put more trust in the consensus of
> the group as to what is correct ... she's not going to do that. I
> think I understand your point.
Yes, I probably should have made more clear what I was asking. I don't mean
that you should necessarily *believe* what the rest of the group believes,
Zoe -- I just think that you should make more of an effort to get the
*content* and the *background* of those beliefs right.
And yes, I recall your argument from some time back about thinking that
you're able to tell when you're on the wrong road and not needing to go any
further to *know* that it's wrong. Ignore that feeling -- it's very
unreliable. Anyway, don't you owe it to yourself to figure out why so many
people think they're on the *right* road?
Jon Fleming
<derekste...@yahoo.com> wrote:
>> >her up to speed. I can't have been the only one to think that, can I?
>>
>> Nope, you're not the only one.
>>
>> >So
>> >why didn't any of us ever say that?
>>
>> I either did or intimated it, but I don't know in which messages.
>> It's a tad tricky finding things in these threads.
>>
>> Of course, going back to the basics and trying to teach her al this
>> stuff on line is impractical.
>
>More so than spending -- what is it now, five months -- embroiled in an
>argument which we know realize she never understood?
Eight months, I think.
Yes. It would take years in a classroom setting to teach her what she
should know to discuss these issues intelligently. More time on-line.
Derek Stevenson
news:ajeebb$1a72vp$1...@ID-139629.news.dfncis.de...
> And yes, I recall your argument from some time back about thinking that
> you're able to tell when you're on the wrong road and not needing to go
any
> further to *know* that it's wrong. Ignore that feeling -- it's very
> unreliable. Anyway, don't you owe it to yourself to figure out why so
many
> people think they're on the *right* road?
And, I should have added, don't you owe it to *them*? It seems to me that
if someone holds beliefs that you don't share, it's a matter of simple
human courtesy to try to determine *why* he or she thinks that way, rather
than simply dismissing them as deluded.
zoe_althrop
<jo...@fleming-nospam.com> wrote:
snip>
>>it's speculation at this point, right? And how do you determine that
>>supernovas create D from the start?
>
>The current scientific theory, well investigated and supported, is
>that _all_ elements heavier than iron are created in supernovae, and
>in a few other rare conditions. See
><http://www.aoc.nrao.edu/~kdyer/supernova.html> and
><http://imagine.gsfc.nasa.gov/docs/ask_astro/answers/010125a.html>.
how do they determine how high a temperature should be in order to
create elements? I mean, 1 billion degrees to form carbon or 5
million degrees to form hydrogen....how are such experiments
conducted?
snip>
>>is there any evidence that D can form outside of parent decay?
>
>Yes. There is lots of evidence that D can be and is formed in
>supernovae. See <http://zebu.uoregon.edu/textbook/energygen.html> and
><http://helios.gsfc.nasa.gov/nucleo.html> for starters. The key word
>is "nucleosynthesis".
these supernovae that suddenly appear out there in the universe, is
there a way to tell if their light is truly a result of an explosion,
or if, instead, it just might be that their light has finally reached
us, but that the galaxy's pathway has taken it into our view and then
away again as it continues its orbit further out into space? Or what
if two galaxies are passing each other so that the doubling of their
light causes a temporary brightening which then fades as they pull
away from each other?
<snip my misunderstanding of electrons, which I hope I now understand
at some minimal level>
thanks for the links. They all talk very authoritatively about the
death of stars and so forth. I wonder how they go about determining
that the appearance and disappearance of these lights out there mean
death? Distance, motion, and time can cause illusions, especially as
they become more exaggerated.
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
>On Tue, 13 Aug 2002 23:55:14 +0000 (UTC), muz...@aol.com (zoe_althrop)
>wrote:
snip>
>>but if a comparison were made among minerals of the same type, they
>>should all carry an equivalent amount of P?
>
>Probably not. The amount of P taken up in the mineral depends on the
>chemistry and crystal structure of the mineral _and_ the amount of P
>in the melt. The amount of P in the melt will vary from one melt to
>another.
I meant, within the same melt.
snip>
>It's difficult to explain, or illustrate with an example, what happens
>step-by-step as a mineral solidifies without getting into some pretty
>complex probability concepts that you need years of work to comprehend
>(and I'd have to study up some before doing the numbers).
>
>However, the step-by-step scenario doesn't matter. No matter how D and
>Di are incorporated into the each mineral that solidifies as the melt
>cools, they will be incorporated in that mineral in the same
>proportion that they are in the melt. This follows immediately from
>the fact that the solidification process cannot distinguish between D
>and Di.
this is a big let-down for me, Jon. Here I was, trusting that your
claim for same ratios of D to Di was based on laws of chemistry.
Instead, I see your claim is based on probability and statistics.
My misconception was that in order for any mineral to form, it HAD to
have a certain quantity of D and Di before it would be able to finish
its formation and be recognized as a particular mineral. I had
visions of the mineral waiting patiently until the right quantity of
Di came along and then, zap, it closed up and became, say, zircon or
biotite or whatever. Now I see that the concept of same ratios is
based strictly on a concept of probability, not laws of the chemical
construction of the particular mineral. In other words, a mineral will
still be that mineral, whether or not it contained D or P; am I
correct?
So tell me, then, will this probability fail to work if the melt were
not thoroughly mixed? What if the original elements remained in their
original locations, melted, ejected and cooled -- with no thorough
mixing of the slurry?
>When an atom moves from the melt to a solid mineral, it is either an
>atom of that mineral's normal constituents or it is an atom of D or it
>is an atom of Di. We don't care about the case where it is an atom of
>the mineral's normal constituent. In the case where it is an atom of
>D or an atom of Di, the probability that it is an atom of D is the
>same as the number of atoms of D divided by the total number of atoms
>of D and Di, and similarly for Di. (In your example, the probability
>that it is an atom of D is 210/(100+210) = 210/310 = 67.74% and the
>probability that it is an atom of Di is100/310 = 32.36%).
>
>As the mineral solidifies (incorporating trillions of trillions of
>trillions of atoms in a real mineral) this "coin flip" of selecting an
>atom of D or Di randomly is repeated trillions of trillions of
>trillions of times. The ratio of D to Di in the mineral will wind up
>in exactly (or indistinguishably close to) the same proportion as the
>ratio was in the melt, just as if you toss a fair coin trillions of
>trillions of trillions of times you will wind up with 50% heads and
>50% tails.
are you equating same ratios of D/Di to heads and non-same ratios of
D/Di to tails? Then that would say that with sufficient flips, you
would have half your minerals that are in same ratios and the other
half are not in same ratios.
And if this is the case, what is the probability that the samples you
choose to test for your isochron are from that 50 percent that have
same ratios versus those that don't?
Or, again, are you equating D to heads and Di to tails? If so, then
you would have half the minerals carrying D and the other half
carrying Di, and it wouldn't matter since D and Di are
indistinguishable in the solidification process.
Or are you telling me that probability and statistics will ensure that
every mineral that forms will take up exactly the same proportion of D
to Di, regardless of how well mixed or poorly mixed the melt is?
That is like saying, here we have two trillion atoms of D and one
trillion atoms of Di. Since they are indistinguishable from each
other, it is predicted that the first mineral that forms will happen
to take up two atoms of D and one atom of Di, and the next mineral
will do the same, and the third mineral will do the same, again and
again and again, without missing a beat. That would be like tossing a
coin a million times and each time it lands on heads, never tails.
Wouldn't real probability work like this? A mineral takes up 4 atoms
of D and 1 atom of Di in one mineral, and the next mineral, since it
cannot distinguish, will take up 10 atoms of D and 5 atoms of Di, and
eventually, with sufficient tries, you may even find a mineral taking
up the quantity of D to Di in the same ratios as found in the melt.
And since this is not a law of the chemical makeup of the mineral,
there is no reason for each mineral to take up exactly the same ratios
of each isotope. But overall, you would probably find that all of D
and Di has been absorbed by the solidifying minerals in the same ratio
of 2:1 COLLECTIVELY for all the minerals. But individually, they
would carry whatever quantity of oldD and Di happened to be close at
hand before they solidified.
And this would be especially true if the remelt is not stirred and
thoroughly mixed, but simply melted in place and ejected.
Can you explain why this should not be the case?
I have vague memories of taking statistics at the beginning of
college, but have forgotten how it works, so consider the above a real
layman's attempt to understand probability and statistics. :-\
>So what happens to the melt? As long as atoms of D and Di are removed
>from the melt (into minerals) in the same proportion as they are in
>the melt, the ratio of D to Di in the melt will not change.
agreed....that is, if your assumption is correct. But I think that
assumption can be questioned. Are atoms of D and Di consistently
removed from a melt in the same proportion as they are in the melt,
from one mineral to the next? I see no reason why they should for
each individual mineral, but the ratio should remain the same for the
OVERALL resolidified rock.
> A few
>examples:
>
>Suppose we have 210 atoms of D and 100 atoms of Di, or a ratio of 2.1.
>Say 21 atoms of D go into a mineral, and therefore 10 atoms of Di also
>go into that mineral (so the proportion of D to Di in the mineral is
>21/10 or 2.1). There are 210 - 21 = 189 atoms of D left in the melt
>and 100 - 10 = 90 atoms of Di left in the melt. The D/Di ratio in the
>melt is 189/90 = 2.1, the same as before.
>
>Or suppose instead that 42 atoms of D go into the mineral, and
>therefore 20 atoms of Di go into the mineral. The D/Di ratio of the
>mineral is 2.1. There are 168 atoms of D left in the melt, and 80
>atoms of Di. 168/80 = 2.1. There are 42 atoms of D in the mineral
>and 20 atoms of Di in the mineral. 42/20 = 2.1.
>
>It's easy to prove with simple algebra that this happens no matter
>_how_ _many_ atoms of D and Di go into each mineral, as long as they
>go into the minerals with probabilities that depend on their
>proportion in the melt. Since solidification cannot distinguish
>between D and Di, they go into the minerals with probabilities that
>depend on their proportion in the melt, and the ratio of D to Di is
>the same in the melt and all minerals solidified from that melt.
wouldn't the melt have to be thoroughly stirred to distribute the
isotopes evenly and ensure an even mix? In spite of probability and
statistics, if the melt contains lots of D in one portion and sparse
Di in that same section, or sparse D in another location but lots of
Di over there, the minerals don't distinguish between them and simply
absorbs whatever D or Di is within its reach. Overall, the ratio for
all minerals would be 2:1, but for individual minerals, the ratios
would be all over the map.
What is the probability that the isochronists would happen to choose
only those samples that happen to have collected in the same 2:1
proportions as the entire melt?
>>Would there be some oldD left unattached?
>
>There might be oldD left that is not incorporated into other minerals.
>It depends on how easily D is taken up by the last few minerals that
>solidify.
>
>>What happens to those atoms?
>
>They solidify into solid D, tiny grains of D and Di intermixed in the
>ratio they were in the melt and in all other minerals that solidified
>from that melt, nestled in between the other minerals.
how can D be intermixed with Di in these last few minerals if all the
D has been absorbed?
snip>
>>have supernovae been sampled for original D?
>
>Yes. Each element gives off a distinctive pattern of light when it is
>hot and absorbs a distinctive pattern of light when it is cold. By
>analyzing the content of light from supernovae, we can detect what
>elements are there. References below.
I've been looking at your references. Are you saying that they are
able to detect the light patterns of all varieties of D, such as
lead-206, 207, 208, strontium-87, argon-40, Neodymium-143, to name
just a few? Fascinating.
>(Interesting side factoid: helium was first discovered when
>scientists used this method on the Sun, and was only discovered on
>Earth afterwards. The name "helium" comes from "helios", the Greek
>for "Sun". The method works. See
><http://antwrp.gsfc.nasa.gov/apod/ap010120.html>.)
I can't seem to acess this one.
snip>
>>what evidence do you have that supernovae produce D outside of decay
>>from P?
>
>Detection of D in supernovae.
and how are the light patterns in the supernovae detected? Not
through that electron beam, right?
>Theoretical physics, which predicts that D can only be produced by
>means other than radioactive decay under conditions such as found in
>supernovae, and further predicts that D _must_ be produced in such
>conditions.
well, now that is more understandable -- theoretical stuff.
Predictions. But no hard evidence to support such predictions, right?
>>>The other possibility is true; supernovae produce oldD, just like they
>>>produce most of the elements that make up the Earth.
>>
>>how does one go about measuring the products of a supernova?
>
>By analyzing the light (including many wavelengths that we don't see
>directly). See
><http://imagine.gsfc.nasa.gov/docs/teachers/lessons/supernova/supernova_chemistry.html>,
><http://cosmology.berkeley.edu/Education/DEMOS/Desk_Top_Stars/LS.html>,
>and
><http://imagine.gsfc.nasa.gov/docs/teachers/lessons/xray_spectra/spectra_cover.html>.
not all of these came up for me, but thanks for sharing, anyway.
snip>
----
zoe
zoe_althrop
<derekste...@yahoo.com> wrote:
snip>
zoe wrote:
>> >but if a comparison were made among minerals of the same type, they
>> >should all carry an equivalent amount of P?
Jon:
>> Probably not. The amount of P taken up in the mineral depends on the
>> chemistry and crystal structure of the mineral _and_ the amount of P
>> in the melt. The amount of P in the melt will vary from one melt to
>> another.
>
>I'm not sure you understood Zoe's question. I suspect that by "minerals of
>the same type", she actually means "the same mineral", and by "an
>equivalent amount of P", she actually means "the same proportion of P". In
>other words, she's asking whether all crystals of, say, shazamite should
>contain the same ratio of P to all other elements.
>And the answer to that, I think, should be "yes". By definition. All
>crystals of shazamite should contain one atom of P,
what element is P here?
> two atoms of carbon,
>three of silicon and one of oxygen, because that's what *makes* it
>shazamite. If its composition was two atoms of P, three of carbon, no
>silicon and two oxygen, it wouldn't be shazamite, it would be sivanite.
>
>Unless I'm as confused as Zoe here ...
no way.
>[snip]
>
>> However, the step-by-step scenario doesn't matter. No matter how D and
>> Di are incorporated into the each mineral that solidifies as the melt
>> cools, they will be incorporated in that mineral in the same
>> proportion that they are in the melt. This follows immediately from
>> the fact that the solidification process cannot distinguish between D
>> and Di.
>>
>> When an atom moves from the melt to a solid mineral, it is either an
>> atom of that mineral's normal constituents or it is an atom of D or it
>> is an atom of Di. We don't care about the case where it is an atom of
>> the mineral's normal constituent.
>
>Okay, now *I'm* confused. Why this distinction between D, Di and "normal
>constituents"? Aren't D and Di "normal constituents" of the minerals
>they're incorporated into?
oh, yes, that would be my question, too.
>The best way I can describe the way I understand minerals form in a cooling
>melt is with a (somewhat bizarre) sports analogy. Let's say I'm in charge
>of creating a sports team to compete on my country's behalf in the
>Olympics. It hasn't yet been decided *which* sport my team is going to
>compete in -- that will be decided by the collective skills of the people I
>manage to recruit for the team. (Well, I said it was a bizarre analogy!)
>
>So, I have a large pool of kickers, catchers, throwers, runners, hitters
>and swimmers to choose from. All of their names are thrown into a hat, and
>I have to choose nine of them, without knowing who they are.
>
>If I pull out the names of a catcher, a thrower, four hitters, three
>runners, no kickers and no swimmers, I can field a baseball team. If I get
>two catchers, two throwers and five swimmers, and no kickers, runners or
>hitters, I have a water polo team. What I actually end up with is five
>kickers and four runners, which means that I have a football (American or
>Canadian football) team.
>
>As it happens, there are two kinds of kickers -- kickers who learned their
>skills playing football, and kickers who learned their skills playing
>soccer. As I recruit my team, I'm not aware of this distinction: the
>information doesn't appears on the players' resumes, and it makes no
>difference to me anyway. Since the ratio of football-kickers to
>soccer-kickers in the talent pool is 4:1, the odds are good that the same
>ratio will apply to the five kickers on my football team.
>
>But in what way are the kickers not "normal constituents" of a football
>team?
oh, dear...
>Now, just to clarify my own understanding, let's continue the analogy.
>
>Runners who play football, it turns out, eventually strengthen their legs
>to the point where they become kickers. The amount of time this takes
>varies depending on the individual, but if you take a group of four runners
>and have them play football, a year later you'll have two runners and two
>kickers. So if my football team plays together for a year, at the end of
>that time I'll have gone from five kickers and four runners to seven
>kickers and two runners.
>
>A journalist writing a profile of the team wants to figure out how long
>we've been together. Since we went in for football, he knows that we must
>have started out with five kickers and four runners. He knows that we know
>have seven kickers and two runners. He knows that of the seven kickers, six
>are football-kickers and one is a soccer-kicker (because football kickers
>always get a tattoo of a football on their arms, while soccer-kickers
>always get one of a soccer ball). He knows that in one year, two out of
>every four runners who play football become football-kickers. Therefore,
>since I had to have started out with five kickers, two of my seven kickers
>must be former runners and that the team has been together a year.
>
>Am I on the right track here? And Zoe, is this making any sense to you?
well, I don't follow sports much so you kind of lost me along the way.
How about a more feminine analogy?
don't bother. I don't want you to have to struggle. (grin)
>(If
>you can't figure out how the elements of this analogy correspond to
>isochron dating, just ask. Note that this analogy applies to figuring out
>the history of a *single sample*, not to the multiple samples required to
>plot an isochron.)
okay. Maybe you can explain the history of oldD from its creation?
----
zoe
zoe_althrop
(Eric) wrote:
snip>
zoe wrote:
>> Would there be some oldD left
>> unattached? What happens to those atoms?
>
>They don´t attach to each other, imagine taking 10 tons of sand,
>dye 6 tons green and 4 tons red, mix it all together and dump
>it in a river. Some of the sand will be added to sandbanks in
>the river, some will get washed out to sea and form part of a
>beach. Get a bucket of sand from any sandbank or the beach and
>you should find that the ratio of green to red sand is 3:2 that
>is how the D/Di ratio stays constant too.
without thorough mixing, the ratio overall would be 3:2, but one
bucket of sand might contain a ratio of 1:1 and another bucket from a
different area might contain a ratio of 2:1, and so on. But if all
ratios were somehow added together, the ratio would be 3:2.
snip>
----
zoe
zoe_althrop
<derekste...@yahoo.com> wrote:
snip>
>And, I should have added, don't you owe it to *them*? It seems to me that
>if someone holds beliefs that you don't share, it's a matter of simple
>human courtesy to try to determine *why* he or she thinks that way, rather
>than simply dismissing them as deluded.
ahhh, if you would but follow your own advice. You do dismiss
creationists as deluded, don't you?
----
zoe
zoe_althrop
<derekste...@yahoo.com> wrote:
snip>
>This is your fault too, Zoe. As much as you might claim to be a simple,
>unsophisticated country girl,
you mean, my claiming to be a nonscientist makes me into an
unsophisticated country girl? Hmmm...I see that the world is divided
into sophisticated city-slicker scientists and unsophisticated country
folk nonscientists -- at least in your mind.
>you have a degree of arrogance and
>self-satisfaction that's sadly misplaced in this sort of discussion.
yes, sir, yes, sir, whatever you say, sir...
better?
>You
>seem to assume right off the bat that whatever you *think* you know about a
>subject is:
>a) accurate and
>b) so detailed as to constitute a sufficient understanding.
>This is rarely the case, and it almost always requires a monumental
>struggle to get you to see it.
don't struggle, Derek, just relax. Apparently, Jon does not have to
struggle. He just smoothly rolls out his information and doesn't even
break a sweat.
> You're like the would-be flying sheep in the
>Monty Python episode, of whom the farmer says, "Once they gets an oidear in
>their 'eads, there's no shiftin' it."
hmmm, I don't see YOUR ideas shifting either, Derek.
>You should rely a little less on your own sense of satisfaction with your
>comprehension of an idea (I've never yet seen you announce your
>satisfaction with an idea that was *right*), and put a little more trust in
>the consensus of the group as to whether you've understood. Ask more
>questions,even if they seem very simple and basic. *Especially* if they
>seem very simple and basic, because this is where you tend to go wrong.
>
>Above all, don't -- as I keep trying to remind you -- assume that just
>because an idea makes you feel good, that it's right. Or vice versa.
One of my mottos: Live by principle, not by feeling, for the most
part.
----
zoe
zoe_althrop
(Eric) wrote:
snip>
zoe wrote:
>> it's speculation at this point, right? And how do you determine
>> that supernovas create D from the start?
>
>A: See what elements are in stars and in the material left after
>an exploding supernova, and yes it is possible to tell what elements
>there are in stars and out in space millions of miles from here
>(look up spectrometry)
looked it up, and it seems to say that a substance is bombarded with
an electron beam in order to determine its characteristics. How do
you get your electron beam out there in those far-flung galaxies?
>B: Determine what conditions are likely in stars before and during
>a supernova explosion, experiment to see how atoms behave under
>various conditions and do a lot of calculations.
how do you determine what conditions are likely, having never been in
a position to personally check the stars or experienced supernovae?
How is the extrapolation done?
maybe you shouldn't bother to answer that since I'm sure there is a
large body of work that explains it, and I should read up on it
myself.
>> Are you saying that D
>> does not necessarily have to be a result of parent decay?
>
>Yes.
>However any changes to the amount of D in rocks here on earth
>are a result of "parent decay"
well, after all, isn't that all that we were discussing, conditions
here on earth? So why introduce supernovae and their supposed
production of D?
>> And if so, how so?
>
>It means that OldD isn´t a reliable indicator of the earths age.
so those rocks that have been dated to 3.8 billion years are not
reliable sources for age of the earth, either?
><snip>
>> >It seems like you are trying to sneak your "accumulated age
>> >product" back into the equation?
>
>> sneak? I'm openly investigating such a possibility as accumulated
>> age product being mistaken for newD. For now, I'm looking at the
>> actual premises or assumptions upon which the isochron is based,
>> not whether oldD/Di can be determined if the starting assumption
>> is correct.
>
>OK, but the relative amounts of OriginalP and OldD are not among
>those starting assumptions.
maybe they should be?
<snip your good electron explanation that hopefully will sink in this
time>
----
zoe
June
The most recent (and exhaustively studied) nearby supernova is 1987A.
below are some links to info and graphics wrt this. Look at the
pictures, read the texts. Not only visible light rays come from
supernova but gamma rays, x-rays, neutrinos, etc. also are
detected...*not* mistaken images of galaxies that are passing close to
each other!
Observations and evidence from supernova 1987A have confirmed various
hypotheses and theories from physics/astronomy including that supernovae
*do* create the heavier elements.
Here's a link to info on supernovae in general. The picture at the top
is a before (on right w/ arrow to original star that exploded) and after
(on left) of SN 1987A. There is a *huge* difference in the visible light
output.
http://www.rog.nmm.ac.uk/leaflets/supernovae/supernovae.html
Here's a link with a history of SN 1987A and some current research by a
Richard McCray. I strongly recommend you read this link. It delineates,
in fairly simple terms, many of the observations that were and are being
made about this explosion. This evidence cannot be explained away by
galaxy's passing each other! Note that these people made predictions
based on astrophysics in 1997 about the illumination of rings from this
explosion. Guess what! That prediction was confirmed in 2000 when rings
began to glow around SN 1987A.
http://cosmos.colorado.edu/astr1120/l6S6.htm
This link is to a bunch of images, graphics of the supernova with text
and links to more information. There's also images/text for other
supernova also. Click on the images themselves after you open each
window to get more info.
http://oposite.stsci.edu/pubinfo/novaesupernovae.html
Here's a brief chronological text on this supernova.
http://www.floridatoday.com/space/explore/stories/1998/021198g.htm
Look at the info on SN 1987A. There's more than just 'disappearing
lights' involved, Zoe. There are detailed observations of almost all
rays/particles from this explosion from the very beginning. They're
still getting new information from this.
>
> ----
> zoe
--
My 2¢ ß-}
June
Jon Fleming
wrote:
>On Wed, 14 Aug 2002 00:34:26 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>snip>
>
>>>it's speculation at this point, right? And how do you determine that
>>>supernovas create D from the start?
>>
>>The current scientific theory, well investigated and supported, is
>>that _all_ elements heavier than iron are created in supernovae, and
>>in a few other rare conditions. See
>><http://www.aoc.nrao.edu/~kdyer/supernova.html> and
>><http://imagine.gsfc.nasa.gov/docs/ask_astro/answers/010125a.html>.
>
>how do they determine how high a temperature should be in order to
>create elements? I mean, 1 billion degrees to form carbon or 5
>million degrees to form hydrogen....how are such experiments
>conducted?
Such experiments are not conducted on Earth, obviously. THe
temperature of stars in which this is happening can be measured. In
addition, the theoretical physics of nuclear interactions is used to
predict the temperature at which it can happen, and these predictions
agree with the astronomical measurements. (We have done a _lot_ of
research on theoretical nuclear physics, and we're very good at
predicting reality based on theory ... bombs, you know)
>snip>
>
>>>is there any evidence that D can form outside of parent decay?
>>
>>Yes. There is lots of evidence that D can be and is formed in
>>supernovae. See <http://zebu.uoregon.edu/textbook/energygen.html> and
>><http://helios.gsfc.nasa.gov/nucleo.html> for starters. The key word
>>is "nucleosynthesis".
>
>these supernovae that suddenly appear out there in the universe, is
>there a way to tell if their light is truly a result of an explosion,
>or if, instead, it just might be that their light has finally reached
>us, but that the galaxy's pathway has taken it into our view and then
>away again as it continues its orbit further out into space?
Yes. The two cases look totally different.
>Or what
>if two galaxies are passing each other so that the doubling of their
>light causes a temporary brightening which then fades as they pull
>away from each other?
Not even close to the amount of light from a supernova, and the
characteristics of the light are totally different.
Supernovae are identifiable.
Yup, but study and investigation and loads of data dispel the
illusions.
Jon Fleming
wrote:
>On Wed, 14 Aug 2002 20:27:24 +0000 (UTC), "Derek Stevenson"
No. He notes that creationists are deluded, but he does not dismiss
them.
sds
> On Wed, 14 Aug 2002 18:26:25 +0000 (UTC), eric....@FAKEADRESS.com
> (Eric) wrote:
>
> snip>
>
> zoe wrote:
>
> >> it's speculation at this point, right? And how do you determine
> >> that supernovas create D from the start?
> >
> >A: See what elements are in stars and in the material left after
> >an exploding supernova, and yes it is possible to tell what elements
> >there are in stars and out in space millions of miles from here
> >(look up spectrometry)
>
> looked it up, and it seems to say that a substance is bombarded with
> an electron beam in order to determine its characteristics. How do
> you get your electron beam out there in those far-flung galaxies?
We are all uneducated in everything at some point in our lives, and we are
all uneducated in some things at every point in our lives.
What's exceptional about Zoe is her lack of self-consciousness about this
common denominator of human kind (or her ability to ignore what
self-consciousness is there).
It's admirable, IMO.
Jon Fleming
wrote:
>On Wed, 14 Aug 2002 18:26:25 +0000 (UTC), eric....@FAKEADRESS.com
>(Eric) wrote:
>
>snip>
>
>zoe wrote:
>
>>> it's speculation at this point, right? And how do you determine
>>> that supernovas create D from the start?
>>
>>A: See what elements are in stars and in the material left after
>>an exploding supernova, and yes it is possible to tell what elements
>>there are in stars and out in space millions of miles from here
>>(look up spectrometry)
>
>looked it up, and it seems to say that a substance is bombarded with
>an electron beam in order to determine its characteristics. How do
>you get your electron beam out there in those far-flung galaxies?
That's only one of may ways of doing spectrometry. The light given
off by stars contains the information required for spectrometry. We
just analyze the light. The light given off or absorbed by each
element is characteristic of that element.
>>B: Determine what conditions are likely in stars before and during
>>a supernova explosion, experiment to see how atoms behave under
>>various conditions and do a lot of calculations.
>
>how do you determine what conditions are likely, having never been in
>a position to personally check the stars or experienced supernovae?
>How is the extrapolation done?
>
>maybe you shouldn't bother to answer that since I'm sure there is a
>large body of work that explains it, and I should read up on it
>myself.
Yes. However, I want to note that "extrapolation" is not the correct
word. Extrapolation is extending information from known data into
areas where there is no data. We _do_ have data for what goes on in a
supernova explosion, so we are not extending into an area where there
is no data.
What we are doing is _calculating_ using well-known and rock-solid
principles of physics, such as the laws of gravity, and _verifying_
against measured data.
>>> Are you saying that D
>>> does not necessarily have to be a result of parent decay?
>>
>>Yes.
>>However any changes to the amount of D in rocks here on earth
>>are a result of "parent decay"
>
>well, after all, isn't that all that we were discussing, conditions
>here on earth? So why introduce supernovae and their supposed
>production of D?
Because supernovae and their measured (not supposed) production of D
contributed to the oldD present on Earth, and this immediately leads
to the conclusion that the amount of oldD is not an indicator of the
age of the Earth.
>>> And if so, how so?
>>
>>It means that OldD isn´t a reliable indicator of the earths age.
>
>so those rocks that have been dated to 3.8 billion years are not
>reliable sources for age of the earth, either?
The rocks that have been dated to 3.8 billion years are reliable
sources for the age of the Earth, because either they were dated with
methods where oldD was not present or they were dated with methods
that are independent of oldD, such as isochron dating. See the list
at <http://www.gate.net/~rwms/AgeEarth.html>, the first figure in
<http://www.talkorigins.org/faqs/faq-age-of-earth.html>, and the
graphs (on the sixth page) and the table (on the seventh page) of
<http://lordibelieve.org/time/age4.PDF>. Note that _all_ these pages
took their data from Dalrymple; that book is _the_ standard resource.
>><snip>
>
>>> >It seems like you are trying to sneak your "accumulated age
>>> >product" back into the equation?
>>
>>> sneak? I'm openly investigating such a possibility as accumulated
>>> age product being mistaken for newD. For now, I'm looking at the
>>> actual premises or assumptions upon which the isochron is based,
>>> not whether oldD/Di can be determined if the starting assumption
>>> is correct.
>>
>>OK, but the relative amounts of OriginalP and OldD are not among
>>those starting assumptions.
>
>maybe they should be?
Nope. The list of assumptions that we have already are necessary and
sufficient, and you don't add assumptions just for grins.
"Necessary and sufficient" is a phrase that is often used, and it's
important to understand. "Necessary" means that, if you remove any of
the assumptions, what remains is not enough to completely solve the
problem. "Sufficient" means that this particular set of assumptions
_is_ enough to completely solve the problem. When your assumptions
are necessary and sufficient, you have the minimum number of
assumptions that you have to have in order to solve the problem. You
_never_ add assumptions to a necessary and sufficient set.
Jon Fleming
wrote:
>On Wed, 14 Aug 2002 16:20:04 +0000 (UTC), "Derek Stevenson"
><derekste...@yahoo.com> wrote:
>
>snip>
>
>zoe wrote:
>
>>> >but if a comparison were made among minerals of the same type, they
>>> >should all carry an equivalent amount of P?
>
>Jon:
>
>>> Probably not. The amount of P taken up in the mineral depends on the
>>> chemistry and crystal structure of the mineral _and_ the amount of P
>>> in the melt. The amount of P in the melt will vary from one melt to
>>> another.
>>
>>I'm not sure you understood Zoe's question. I suspect that by "minerals of
>>the same type", she actually means "the same mineral", and by "an
>>equivalent amount of P", she actually means "the same proportion of P". In
>>other words, she's asking whether all crystals of, say, shazamite should
>>contain the same ratio of P to all other elements.
>>And the answer to that, I think, should be "yes". By definition. All
>>crystals of shazamite should contain one atom of P,
>
>what element is P here?
He's making up a mineral name just so he can have a short word to use
to mean "any arbitrary mineral". It doesn't matter in the slightest
what element is P here. All he's saying that each minimum unit of a
mythical mineral that he made up includes one atom of P.
<snip>
>>> When an atom moves from the melt to a solid mineral, it is either an
>>> atom of that mineral's normal constituents or it is an atom of D or it
>>> is an atom of Di. We don't care about the case where it is an atom of
>>> the mineral's normal constituent.
>>
>>Okay, now *I'm* confused. Why this distinction between D, Di and "normal
>>constituents"? Aren't D and Di "normal constituents" of the minerals
>>they're incorporated into?
>
>oh, yes, that would be my question, too.
OK, look at my answer to Derek's post.
<snip>
>okay. Maybe you can explain the history of oldD from its creation?
Long before the Earth was formed, supernovae explosions and some
extremely massive stars produced P and D. The P immediately started
decaying into D. The P and D produced in stars and the D decayed from
the P went into interstellar gas, and the P continued to decay to D.
When the Earth formed by condensing form interstellar gas, it
incorporated some P and some D. P continued to decay to D.
OldD therefore consists of D that was created directly as D, D that
formed by decay of P before the Earth formed, and D that formed from
the decay of P after the Earth formed.
Richard Uhrich
We can see the debris of supernovae seen by the Chinese 1000 years ago,
or in Europe 400 years ago. And in mmy sophomore asttroonomy class we
calculated thhe life cycle of stars based on their mass, brightness and
chhemical composition. Supernovae are physics, understood and
predictable. Like the isochron method of dating.
--
Richard Uhrich
---
Ignorance more frequently begets confidence than does knowledge. --
Charles Darwin
Jon Fleming
wrote:
Zoe, m' dear, quantum mechanics tells us that _everything_ is based on
probability and statistics. When large numbers of atoms are involved
(as they are in any macroscopic solid), the predictions of probability
and statistics are very precise (there's very little possibility of
variation). It's possible for you to drop and break a coffee cup and
then watch in astonishment as the parts of the cup reassemble
themselves into a whole cup and jump back up into your hand; you would
be astonished if that happens because probability tells us that's very
unlikely to happen.
>My misconception was that in order for any mineral to form, it HAD to
>have a certain quantity of D and Di before it would be able to finish
>its formation and be recognized as a particular mineral. I had
>visions of the mineral waiting patiently until the right quantity of
>Di came along and then, zap, it closed up and became, say, zircon or
>biotite or whatever. Now I see that the concept of same ratios is
>based strictly on a concept of probability, not laws of the chemical
>construction of the particular mineral. In other words, a mineral will
>still be that mineral, whether or not it contained D or P; am I
>correct?
Yes, that's true for many minerals that are used in radiometric
dating.
>So tell me, then, will this probability fail to work if the melt were
>not thoroughly mixed?
Yes.
>What if the original elements remained in their
>original locations, melted, ejected and cooled -- with no thorough
>mixing of the slurry?
"Slurry" is the wrong word. It's a fact that the original elements do
_not_ remain in their original locations in the melt. There's a
certain amount of mechanical mixing, but most of the mixing is from
diffusion; the atoms are hot so they move around a lot.
So the answer is that the constant oldD/Di assumption would fail if
the original elements remained in their original locations, but it is
impossible to have both "molten" and "the elements remain in their
original locations".
>>When an atom moves from the melt to a solid mineral, it is either an
>>atom of that mineral's normal constituents or it is an atom of D or it
>>is an atom of Di. We don't care about the case where it is an atom of
>>the mineral's normal constituent. In the case where it is an atom of
>>D or an atom of Di, the probability that it is an atom of D is the
>>same as the number of atoms of D divided by the total number of atoms
>>of D and Di, and similarly for Di. (In your example, the probability
>>that it is an atom of D is 210/(100+210) = 210/310 = 67.74% and the
>>probability that it is an atom of Di is100/310 = 32.36%).
>>
>>As the mineral solidifies (incorporating trillions of trillions of
>>trillions of atoms in a real mineral) this "coin flip" of selecting an
>>atom of D or Di randomly is repeated trillions of trillions of
>>trillions of times. The ratio of D to Di in the mineral will wind up
>>in exactly (or indistinguishably close to) the same proportion as the
>>ratio was in the melt, just as if you toss a fair coin trillions of
>>trillions of trillions of times you will wind up with 50% heads and
>>50% tails.
>
>are you equating same ratios of D/Di to heads and non-same ratios of
>D/Di to tails?
No, I'm equating "grabbing" an atom of D to heads and "grabbing" an
atom of Di to tails. For a fair coin, this corresponds to a D/Di
ratio in the melt of 1 (half D and half Di). If the D/Di ratio is
something other than 1, then you'd have to equate it to a coin that is
biased one way or the other (or use some more complex mapping between
the D, Di, and patterns of heads and tails).
For example, if the D/Di ratio in the melt is 2 (66.6667% D and
33.3333% Di), and D is equated to heads and Di is equated to tails,
then you could a weird coin that had a 66.666667% chance of landing on
"heads" and a 33.33333% chance of landing "tails".
>Then that would say that with sufficient flips, you
>would have half your minerals that are in same ratios and the other
>half are not in same ratios.
Not at all. With a sufficient number of flips, you would have all
your minerals and the melt in the same ratios.
>And if this is the case, what is the probability that the samples you
>choose to test for your isochron are from that 50 percent that have
>same ratios versus those that don't?
Since this is not the case, the question is meaningless.
>Or, again, are you equating D to heads and Di to tails?
Yes.
>If so, then
>you would have half the minerals carrying D and the other half
>carrying Di,
Nope. All the minerals would carry both D and Di. The probability,
and the fact that the number of atoms of D and Di is large for any
sample we can see, would ensure that the ratio of D to Di in each
mineral would be the same as in the melt.
>and it wouldn't matter since D and Di are
>indistinguishable in the solidification process.
>
>Or are you telling me that probability and statistics will ensure that
>every mineral that forms will take up exactly the same proportion of D
>to Di, regardless of how well mixed or poorly mixed the melt is?
Not quite. The probability and statistics will ensure that every
mineral that forms will take up exactly the same proportion of D to Di
_provided_ that the melt is well mixed. The physics of hot molten
materials, and the reality of how molten rock is formed and moved,
ensure that the melt will be well mixed (over distances of a few tens
of feet or so, which is sufficient since a rock that solidifies will
only grab material form a volume about as large as the final rock).
>That is like saying, here we have two trillion atoms of D and one
>trillion atoms of Di. Since they are indistinguishable from each
>other, it is predicted that the first mineral that forms will happen
>to take up two atoms of D and one atom of Di, and the next mineral
>will do the same, and the third mineral will do the same, again and
>again and again, without missing a beat. That would be like tossing a
>coin a million times and each time it lands on heads, never tails.
Not quite. The numbers you are using are too small, and your "that
would be like ..." is incorrect. Heads is D, tails is Di. Tossing a
coin many times and getting only heads corresponds to a mineral
solidifying containing only D and no Di.
Here we have two trillion trillion trillion trillion atoms of D and
one trillion trillion trillion trillion atoms of Di. Since they are
indistinguishable from each other, it is predicted that the first
mineral that forms will happen to take up two trillion trillion
trillion atoms of D and one trillion trillion trillion atoms of Di.
That would be like tossing a fair coin three trillion trillion
trillion times and getting 1.5 trillion trillion trillion heads and
1.5 trillion trillion trillion tails. In reality, you probably
wouldn't get _exactly_ those numbers, you might get
1,500,000,000,000,000,000,000,000,123,456,789 heads (atoms of D) and
1,499,999,999,999,999,999,999,999,876,543,211 tails (atoms of Di), or
some number like that. Probability tells us that, in three trillion
flips of a fair coin, you are essentially certain of getting a ratio
of heads to tails of 1.0000 plus or minus a very very very small
number; similarly probability tells us that after trillions of
trillions of trillions of random selections of D or Di from a melt
containing D and Di in some proportion, the ratio of D to Di that you
wind up with is in the same proportion as in the melt plus or minus a
very very very small number.
>Wouldn't real probability work like this? A mineral takes up 4 atoms
>of D and 1 atom of Di in one mineral, and the next mineral, since it
>cannot distinguish, will take up 10 atoms of D and 5 atoms of Di, and
>eventually, with sufficient tries, you may even find a mineral taking
>up the quantity of D to Di in the same ratios as found in the melt.
Sort of, except the numbers you are working with are _way_ too small.
>And since this is not a law of the chemical makeup of the mineral,
>there is no reason for each mineral to take up exactly the same ratios
>of each isotope. But overall, you would probably find that all of D
>and Di has been absorbed by the solidifying minerals in the same ratio
>of 2:1 COLLECTIVELY for all the minerals. But individually, they
>would carry whatever quantity of oldD and Di happened to be close at
>hand before they solidified.
Sort of, except to talk realistically about the probabilities you have
to do it with realistic numbers. You're just making the numbers up
with no constraints. There are constraints. The numbers must be
incredibly large, because there is an incredibly large number of atoms
in the smallest sample, _and_ those numbers must be in about the
same ratio as the ratio in the melt, because probability guarantees us
that there's no need to worry about cases when the atoms in the solid
are not in about the same proportion as in the melt (just like there's
no need to worry about your house jumping off its foundation and
spinning around upside down and falling back onto its foundation).
For example, from a melt that contains a D/Di ratio of 1.0 (half D and
half Di), a mineral takes up 4,000,000,000,153,896,000,111 atoms of D
and 3,999,999,999,842,103,999,889 atoms of Di (ratio 1.000000000000) .
The next mineral takes up 10,000,000,000,000,762,301,432 atoms of D
and 9,999,999,999,999,237,698,568 atoms of Di (ratio 1.000000000000).
And so on.
>
>And this would be especially true if the remelt is not stirred and
>thoroughly mixed, but simply melted in place and ejected.
Nope. In molten rock, atoms move around quite freely without any
mechanical mixing. Of course, you never see "melted in place"
(there's always some movement and mechanical mixing) and "ejected"
involves mechanical mixing.
>Can you explain why this should not be the case?
>
>I have vague memories of taking statistics at the beginning of
>college, but have forgotten how it works, so consider the above a real
>layman's attempt to understand probability and statistics. :-\
>
>>So what happens to the melt? As long as atoms of D and Di are removed
>>from the melt (into minerals) in the same proportion as they are in
>>the melt, the ratio of D to Di in the melt will not change.
>
>agreed....that is, if your assumption is correct. But I think that
>assumption can be questioned. Are atoms of D and Di consistently
>removed from a melt in the same proportion as they are in the melt,
>from one mineral to the next?
Yes,
Nope, although in practice melts _are_ thoroughly stirred over the
distances that are important.
>In spite of probability and
>statistics, if the melt contains lots of D in one portion and sparse
>Di in that same section, or sparse D in another location but lots of
>Di over there, the minerals don't distinguish between them and simply
>absorbs whatever D or Di is within its reach. Overall, the ratio for
>all minerals would be 2:1, but for individual minerals, the ratios
>would be all over the map.
Yup. Except that doesn't happen. There's too much mechanical mixing
in real situations, and there's too much diffusional mixing due to the
temperature.
>What is the probability that the isochronists would happen to choose
>only those samples that happen to have collected in the same 2:1
>proportions as the entire melt?
Essentially zero. If it did happen, there is essentially a zero
probability that the sample points would lie on a straight line.
Overall, the chance that it will happen _and_ it will happen in
exactly the right way so as to line up the samples is zero.
>>>Would there be some oldD left unattached?
>>
>>There might be oldD left that is not incorporated into other minerals.
>>It depends on how easily D is taken up by the last few minerals that
>>solidify.
>>
>>>What happens to those atoms?
>>
>>They solidify into solid D, tiny grains of D and Di intermixed in the
>>ratio they were in the melt and in all other minerals that solidified
>>from that melt, nestled in between the other minerals.
>
>how can D be intermixed with Di in these last few minerals if all the
>D has been absorbed?
All the D has not been absorbed. The ratio of D to Di is the same in
the melt as it always was.
>
>snip>
>
>>>have supernovae been sampled for original D?
>>
>>Yes. Each element gives off a distinctive pattern of light when it is
>>hot and absorbs a distinctive pattern of light when it is cold. By
>>analyzing the content of light from supernovae, we can detect what
>>elements are there. References below.
>
>I've been looking at your references. Are you saying that they are
>able to detect the light patterns of all varieties of D, such as
>lead-206, 207, 208, strontium-87, argon-40, Neodymium-143, to name
>just a few? Fascinating.
Yup. Each isotope of each element has its own pattern. It is indeed
fascinating.
>>(Interesting side factoid: helium was first discovered when
>>scientists used this method on the Sun, and was only discovered on
>>Earth afterwards. The name "helium" comes from "helios", the Greek
>>for "Sun". The method works. See
>><http://antwrp.gsfc.nasa.gov/apod/ap010120.html>.)
>
>I can't seem to acess this one.
I can't either ... and several other NASA sites are not findable
either. They probably have a problem. Try
<http://www-solar.mcs.st-andrews.ac.uk/~clare/Lockyer/helium.html>,
and <http://www.xware.ru/db/msg/apod/2001-01-20>.
>
>snip>
>
>>>what evidence do you have that supernovae produce D outside of decay
>>>from P?
>>
>>Detection of D in supernovae.
>
>and how are the light patterns in the supernovae detected? Not
>through that electron beam, right?
Nope. <http://www.synapses.co.uk/astro/starstru.html> has a good
explanation and some pictures.
>>Theoretical physics, which predicts that D can only be produced by
>>means other than radioactive decay under conditions such as found in
>>supernovae, and further predicts that D _must_ be produced in such
>>conditions.
>
>well, now that is more understandable -- theoretical stuff.
>Predictions. But no hard evidence to support such predictions, right?
Wrong. We have hard evidence from our observations of stars and our
many other experiments that test the theory.
Derek Stevenson
[snip]
> >I'm not sure you understood Zoe's question. I suspect that by "minerals
of
> >the same type", she actually means "the same mineral", and by "an
> >equivalent amount of P", she actually means "the same proportion of P".
In
> >other words, she's asking whether all crystals of, say, shazamite should
> >contain the same ratio of P to all other elements.
> >And the answer to that, I think, should be "yes". By definition. All
> >crystals of shazamite should contain one atom of P,
>
> what element is P here?
As Jon points out in his reply, "shazamite" is a made-up mineral, so P can
be whatever element you want it to be.
[snip]
> >A journalist writing a profile of the team wants to figure out how long
> >we've been together. Since we went in for football, he knows that we
must
> >have started out with five kickers and four runners. He knows that we
know
> >have seven kickers and two runners. He knows that of the seven kickers,
six
> >are football-kickers and one is a soccer-kicker (because football
kickers
> >always get a tattoo of a football on their arms, while soccer-kickers
> >always get one of a soccer ball). He knows that in one year, two out of
> >every four runners who play football become football-kickers. Therefore,
> >since I had to have started out with five kickers, two of my seven
kickers
> >must be former runners and that the team has been together a year.
> >
> >Am I on the right track here? And Zoe, is this making any sense to you?
>
> well, I don't follow sports much so you kind of lost me along the way.
> How about a more feminine analogy?
I didn't think you had to know much about sports to follow the analogy --
heck, I'm probably less interested in sports than you. Where did you get
lost, and what was your difficulty?
But as for a "more feminine analogy", how about cooking? I've entered a
baking competition, and the recipe I prepare will depend on the mix of
ingredients I'm assigned -- I'll make one thing if I end up with four cups
of flour, two eggs and a cup of sugar, and another if I end up with six
eggs, a cup of chocolate and half a cup of flour.
As for an analogy for D/Di: you can buy a carton of buttermilk from the
store, or you can sour some regular milk with a little vinegar -- for
purposes of the analogy, we'll assume that it makes no difference to the
recipe which you use, but that the difference can be detected later.
How's that for a start?
> don't bother. I don't want you to have to struggle. (grin)
>
> >(If
> >you can't figure out how the elements of this analogy correspond to
> >isochron dating, just ask. Note that this analogy applies to figuring
out
> >the history of a *single sample*, not to the multiple samples required
to
> >plot an isochron.)
>
> okay. Maybe you can explain the history of oldD from its creation?
Jon's done that, but what does it have to do with the analogy?
Derek Stevenson
> >So tell me, then, will this probability fail to work if the melt were
> >not thoroughly mixed?
>
> Yes.
>
> >What if the original elements remained in their
> >original locations, melted, ejected and cooled -- with no thorough
> >mixing of the slurry?
>
> "Slurry" is the wrong word. It's a fact that the original elements do
> _not_ remain in their original locations in the melt. There's a
> certain amount of mechanical mixing, but most of the mixing is from
> diffusion; the atoms are hot so they move around a lot.
>
> So the answer is that the constant oldD/Di assumption would fail if
> the original elements remained in their original locations, but it is
> impossible to have both "molten" and "the elements remain in their
> original locations".
Here's a little experiment you can try at home, Zoe.
Make some nice hot coffee, and pour a cup. Add some milk. *Don't* stir --
just let it sit for half an hour or so. Now come back to it and have a
look.
In the absence of mixing, did the milk and the black coffee remain in their
"original locations"?
[snip]
Derek Stevenson
No. I (and others) have speculated numerous times about why it is that
creationists think the way that they do. We're often better at explaining
the usual creationist arguments than the creationists themselves, while the
reverse is seldom true. The t.o. archive is, as I've noted before, probably
the 'net's most comprehensive repository of creationist arguments.
We *do* follow our own advice.
Derek Stevenson
> >This is your fault too, Zoe. As much as you might claim to be a simple,
> >unsophisticated country girl,
>
> you mean, my claiming to be a nonscientist makes me into an
> unsophisticated country girl? Hmmm...I see that the world is divided
> into sophisticated city-slicker scientists and unsophisticated country
> folk nonscientists -- at least in your mind.
No, I mean that you try to portray yourself as a humble, uninformed honest
seeker of information, eager to fill the gaps in her knowledge and ready to
fully and fairly consider new ideas. I'm not saying it's an act -- I'm sure
that's how you see yourself, and to some extent it's even true.
But you're not as humble as you think. You have some hard-wired notions
that no evidence is ever going to shake -- or so it appears from out here,
at any rate -- and some other ideas that you've latched on to and decided
are correct, without ever really examining them closely.
> >you have a degree of arrogance and
> >self-satisfaction that's sadly misplaced in this sort of discussion.
>
> yes, sir, yes, sir, whatever you say, sir...
>
> better?
No.
Try thinking about *why* someone might say this about you -- or why "Zoe"
has become a byword for arrogant ignorance in this group. (Do a Google
search for "Zoe" sometime, and see how your name is evoked in the threads
you're not a participant in.)
> >You
> >seem to assume right off the bat that whatever you *think* you know
about a
> >subject is:
> >a) accurate and
> >b) so detailed as to constitute a sufficient understanding.
> >This is rarely the case, and it almost always requires a monumental
> >struggle to get you to see it.
>
> don't struggle, Derek, just relax. Apparently, Jon does not have to
> struggle. He just smoothly rolls out his information and doesn't even
> break a sweat.
Just how long has this isochron discussion been going on, again?
> > You're like the would-be flying sheep in the
> >Monty Python episode, of whom the farmer says, "Once they gets an oidear
in
> >their 'eads, there's no shiftin' it."
>
> hmmm, I don't see YOUR ideas shifting either, Derek.
No, of course you don't see it. How would you recognize it if you did?
> >You should rely a little less on your own sense of satisfaction with
your
> >comprehension of an idea (I've never yet seen you announce your
> >satisfaction with an idea that was *right*), and put a little more trust
in
> >the consensus of the group as to whether you've understood. Ask more
> >questions,even if they seem very simple and basic. *Especially* if they
> >seem very simple and basic, because this is where you tend to go wrong.
> >
> >Above all, don't -- as I keep trying to remind you -- assume that just
> >because an idea makes you feel good, that it's right. Or vice versa.
>
> One of my mottos: Live by principle, not by feeling, for the most
> part.
You have to do more than *say* it. You have to *live* it. And you don't.
Look again at how many discussions you're tried to set aside with arguments
like "well, it's an explanation that *I* find satisfactory".
Derek Stevenson
news:ajiqln$m3m$1...@slb3.atl.mindspring.net...
> > looked it up, and it seems to say that a substance is bombarded with
> > an electron beam in order to determine its characteristics. How do
> > you get your electron beam out there in those far-flung galaxies?
>
> We are all uneducated in everything at some point in our lives, and we
are
> all uneducated in some things at every point in our lives.
>
> What's exceptional about Zoe is her lack of self-consciousness about this
> common denominator of human kind (or her ability to ignore what
> self-consciousness is there).
>
> It's admirable, IMO.
Indeed it is, and Zoe deserves credit for the many instances in which she
has asked for more information, or gone to the effort of researching it for
herself. Would that more creationists were like her.
I only wish that she were more consistent about this. While she *will* seek
out more information on a topic about which she is uninformed, and *will*
think carefully about it, the process often comes to a halt when she thinks
she understands, or when she's come to a conclusion that she likes. Once
she's reached this point, it's very difficult to persuade her that there is
additional information that she hasn't considered, or that some of the
"information" she *has* taken into account is mistaken.
> > how do you determine what conditions are likely, having never been in
> > a position to personally check the stars or experienced supernovae?
> > How is the extrapolation done?
> >
> > maybe you shouldn't bother to answer that since I'm sure there is a
> > large body of work that explains it, and I should read up on it
> > myself.
If this is a sincere response, and not a sarcastic regurgitation of a
common response to her posts, then kudos are due her once again. With the
caveat that it *is* a large body of work, and that any conclusions she
reaches on the basis of a quick survey should be regarded as highly
tentative.
[snip]
John Harshman
<jo...@fleming-nospam.com> wrote:
It might be good to explain why they look totally different. A supernova
lasts a few weeks at most. Sometimes, you can look at photos made before
it appeared and see the star as it was before the supernova. After it's
over, there's a clearly visible, expanding gas cloud. On the other hand,
the motion of distant stars and galaxies, if enough to explain detectable
differences in light reaching us, would be on the scale of millions of
years. Think about it. If a galaxy is, say, 100 million light years from
us, then in order to get 1% farther away (and for its light to get about
2% fainter for that reason) it would have to move a million light years.
And galaxies don't travel at anything near the speed of light, either.
--
*Note the obvious spam-defeating modification
to my address if you reply by email.
zoe_althrop
wrote:
snip>
I've looked at all your sites, June. Thank you. Absolutely
fascinating stuff. I see now that those "explosions" could not be
passing galaxies.
Of course, you might as well know that my creationist perspective is
already interpreting these supernova differently. Are we seeing
(finally) works of creation that occurred way back in time? The
movements of a Creator?
snip>
----
zoe
zoe_althrop
<derekste...@yahoo.com> wrote:
snip>
>But as for a "more feminine analogy", how about cooking? I've entered a
>baking competition, and the recipe I prepare will depend on the mix of
>ingredients I'm assigned -- I'll make one thing if I end up with four cups
>of flour, two eggs and a cup of sugar, and another if I end up with six
>eggs, a cup of chocolate and half a cup of flour.
>
>As for an analogy for D/Di: you can buy a carton of buttermilk from the
>store, or you can sour some regular milk with a little vinegar -- for
>purposes of the analogy, we'll assume that it makes no difference to the
>recipe which you use, but that the difference can be detected later.
>
>How's that for a start?
well, now we're cooking.:-)
except that I don't think your heart's in this feminine analogy,
Derek. Your above description is brief, disinterested, and not
explaining much at all. I think you were doing much better with the
sports analogy. I'll just have to go back to that one and pay more
attention if I want to work with you on this.
----
zoe
zoe_althrop
except that volcanic eruptions are not of the consistency of milk and
black coffee. Apart from steam and water, the part of the eruption
that forms rock is thick and sludgey. Try melting a bar of dark
chocolate and a bar of white chocolate together. Do not stir, and the
colors remain fairly distinct. You should get a mixture of black and
white. Well, at least that is what I think should happen. Maybe I'll
go do that experiment next.
And Jon's post above insists that much mechanical mixing goes on.
What is the cause of this mechanical mixing? Doesn't the molten stuff
just sit quietly there, under high temperature, below the crust -- or
is it the mantle?
Which brings up another question. Shouldn't there be 4.5 billion
years worth of oldD down under? And when we talk about remelts, what
exactly gets remelted? Not the volcanic stuff that's under the
mantle, right? Does remelting refer to rock on the surface that gets
softened again under a new volcanic eruption? Is that what partial
remelting does, cause partial crystallization? Jon?
----
zoe
zoe_althrop
harshman....@sjm.infi.net (John Harshman) wrote:
snip>
>It might be good to explain why they look totally different. A supernova
>lasts a few weeks at most. Sometimes, you can look at photos made before
>it appeared and see the star as it was before the supernova. After it's
>over, there's a clearly visible, expanding gas cloud.
interesting -- no, more than that. This stuff is exciting! I want to
go out there and watch what's happening, front-row seat, please.
>On the other hand,
>the motion of distant stars and galaxies, if enough to explain detectable
>differences in light reaching us, would be on the scale of millions of
>years. Think about it. If a galaxy is, say, 100 million light years from
>us, then in order to get 1% farther away (and for its light to get about
>2% fainter for that reason) it would have to move a million light years.
>And galaxies don't travel at anything near the speed of light, either.
I suppose you're right. How intriguing.
----
zoe
Richard Uhrich
> On Fri, 16 Aug 2002 07:24:01 +0000 (UTC), jun...@adelphia.net (June)
> wrote:
>
> snip>
>
> I've looked at all your sites, June. Thank you. Absolutely
> fascinating stuff. I see now that those "explosions" could not be
> passing galaxies.
>
> Of course, you might as well know that my creationist perspective is
> already interpreting these supernova differently. Are we seeing
> (finally) works of creation that occurred way back in time? The
> movements of a Creator?
A star was blown to smithereens, a phenomenon predicted by astronomers.
How was this a work of a creation? Do you suggest this is how this
creator must make heavy elements?
>
> snip>
>
> ----
> zoe
Jon Fleming
wrote:
Not when it's underground before the eruption, and not so much when it
actually erupts. I've _seen_ Kiluea. I've _seen_ lots of film
footage of volcanic eruptions. None of them can be described as
"thick and sludgy".
>Try melting a bar of dark
>chocolate and a bar of white chocolate together. Do not stir, and the
>colors remain fairly distinct. You should get a mixture of black and
>white. Well, at least that is what I think should happen. Maybe I'll
>go do that experiment next.
>
>And Jon's post above insists that much mechanical mixing goes on.
>What is the cause of this mechanical mixing? Doesn't the molten stuff
>just sit quietly there, under high temperature, below the crust -- or
>is it the mantle?
Nope. It originates miles below the surface and is extruded, under
tremendous pressure, through small cracks and channels, mixing as it
goes.
>Which brings up another question. Shouldn't there be 4.5 billion
>years worth of oldD down under?
There's more than that ... supernovae, remember.
>And when we talk about remelts, what
>exactly gets remelted? Not the volcanic stuff that's under the
>mantle, right?
Wrong. Remelting happens to the stuff that's under the mantle. It
gets down there by subduction as the continents move around.
> Does remelting refer to rock on the surface that gets
>softened again under a new volcanic eruption?
That is one possibility for a partial remelt.
>Is that what partial
>remelting does, cause partial crystallization?
Not so much partial crystallization as partial redistribution of the
contents, disturbing the oldD/Di ratio (because Di moves around).
zoe_althrop
<uhr...@san.rr.com> wrote:
snip>
>A star was blown to smithereens, a phenomenon predicted by astronomers.
>How was this a work of a creation? Do you suggest this is how this
>creator must make heavy elements?
a star was created out of smithereens. There's a lot we don't know
yet.
snip>
----
zoe
zoe_althrop
<jo...@fleming-nospam.com> wrote:
>>On Fri, 16 Aug 2002 14:20:32 +0000 (UTC), "Derek Stevenson"
>><derekste...@yahoo.com> wrote:
snip>
>>>Here's a little experiment you can try at home, Zoe.
>>>
>>>Make some nice hot coffee, and pour a cup. Add some milk. *Don't* stir --
>>>just let it sit for half an hour or so. Now come back to it and have a
>>>look.
>>>
>>>In the absence of mixing, did the milk and the black coffee remain in their
>>>"original locations"?
>>
>>except that volcanic eruptions are not of the consistency of milk and
>>black coffee. Apart from steam and water, the part of the eruption
>>that forms rock is thick and sludgey.
>
>Not when it's underground before the eruption, and not so much when it
>actually erupts. I've _seen_ Kiluea.
me, too.
>I've _seen_ lots of film
>footage of volcanic eruptions. None of them can be described as
>"thick and sludgy".
it sure looks thick and sludgy to me...like hot fudge slowly
advancing.
>>Try melting a bar of dark
>>chocolate and a bar of white chocolate together. Do not stir, and the
>>colors remain fairly distinct. You should get a mixture of black and
>>white. Well, at least that is what I think should happen. Maybe I'll
>>go do that experiment next.
>>
>>And Jon's post above insists that much mechanical mixing goes on.
>>What is the cause of this mechanical mixing? Doesn't the molten stuff
>>just sit quietly there, under high temperature, below the crust -- or
>>is it the mantle?
>
>Nope. It originates miles below the surface and is extruded, under
>tremendous pressure, through small cracks and channels, mixing as it
>goes.
what causes the mixing? Tornadic activity? Squeezing and pumping
activity, like a stomach? What would cause this mixing, Jon?
>>Which brings up another question. Shouldn't there be 4.5 billion
>>years worth of oldD down under?
>
>There's more than that ... supernovae, remember.
yes. So any never-remelted, genuinely old rock should register at
least 15 billion years, not 4.5 billion years, right?
>>And when we talk about remelts, what
>>exactly gets remelted? Not the volcanic stuff that's under the
>>mantle, right?
>
>Wrong. Remelting happens to the stuff that's under the mantle. It
>gets down there by subduction as the continents move around.
>
>> Does remelting refer to rock on the surface that gets
>>softened again under a new volcanic eruption?
>
>That is one possibility for a partial remelt.
>
>>Is that what partial
>>remelting does, cause partial crystallization?
>
>Not so much partial crystallization as partial redistribution of the
>contents, disturbing the oldD/Di ratio (because Di moves around).
so how do you determine when a rock is partially melted?
actually, consider that a rhetorical question, since I think that
you've had enough of me and the isochron.
----
zoe
Richard Uhrich
Yes, all were. That proves gravity. A star going supernova proves
nuclear physics. So again I ask how a supernova explosion is evidence of
creation.
June
> On Sun, 18 Aug 2002 00:24:40 +0000 (UTC), Jon Fleming
> <jo...@fleming-nospam.com> wrote:
>
> >>On Fri, 16 Aug 2002 14:20:32 +0000 (UTC), "Derek Stevenson"
> >><derekste...@yahoo.com> wrote:
>
> snip>
>
> >>>Here's a little experiment you can try at home, Zoe.
> >>>
> >>>Make some nice hot coffee, and pour a cup. Add some milk. *Don't* stir --
> >>>just let it sit for half an hour or so. Now come back to it and have a
> >>>look.
> >>>
> >>>In the absence of mixing, did the milk and the black coffee remain in their
> >>>"original locations"?
> >>
> >>except that volcanic eruptions are not of the consistency of milk and
> >>black coffee. Apart from steam and water, the part of the eruption
> >>that forms rock is thick and sludgey.
> >
> >Not when it's underground before the eruption, and not so much when it
> >actually erupts. I've _seen_ Kiluea.
>
> me, too.
[sigh] not me yet, I think/hope we're going in March for our wedding
anniversary, specifically to see Kiluea!
OT: We went to Oregon for vacation a couple of weeks ago and spent a day
at Mt. St. Helens. Definitely worth the price of admission if you ever
get up there. Went to Crater Lake, too, but all the smoke from forest
fires obscured the view, bummer.
>
> >I've _seen_ lots of film
> >footage of volcanic eruptions. None of them can be described as
> >"thick and sludgy".
>
> it sure looks thick and sludgy to me...like hot fudge slowly
> advancing.
I think the viscosity has a lot to do with what part of the lava flow
you're observing. A leading edge that is cooling rapidly can look pretty
sludgey, but I know I've seen pictures of it flowing fairly rapidly
under some circumstances. FWIU, though, the mixing and diffusion of
molecules would have been done before the lava ever reaches the surface
anyway. It is even less viscous while in the magma chamber, I would
think (no cooling yet).
Here's a site I found that lists the viscosity of lavas as compared to
everyday liquids. The info appears to be provide by the U of Hawaii,
Institute of Geophysics and Planetology, so I guess they know what
they're talking about.
http://www.spacegrant.hawaii.edu/class_acts/ViscosityTe.html
From the table at the bottom of the page, it looks like pahoehoe lava
(the kind that Kiluea spills) is 100 to 1,000 times more viscous than
motor oil (Motor oil is listed at 1 Pascal second at room temp and the
lava at 100 to 1,000 Pascal seconds with no temps given). I'm not sure
how that compares to hot fudge, but I found a reference for honey at 20º
C at around 200 poise or 20 Pascal seconds.
>
> >>Try melting a bar of dark
> >>chocolate and a bar of white chocolate together. Do not stir, and the
> >>colors remain fairly distinct. You should get a mixture of black and
> >>white. Well, at least that is what I think should happen. Maybe I'll
> >>go do that experiment next.
> >>
> >>And Jon's post above insists that much mechanical mixing goes on.
> >>What is the cause of this mechanical mixing? Doesn't the molten stuff
> >>just sit quietly there, under high temperature, below the crust -- or
> >>is it the mantle?
> >
> >Nope. It originates miles below the surface and is extruded, under
> >tremendous pressure, through small cracks and channels, mixing as it
> >goes.
>
> what causes the mixing? Tornadic activity? Squeezing and pumping
> activity, like a stomach? What would cause this mixing, Jon?
Convection currents; being squeezed through holes, tunnels, chambers in
the mantle on the way to the surface; expansion and reduction of
viscosity with reduced pressure as the magma rises; etc.
>
> >>Which brings up another question. Shouldn't there be 4.5 billion
> >>years worth of oldD down under?
> >
> >There's more than that ... supernovae, remember.
>
> yes. So any never-remelted, genuinely old rock should register at
> least 15 billion years, not 4.5 billion years, right?
[blink, blink] Uh, no, Zoe, about the only element that existed 15 bya
was hydrogen. There wasn't anything to make rocks out of at that time.
It wasn't until several billions of years worth of fusion in stars that
there was enough heavier elements to make any rocks anywhere in the
universe.
Furthermore, this planet did not exist until a bit less than 5 bya, so
no Earth rocks could possibly be older than that.
Finally, there can't be 4.5 billion years worth of 'oldD down under'
because, as Jon explained somewhere else, P, D and Di were all created
by the supernova(s) that produced the gas cloud our sun and solar system
condensed out of. We have no way of knowing how old that cloud was and
how much of these elements were already in it and how much P decayed to
D before the solar system formed, cooled and Earth rocks were available
to start an isochron clock ticking.
So there is _NO WAY_ to get any idea of the age of the Earth just from
the amount of D found in any rocks or magma.
>
> >>And when we talk about remelts, what
> >>exactly gets remelted? Not the volcanic stuff that's under the
> >>mantle, right?
> >
> >Wrong. Remelting happens to the stuff that's under the mantle. It
> >gets down there by subduction as the continents move around.
> >
> >> Does remelting refer to rock on the surface that gets
> >>softened again under a new volcanic eruption?
> >
> >That is one possibility for a partial remelt.
> >
> >>Is that what partial
> >>remelting does, cause partial crystallization?
> >
> >Not so much partial crystallization as partial redistribution of the
> >contents, disturbing the oldD/Di ratio (because Di moves around).
>
> so how do you determine when a rock is partially melted?
Depends on the rock and what you mean by partial melt. My understanding
is that partial melt refers to deeply buried rocks that are heated to
the point where only some of their crystals melt. The liquid from these
partial melts rise and become magma, the rest of the rock - less the
melted crystals - stays where it is. Geologists don't generally see
these rocks.
A partially *remelted* rock would be a solid rock that is exposed to
heat but not enough to become completely molten. This would be a
metamorphic rock, I think. Does that make sense?
Here are some links to info on magma, lava, volcanoes and igneous rocks.
Maybe you can get some of your answers at one of these.
http://www.tulane.edu/~sanelson/geol204/volcan&magma.htm
http://geoweb.tamu.edu/Faculty/Wiltschko/Geology104/assets/files/06b.%20
Origin%20of%20Magma.htm
http://geoweb.tamu.edu/courses/geol101/grossman/Ign.rocks.html
http://www.science.ubc.ca/~geol202/rock_cycle/rockcycle.html
http://www.uoregon.edu/~dogsci/dorsey/geo101/lect6.html
>
> actually, consider that a rhetorical question, since I think that
> you've had enough of me and the isochron.
>
> ----
> zoe
Robin Levett
> On Fri, 9 Aug 2002 01:52:10 +0000 (UTC), "Robin Levett"
> <rle...@ibmrlevett.uklinux.net> wrote:
>
> snip>
>
> >zoe; did it ever occur to you to harmonise your worldview with the
> >data?
>
> isn't that what I'm doing right now? Harmonizing my worldview with
> the data? If I come to see sufficient discrepancies in my
worldview,
> then it would have to change. But one or two unanswered questions
> coming from the field of science does not yet begin to undermine a
> worldview which is built on a much wider foundation than just
science.
>
I've just got back from a week in the Algarve seeing stars, so haven't
caught up with the 5,000-odd posts since I left; this may have been
dealt with elsehere - if so, I apologise.
My point was that you said you were seeking to reconcile the facts
with your worldview; which is quite definitely not the same process as
reconciling your worldview with the facts.
More generally, this is why there is no equivalence between your
worldview and a scientist's. A scientist's worldview is a synthesis
of the facts, and so is naturally adjusted by advances in knowledge.
A creationist's worldview is fixed, set by the particular
interpretation they adopt of whichever holy book the subscribe to, and
changes in that worldview are only accomplished by recognition of
unchallengeable facts that unequivocally challenge it. You have
recognised (and it does you credit, however long it has taken (NB that
is not intended to be patronising)) the validity of the isochron
dating method, and therefore, one hopes, of "old" fossiliferous rock.
What we wait to see is whether that forces a change in your worldview.
--
________________________________________________________________
Robin Levett
rle...@ibmrlevett.uklinux.net
(address munged by addition of Big Blue)
Atheist = knows of and uses Occam's Razor
Agnostic = knows of but isn't sure whether to use Occam's Razor
Fundy = what's Ockam's erasure?
___________________________________________________
Jon Fleming
wrote:
>On Sun, 18 Aug 2002 00:24:40 +0000 (UTC), Jon Fleming
><jo...@fleming-nospam.com> wrote:
>
>>>On Fri, 16 Aug 2002 14:20:32 +0000 (UTC), "Derek Stevenson"
>>><derekste...@yahoo.com> wrote:
>
>snip>
>
>>>>Here's a little experiment you can try at home, Zoe.
>>>>
>>>>Make some nice hot coffee, and pour a cup. Add some milk. *Don't* stir --
>>>>just let it sit for half an hour or so. Now come back to it and have a
>>>>look.
>>>>
>>>>In the absence of mixing, did the milk and the black coffee remain in their
>>>>"original locations"?
>>>
>>>except that volcanic eruptions are not of the consistency of milk and
>>>black coffee. Apart from steam and water, the part of the eruption
>>>that forms rock is thick and sludgey.
>>
>>Not when it's underground before the eruption, and not so much when it
>>actually erupts. I've _seen_ Kiluea.
>
>me, too.
>
>>I've _seen_ lots of film
>>footage of volcanic eruptions. None of them can be described as
>>"thick and sludgy".
>
>it sure looks thick and sludgy to me...like hot fudge slowly
>advancing.
Only at the very tip, where it's cooled enough to be almost solid
>>>Try melting a bar of dark
>>>chocolate and a bar of white chocolate together. Do not stir, and the
>>>colors remain fairly distinct. You should get a mixture of black and
>>>white. Well, at least that is what I think should happen. Maybe I'll
>>>go do that experiment next.
>>>
>>>And Jon's post above insists that much mechanical mixing goes on.
>>>What is the cause of this mechanical mixing? Doesn't the molten stuff
>>>just sit quietly there, under high temperature, below the crust -- or
>>>is it the mantle?
>>
>>Nope. It originates miles below the surface and is extruded, under
>>tremendous pressure, through small cracks and channels, mixing as it
>>goes.
>
>what causes the mixing? Tornadic activity? Squeezing and pumping
>activity, like a stomach?
being forced through small cracks and channels, and the shearing
action that results. Got a food mill? Heat your dark-white chocolate
mixture to be liquid and force it through your food mill a few tens
of times.
>What would cause this mixing, Jon?
Shearing.
>>>Which brings up another question. Shouldn't there be 4.5 billion
>>>years worth of oldD down under?
>>
>>There's more than that ... supernovae, remember.
>
>yes. So any never-remelted, genuinely old rock should register at
>least 15 billion years, not 4.5 billion years, right?
Nope. Any never-remelted rock should register the age since it
formed ... 4.5 billion years.
>>>And when we talk about remelts, what
>>>exactly gets remelted? Not the volcanic stuff that's under the
>>>mantle, right?
>>
>>Wrong. Remelting happens to the stuff that's under the mantle. It
>>gets down there by subduction as the continents move around.
>>
>>> Does remelting refer to rock on the surface that gets
>>>softened again under a new volcanic eruption?
>>
>>That is one possibility for a partial remelt.
>>
>>>Is that what partial
>>>remelting does, cause partial crystallization?
>>
>>Not so much partial crystallization as partial redistribution of the
>>contents, disturbing the oldD/Di ratio (because Di moves around).
>
>so how do you determine when a rock is partially melted?
Disorientation between the crystals; the crystals that formed after
the partial remelt will be differently oriented than the crystals that
formed in the original solidification.
Different compositions: the crystals that form after a partial remelt
often have different amounts of trace elements incorporated (such as
P, D, and Di).
For example, if you do an isochron analysis of rock that has partially
remelted and you take a lot of samples, you will often see _two_ lines
of data points on the graph. One line of data points will be made up
of crystals that didn't remelt, and the other line of data points will
be made up of crystals that didn't remelt.
>actually, consider that a rhetorical question, since I think that
>you've had enough of me and the isochron.
Well, I've certainly had enough of your posting untrue and untested
claims as if they were fact.
sds
"Robin Levett" <rle...@ibmrlevett.uklinux.net> wrote in message
news:QCM79.341$un6....@newsfep1-gui.server.ntli.net...
Sorry to interrupt here, but I'm amazed that any thinking person would
assume that one who believes:
A supernatural God poofed everything in to existence
There is a spirit that works in us
Jesus was a man and God simultaneously, and his dead body was resurrected
Various outright miraculous showmanships actually occurred (as per the New
Testament)
would change their mind about any of this because science shows a rock to be
older than a few thousand years!
What an underestimate of the power of faith you are making. And note, I'm
not arguing the sensibility of faith - just it's historically documented
power over human action.
BTW, Hi again, Robin.
David Jensen
"sds" <s...@mp3.com> wrote in <ajo6um$24l$1...@slb6.atl.mindspring.net>:
>Sorry to interrupt here, but I'm amazed that any thinking person would
>assume that one who believes:
>
>A supernatural God poofed everything in to existence
>There is a spirit that works in us
>Jesus was a man and God simultaneously, and his dead body was resurrected
>Various outright miraculous showmanships actually occurred (as per the New
>Testament)
>
>would change their mind about any of this because science shows a rock to be
>older than a few thousand years!
Yet, that is the claim that fundamentalists make time and again. Oh, I'm
sorry, you said 'thinking person'.
>What an underestimate of the power of faith you are making. And note, I'm
>not arguing the sensibility of faith - just it's historically documented
>power over human action.
I was under the impression that faith was a human action.
June
> On Fri, 16 Aug 2002 07:24:01 +0000 (UTC), jun...@adelphia.net (June)
> wrote:
>
> snip>
>
[snip]
>
> I've looked at all your sites, June. Thank you. Absolutely
> fascinating stuff. I see now that those "explosions" could not be
> passing galaxies.
>
> Of course, you might as well know that my creationist perspective is
> already interpreting these supernova differently. Are we seeing
> (finally) works of creation that occurred way back in time? The
> movements of a Creator?
>
> snip>
>
> ----
> zoe
Your 'perspective' no longer surprises me. But maybe this time you
should find out the facts before you jump to a lot of erroneous
conclusions.
Rodjk
> On Tue, 6 Aug 2002 12:00:25 +0000 (UTC), Bjoern Feuerbacher
> <feuerba...@thphys.uni-heidelberg.de> wrote:
>
> snip>
>
> zoe wrote:
>
> >> I see that as long as Di is in the picture, you can indeed determine
> >> oldD/Di and newD/Di, which means you can determine time since
> >> solidification.
> >
> >O.k. Does that mean that you admit that the geologic columns is hundred
> >of millions of years old?
>
> I admit that the earth's material gives evidence of being ancient.
> The arrangement of the "geological column" is something I have yet to
> study into.
>
>
> >> Just keep your fingers crossed that I don't start thinking again in
> >> the wee hours of sleeptime. :-p
> >
> >Thinking about what? Why the method could give wrong ages anyway?
>
> why the old ages do not jibe with the rest of my worldview.
Heck, that is the easy one. It's because your worldview is wrong.
Another challenge, Zoe. Use your new found knowledge to convince other
creationist that their criticism of geologist dating methods are
wrong.
Lets see how long it takes before they are calling you a heretic and
an athiest.
Rodjk
>
> ----
> zoe
Rodjk
Yes, it is unfortunate that fairy tales and wishful thinking win out
over reason much of the time.
Rodjk #613
Chris Ho-Stuart
[snip]
> Which brings up another question. Shouldn't there be 4.5 billion
> exactly gets remelted? Not the volcanic stuff that's under the
> softened again under a new volcanic eruption? Is that what partial
> remelting does, cause partial crystallization? Jon?
If I can respond here as well. It -does- -not- -matter- how
many years worth of oldD is used to make the rock.
When you mix up melted material, no matter how old that material
might be, you get a flat isochon; which means zero age. That is
why you CANNOT get the "age or the Earth" from isochrons. Saying
"fuzzy" does not help; there is simply no basis at all for
any connection whatsoever with the age of the Earth.
We were dealing with this months ago; when we asked you want
happens when you melt and resolidify rock in a laboratory. The
answer is; you get a flat isochron; and the ages of the Earth,
or the age of the material from which the rock is formed,
make no difference whatsoever.
It is like a stop watch. The isochron is the watch hand. The
position of the watch hand tells you how long since the
reset. It has no relation at all to the age of the watch.
Isochron dating is not dating the material itself. It is dating
the rock; how long it is since the material was all mixed up
in a melt. If it was not mixed up in a melt; then you generally
do not get any isochron at all; the measurements will not fall
along a line.
Chris
Andrew Arensburger
> On Sat, 17 Aug 2002 21:33:29 +0000 (UTC), muz...@aol.com (zoe_althrop)
> wrote:
>>except that volcanic eruptions are not of the consistency of milk and
>>black coffee. Apart from steam and water, the part of the eruption
>>that forms rock is thick and sludgey.
> Not when it's underground before the eruption, and not so much when it
> actually erupts. I've _seen_ Kiluea. I've _seen_ lots of film
> footage of volcanic eruptions. None of them can be described as
> "thick and sludgy".
If I may change the subject for a second, could someone give
me some idea of how quickly the magma under the earth's crust moves?
I have a mental image of the Earth's interior looking like a
pot of boiling oil with convection cells rolling over like crazy, but
since there are entire continents lying on top of it, I suspect that
my mental image is naive. Presumably the magma moves at around the
same speed as continental drift?
--
Andrew Arensburger, Systems guy University of Maryland
arensb.no-...@glue.umd.edu Office of Information Technology
We're doomed.
sds
"Chris Ho-Stuart" <host...@sky.fit.qut.edu.au> wrote in message
news:3d60...@news.qut.edu.au...
> zoe_althrop <muz...@aol.com> wrote:
> [snip]
> > Which brings up another question. Shouldn't there be 4.5 billion
> > years worth of oldD down under? And when we talk about remelts, what
> > exactly gets remelted? Not the volcanic stuff that's under the
> > mantle, right? Does remelting refer to rock on the surface that gets
> > softened again under a new volcanic eruption? Is that what partial
> > remelting does, cause partial crystallization? Jon?
>
> If I can respond here as well. It -does- -not- -matter- how
> many years worth of oldD is used to make the rock.
>
> When you mix up melted material, no matter how old that material
> might be, you get a flat isochon; which means zero age. That is
> why you CANNOT get the "age or the Earth" from isochrons. Saying
> "fuzzy" does not help; there is simply no basis at all for
> any connection whatsoever with the age of the Earth.
>
> We were dealing with this months ago; when we asked you want
> happens when you melt and resolidify rock in a laboratory.
Pardon my ignorance, but has this been done? I mean melting and
resolidifying these sorts of rocks in the lab, and checking the D/Di ratio
to see if it is, indeed, redistributed homogeneously in all the minerals?
I'm sure is has, but with one word, you can save me some time.
I understand "diffusion" in gases and liquids, but melted rock? I'm just
not used to thinking of liquid "rock". How "liquid" does it have to be (and
for how long) to reset the isochron slope to zero during a re-melt. And
more importantly, if it isn't melted "enough", will the samples subsequently
show nonlinearity , thus indicating an inappropriate specimen? Seems they
would have to. That's one of the strong points of this graphical,
multi-sample method, I think. It's usually relatively easy to detect poor
specimens.
Anybody know how long "isochrons" have been used? I'll bet the same thing
was done well before the term was coined.
Chris Ho-Stuart
> "Chris Ho-Stuart" <host...@sky.fit.qut.edu.au> wrote in message
> news:3d60...@news.qut.edu.au...
>> zoe_althrop <muz...@aol.com> wrote:
>> [snip]
>> > Which brings up another question. Shouldn't there be 4.5 billion
>> > years worth of oldD down under? And when we talk about remelts, what
>> > exactly gets remelted? Not the volcanic stuff that's under the
>> > mantle, right? Does remelting refer to rock on the surface that gets
>> > softened again under a new volcanic eruption? Is that what partial
>> > remelting does, cause partial crystallization? Jon?
>>
>> If I can respond here as well. It -does- -not- -matter- how
>> many years worth of oldD is used to make the rock.
>>
>> When you mix up melted material, no matter how old that material
>> might be, you get a flat isochon; which means zero age. That is
>> why you CANNOT get the "age or the Earth" from isochrons. Saying
>> "fuzzy" does not help; there is simply no basis at all for
>> any connection whatsoever with the age of the Earth.
>>
>> We were dealing with this months ago; when we asked you want
>> happens when you melt and resolidify rock in a laboratory.
>
> Pardon my ignorance, but has this been done? I mean melting and
> resolidifying these sorts of rocks in the lab, and checking the D/Di ratio
> to see if it is, indeed, redistributed homogeneously in all the minerals?
>
> I'm sure is has, but with one word, you can save me some time.
Frankly, no idea. But as an experiment, is something parallel
with wondering what would happen if you dropped a feather and
a hammer on the moon. Some Apollo astronauts did just that; but
it was not for the sake of curiosity about the outcome. Same
with the melted rock. The result is not in any doubt.
> I understand "diffusion" in gases and liquids, but
> melted rock? I'm just not used to thinking of liquid
> "rock". How "liquid" does it have to be (and for how
> long) to reset the isochron slope to zero during a re-melt.
> And more importantly, if it isn't melted "enough", will the
> samples subsequently show nonlinearity , thus indicating an
> inappropriate specimen? Seems they would have to. That's one
> of the strong points of this graphical, multi-sample method, I
> think. It's usually relatively easy to detect poor specimens.
Yes, I think you are right on the money here. It would probably have
to be well melted and mixed up; as would occur in a volcano.
> Anybody know how long "isochrons" have been used? I'll bet the same
> thing was done well before the term was coined.
Good question. Excuse me while I fire up a search engine...
OK. Sorry to keep you waiting... :-)
<http://www.talkorigins.org/faqs/geohist.html#Radiometric>
gives 1946 (Houtermans) as the first to use isochon methods. But I
would guess that the basic work on isotopes and identical chemistry
established and confirmed the theoretical basis for constant D/Di
in a mix well before this. Nodel prizes in 1921 and 1922 were to do
with discovery of isotopes and their properties.
Cheers -- Chris
Jon Fleming
Don't forget that the rate of diffusion is exponential with
temperature.
>How "liquid" does it have to be (and
>for how long) to reset the isochron slope to zero during a re-melt.
I don't know, but here's some indications:
<http://www.geo.cornell.edu/geology/classes/Geo656/98notes/98Lecture04.pdf>:
"The agent accomplishing isotopic homogenization of a `system' is
usually diffusion, the rate of which, like other reaction rates,
increases exponentially with temperature. Diffusion rates will vary
depending on the element and the properties of the material through
which the element diffuses. We can nevertheless make the general
observation that the greater the length scale the greater will be the
time (or the higher the temperature required) for isotopic
homogenization to be achieved. For the same temperature and duration
of a thermal event, diffusion will more readily achieve isotopic
homogenization on a small scale than on a large one. Thus, if our
samples or subsystems are 'whole rocks' collected meters or perhaps
kilometers apart, the event dated will generally be a higher
temperature one than an event dated by analysis of individual minerals
from a rock specimen whose scale is only a few centimeters."
<http://www.magnet.fsu.edu/publications/1999annualreview/pdfs/geochemistry.pdf>:
"These data point to an extraordinary scale of strontium isotopic
homogenization in non-convecting rocks. The coincidence (within
uncertainties) of the isochron ages indicate that during the regional
metamorphic episode, the strontium in these rocks were isotopically
homogenized over a scale of 10 to 400 meters (the range of distance
between samples at a single locality). The distinctly different
87Sr/86Sr initial ratio values of the three isochrons indicate that
the scale of isotopic homogenization was less than a few kilometers
(the minimum distance between collecting localities."
>And
>more importantly, if it isn't melted "enough", will the samples subsequently
>show nonlinearity , thus indicating an inappropriate specimen?
Almost certainly. Assuming there are no lithological differences
between "original" and "remelted" rock that allows separation, what
you would be looking at is a random sample of points from two or more
lines. Such a random sampling of points with about 6 or 7 samples is
extremely unlikely to form a line.
>Seems they
>would have to. That's one of the strong points of this graphical,
>multi-sample method, I think. It's usually relatively easy to detect poor
>specimens.
Exactly.
<snip>
zoe_althrop
wrote:
>> On Sun, 18 Aug 2002 00:24:40 +0000 (UTC), Jon Fleming
>> <jo...@fleming-nospam.com> wrote:
snip>
>>>I've _seen_ Kiluea.
>>
>> me, too.
>
>[sigh] not me yet, I think/hope we're going in March for our wedding
>anniversary, specifically to see Kiluea!
congrats (in advance).
but no, I didn't mean I was there. I saw it on TV, and it looked like
hot fudge. I bet my long-distance sleuthing impresses you, eh?
>OT: We went to Oregon for vacation a couple of weeks ago and spent a day
>at Mt. St. Helens. Definitely worth the price of admission if you ever
>get up there. Went to Crater Lake, too, but all the smoke from forest
>fires obscured the view, bummer.
now THERE, I have been. Lived there for about 11 years, outside of
Beaverton, across the border from Washington -- and went through the
Mt. St. Helens eruption -- ash all over every inch of our 31 acres.
>> >I've _seen_ lots of film
>> >footage of volcanic eruptions. None of them can be described as
>> >"thick and sludgy".
>>
>> it sure looks thick and sludgy to me...like hot fudge slowly
>> advancing.
>
>I think the viscosity has a lot to do with what part of the lava flow
>you're observing. A leading edge that is cooling rapidly can look pretty
>sludgey, but I know I've seen pictures of it flowing fairly rapidly
>under some circumstances. FWIU, though, the mixing and diffusion of
>molecules would have been done before the lava ever reaches the surface
>anyway. It is even less viscous while in the magma chamber, I would
>think (no cooling yet).
but will churning take place? Mixing? Shouldn't it just sit there,
unmoving?
>Here's a site I found that lists the viscosity of lavas as compared to
>everyday liquids. The info appears to be provide by the U of Hawaii,
>Institute of Geophysics and Planetology, so I guess they know what
>they're talking about.
>
>http://www.spacegrant.hawaii.edu/class_acts/ViscosityTe.html
>
>From the table at the bottom of the page, it looks like pahoehoe lava
>(the kind that Kiluea spills) is 100 to 1,000 times more viscous than
>motor oil (Motor oil is listed at 1 Pascal second at room temp and the
>lava at 100 to 1,000 Pascal seconds with no temps given). I'm not sure
>how that compares to hot fudge, but I found a reference for honey at 20º
>C at around 200 poise or 20 Pascal seconds.
I would think that hot fudge might be the consistency of honey, yes,
so maybe it's more the consistency of cooling fudge.
snip>
>> what causes the mixing? Tornadic activity? Squeezing and pumping
>> activity, like a stomach? What would cause this mixing, Jon?
>
>Convection currents; being squeezed through holes, tunnels, chambers in
>the mantle on the way to the surface; expansion and reduction of
>viscosity with reduced pressure as the magma rises; etc.
isn't this rise to the surface a single instance in time? -- not
sufficient time to thoroughly mix lava, is it? While it sits trapped
underground, why should there be a mixing motion?
snip>
>> yes. So any never-remelted, genuinely old rock should register at
>> least 15 billion years, not 4.5 billion years, right?
>
>[blink, blink] Uh, no, Zoe, about the only element that existed 15 bya
>was hydrogen. There wasn't anything to make rocks out of at that time.
>It wasn't until several billions of years worth of fusion in stars that
>there was enough heavier elements to make any rocks anywhere in the
>universe.
>
>Furthermore, this planet did not exist until a bit less than 5 bya, so
>no Earth rocks could possibly be older than that.
>
>Finally, there can't be 4.5 billion years worth of 'oldD down under'
>because, as Jon explained somewhere else, P, D and Di were all created
>by the supernova(s) that produced the gas cloud our sun and solar system
>condensed out of. We have no way of knowing how old that cloud was and
>how much of these elements were already in it and how much P decayed to
>D before the solar system formed, cooled and Earth rocks were available
>to start an isochron clock ticking.
I meant that there has to be at least 15 billion years worth of D in
the rock. And if D were supposedly formed independently of P in
supernovae, then there would be more than 15 billion years worth of D
in the earth's rocks. Should not the isochrons show this
disproportionate amount of D to P and D to Di?
>So there is _NO WAY_ to get any idea of the age of the Earth just from
>the amount of D found in any rocks or magma.
you're probably right on that score, but should one expect to see
evidence of 15-plus billion years worth of oldD in the isochrons?
snip>
>> so how do you determine when a rock is partially melted?
>
>Depends on the rock and what you mean by partial melt. My understanding
>is that partial melt refers to deeply buried rocks that are heated to
>the point where only some of their crystals melt. The liquid from these
>partial melts rise and become magma, the rest of the rock - less the
>melted crystals - stays where it is. Geologists don't generally see
>these rocks.
>
>A partially *remelted* rock would be a solid rock that is exposed to
>heat but not enough to become completely molten. This would be a
>metamorphic rock, I think. Does that make sense?
>
>Here are some links to info on magma, lava, volcanoes and igneous rocks.
>Maybe you can get some of your answers at one of these.
>
>http://www.tulane.edu/~sanelson/geol204/volcan&magma.htm
was able to print this one.
>
>http://geoweb.tamu.edu/Faculty/Wiltschko/Geology104/assets/files/06b.%20
>Origin%20of%20Magma.htm
can't access this one.
>http://geoweb.tamu.edu/courses/geol101/grossman/Ign.rocks.html
>
>http://www.science.ubc.ca/~geol202/rock_cycle/rockcycle.html
>
>http://www.uoregon.edu/~dogsci/dorsey/geo101/lect6.html
printed these last three. Thanks for the links, June. I guess I've
got some extra reading to do on my two weeks of vacation -- which
starts this Sunday, btw :-).
snip>
----
zoe
zoe_althrop
wrote:
snip>
>Another challenge, Zoe. Use your new found knowledge to convince other
>creationist that their criticism of geologist dating methods are
>wrong.
and what makes you think I'm no longer criticizing geolgist dating
methods?
>Lets see how long it takes before they are calling you a heretic and
>an athiest.
intolerance can be found in all schools of thought.
----
zoe
zoe_althrop
<rle...@ibmrlevett.uklinux.net> wrote:
snip>
>I've just got back from a week in the Algarve seeing stars,
interesting. What did you see?
>so haven't
>caught up with the 5,000-odd posts since I left; this may have been
>dealt with elsehere - if so, I apologise.
>
>My point was that you said you were seeking to reconcile the facts
>with your worldview; which is quite definitely not the same process as
>reconciling your worldview with the facts.
Robin, I have come to my worldview by a different path than science,
and it rests on, for me, an unshakeable foundation. Therefore, the
stray scientific data coming in from left field has to be reconciled
with my worldview, not vice versa.
It's similar to a foundation based on acceptance of the laws of
gravity. This is a foundation so settled that if a a hunk of metal
flies instead of falls, you don't let go of the gravitational
worldview because of this discrepancy. Instead, the fact of flight
has to be reconciled with a worldview that accepts the laws of
gravity.
>More generally, this is why there is no equivalence between your
>worldview and a scientist's. A scientist's worldview is a synthesis
>of the facts, and so is naturally adjusted by advances in knowledge.
>A creationist's worldview is fixed, set by the particular
>interpretation they adopt of whichever holy book the subscribe to, and
>changes in that worldview are only accomplished by recognition of
>unchallengeable facts that unequivocally challenge it. You have
>recognised (and it does you credit, however long it has taken (NB that
>is not intended to be patronising)) the validity of the isochron
>dating method, and therefore, one hopes, of "old" fossiliferous rock.
>What we wait to see is whether that forces a change in your worldview.
not at all. As I've said earlier, and which you may have missed, my
worldview is based on a broader foundation than just science. If
there are a few scientific discrepancies, they do not shake my
experience or my worldview. There IS more to life than science, you
know.
----
zoe
Jon Fleming
wrote:
>On Sun, 18 Aug 2002 05:36:51 +0000 (UTC), jun...@adelphia.net (June)
>wrote:
>>I think the viscosity has a lot to do with what part of the lava flow
>>you're observing. A leading edge that is cooling rapidly can look pretty
>>sludgey, but I know I've seen pictures of it flowing fairly rapidly
>>under some circumstances. FWIU, though, the mixing and diffusion of
>>molecules would have been done before the lava ever reaches the surface
>>anyway. It is even less viscous while in the magma chamber, I would
>>think (no cooling yet).
>
>but will churning take place?
Yes.
>Mixing?
Yes
> Shouldn't it just sit there,
>unmoving?
No. If it just sat there unmoving, it would remain many many miles
below the surface. It reached the surface, therefore it moved.
<snip>
>>Convection currents; being squeezed through holes, tunnels, chambers in
>>the mantle on the way to the surface; expansion and reduction of
>>viscosity with reduced pressure as the magma rises; etc.
>
>isn't this rise to the surface a single instance in time?
Well, not really, but maybe.
>-- not
>sufficient time to thoroughly mix lava, is it?
Why not?
>While it sits trapped
>underground, why should there be a mixing motion?
It doesn't just sit trapped underground, there are convection currents
and suchlike.
<snip>
>>Finally, there can't be 4.5 billion years worth of 'oldD down under'
>>because, as Jon explained somewhere else, P, D and Di were all created
>>by the supernova(s) that produced the gas cloud our sun and solar system
>>condensed out of. We have no way of knowing how old that cloud was and
>>how much of these elements were already in it and how much P decayed to
>>D before the solar system formed, cooled and Earth rocks were available
>>to start an isochron clock ticking.
>
>I meant that there has to be at least 15 billion years worth of D in
>the rock.
Nope. P and D did not exist 15 billion years ago.
>And if D were supposedly formed independently of P in
>supernovae, then there would be more than 15 billion years worth of D
>in the earth's rocks.
On the average, there is something on the order of 10 billion years of
D from decayed P plus whatever D was created in supernovae.
> Should not the isochrons show this
>disproportionate amount of D to P and D to Di?
Does it not?
Also, remember "on the average" from above. Individual rocks deviate
significantly from the average over all the Earth.
<snip>
Jon Fleming
wrote:
<snip>
>>I think the viscosity has a lot to do with what part of the lava flow
>>you're observing. A leading edge that is cooling rapidly can look pretty
>>sludgey, but I know I've seen pictures of it flowing fairly rapidly
>>under some circumstances. FWIU, though, the mixing and diffusion of
>>molecules would have been done before the lava ever reaches the surface
>>anyway. It is even less viscous while in the magma chamber, I would
>>think (no cooling yet).
>
>but will churning take place? Mixing? Shouldn't it just sit there,
>unmoving?
Actually, while all this discussion of mixing is interesting in its
own right, it has very little to do with the constant oldD/Di
assumption of isochron dating.
The primary mechanism of mixing in magma is not mechanical; it's
diffusion. Because the atoms are hot and moving very fast (on an
atomic scale; you can't easily _see_ them moving) they tend to diffuse
to make the concentration equal throughout the melt, even if the melt
is just sitting there apparently unmoving.
It's difficult to conceive of a "kitchen table" experiment to
illustrate this, because temperature plays so strong a role. The
diffusion rate depends exponentially on temperature; that means that
as the temperature increases a little the diffusion rate increases a
lot. You can't see what diffusion looks like in molten rock unless
you set something up with a temperature very close to that of the
molten rock; and that's not a good thing to do in your kitchen.
<snip>
Andrew Arensburger
> The primary mechanism of mixing in magma is not mechanical; it's
> diffusion. Because the atoms are hot and moving very fast (on an
> atomic scale; you can't easily _see_ them moving) they tend to diffuse
> to make the concentration equal throughout the melt, even if the melt
> is just sitting there apparently unmoving.
> It's difficult to conceive of a "kitchen table" experiment to
> illustrate this, because temperature plays so strong a role.
The best I can come up with is to take two saucepans of water,
put a bouillon cube in the middle, turn on the heat under one of them,
and see how quickly the bouillon dissolves. (I was originally thinking
of sugar, but dissolved bouillon is visible.)
Is this even roughly comparable to what's going on in magma?
One problem that I can think of is that you don't want to turn
the heat up high enough to induce significant convection, since the
point of the exercise is to see how quickly the molecules in the
bouillon get diffused without mixing.
There's also the problem of evaporation.
Any thoughts?
--
Andrew Arensburger, Systems guy University of Maryland
arensb.no-...@glue.umd.edu Office of Information Technology
Carpenter's rule: cut to fit; beat into place.
Jon Fleming
<arensb.no-...@glue.umd.edu> wrote:
>Jon Fleming <jo...@fleming-nospam.com> wrote:
>> The primary mechanism of mixing in magma is not mechanical; it's
>> diffusion. Because the atoms are hot and moving very fast (on an
>> atomic scale; you can't easily _see_ them moving) they tend to diffuse
>> to make the concentration equal throughout the melt, even if the melt
>> is just sitting there apparently unmoving.
>
>> It's difficult to conceive of a "kitchen table" experiment to
>> illustrate this, because temperature plays so strong a role.
>
> The best I can come up with is to take two saucepans of water,
>put a bouillon cube in the middle, turn on the heat under one of them,
>and see how quickly the bouillon dissolves. (I was originally thinking
>of sugar, but dissolved bouillon is visible.)
> Is this even roughly comparable to what's going on in magma?
> One problem that I can think of is that you don't want to turn
>the heat up high enough to induce significant convection, since the
>point of the exercise is to see how quickly the molecules in the
>bouillon get diffused without mixing.
> There's also the problem of evaporation.
>
> Any thoughts?
Yeah, I think convection is going to swamp diffusion in that
experiment.
Florian
> Jon Fleming <jo...@fleming-nospam.com> wrote:
> > The primary mechanism of mixing in magma is not mechanical; it's
> > diffusion. Because the atoms are hot and moving very fast (on an
> > atomic scale; you can't easily _see_ them moving) they tend to
> > diffuse to make the concentration equal throughout the melt, even
> > if the melt is just sitting there apparently unmoving.
>
> > It's difficult to conceive of a "kitchen table" experiment to
> > illustrate this, because temperature plays so strong a role.
>
> The best I can come up with is to take two saucepans of water,
> put a bouillon cube in the middle, turn on the heat under one of
> them, and see how quickly the bouillon dissolves. (I was originally
> thinking of sugar, but dissolved bouillon is visible.)
> Is this even roughly comparable to what's going on in magma?
> One problem that I can think of is that you don't want to turn
> the heat up high enough to induce significant convection, since the
> point of the exercise is to see how quickly the molecules in the
> bouillon get diffused without mixing.
> There's also the problem of evaporation.
>
> Any thoughts?
How about this:
Take two or more ice cube trays, fill them with different flavors of
Kool-Aid, and freeze. Dump the resulting varicolored solids in a
transparent bowl and pour in enough ice-cold water to fill the air
gaps. Cover and place on a quiet shelf in the refrigerator.
For the next day or two you can play around with partially-melted
isochron analogies, and then you can enjoy the homogenous result and
see if it gives you a fuzzy window into the true flavor of Kool-Aid.
--
odoratusque est Dominus odorem suavitatis
Rodjk
> On Sun, 18 Aug 2002 22:19:14 +0000 (UTC), rjk...@yahoo.com (Rodjk)
> wrote:
>
> snip>
>
> >Another challenge, Zoe. Use your new found knowledge to convince other
> >creationist that their criticism of geologist dating methods are
> >wrong.
>
> and what makes you think I'm no longer criticizing geolgist dating
> methods?
The point is, you now know that the criticisms that YEC's have been
using are wrong. You may still disagree with the ages, but you know
that the criticism is wrong.
(Ok, personally, I really don't think you have learned anything, and I
think you will wait a few months and start all over again with the
same nonsense. I think that Bjoern and Jon have wasted their time. My
opinion)
>
> >Lets see how long it takes before they are calling you a heretic and
> >an athiest.
>
> intolerance can be found in all schools of thought.
By Christians??? Perish the thought!!!
ROTFLMAO!!!!
Rodjk #613
>
> ----
> zoe
Andrew Arensburger
> On Sun, 18 Aug 2002 05:36:51 +0000 (UTC), jun...@adelphia.net (June)
> wrote:
>>So there is _NO WAY_ to get any idea of the age of the Earth just from
>>the amount of D found in any rocks or magma.
> you're probably right on that score, but should one expect to see
> evidence of 15-plus billion years worth of oldD in the isochrons?
Just how much _is_ 15 billion years' worth of D, anyway?
As others have pointed out, 15 billion years ago, there was
pretty much just hydrogen, which isn't D in any isochron dating method
that I'm aware of.
I suspect that you meant to ask a somewhat different question:
if the Earth was formed with a certain amount of P, which has been
decaying to D for 4.6 billion years, shouldn't we see this D
somewhere?
If so, then yes, unless the daughter product can escape
somehow (e.g., if something decays to helium, then the helium can
escape the atmosphere).
The problem is, how much D would you expect to see? On Earth
the D that you see either came from
a) "primordial" D that was already around when the Earth was formed
b) radioactive decay from P
c) none of the above
For (a), you'd have to estimate how much D was produced in the
supernova(e) that created the material from which the solar system was
formed. Or at least estimate how much D the earth started out with.
I'm not an astronomer, but I'm pretty sure that supernovae don't
produce equal amounts of, say, iron and calcium. For all I know, they
produce a thousand times more strontium than rubidium.
For (b), you'd have to estimate how much P the earth started
out with.
(c) is a bit more general, and might include meteors falling
to the Earth after some given time. Or, if D is a gas, then (c) might
include D escaping the atmosphere. Hopefully, if your D is strontium,
then (c) is small enough to ignore.
On top of this, there's the problem of trying to measure the
amount of P and D on Earth today.
So unless you know how much P and D the Earth started with,
you can't say how much there ought to be today, if the Earth is 4.6
billion years old. It's like buying something on sale: you can't tell
how much you've saved unless you know how much it originally cost.
--
Andrew Arensburger, Systems guy University of Maryland
arensb.no-...@glue.umd.edu Office of Information Technology
Contains less than 2% U.S. RDA for this newsgroup.
June
> On Sun, 18 Aug 2002 05:36:51 +0000 (UTC), jun...@adelphia.net (June)
> wrote:
>
> >> On Sun, 18 Aug 2002 00:24:40 +0000 (UTC), Jon Fleming
> >> <jo...@fleming-nospam.com> wrote:
>
> snip>
>
> >>>I've _seen_ Kiluea.
> >>
> >> me, too.
> >
> >[sigh] not me yet, I think/hope we're going in March for our wedding
> >anniversary, specifically to see Kiluea!
>
> congrats (in advance).
Thx, just hope we get to go.
>
> but no, I didn't mean I was there. I saw it on TV, and it looked like
> hot fudge. I bet my long-distance sleuthing impresses you, eh?
It would probably be better if you made it clear where you get your info
in the future. :-p
>
> >OT: We went to Oregon for vacation a couple of weeks ago and spent a day
> >at Mt. St. Helens. Definitely worth the price of admission if you ever
> >get up there. Went to Crater Lake, too, but all the smoke from forest
> >fires obscured the view, bummer.
>
> now THERE, I have been. Lived there for about 11 years, outside of
> Beaverton, across the border from Washington -- and went through the
> Mt. St. Helens eruption -- ash all over every inch of our 31 acres.
WOW. That must have been an experience. Did you ever go see the exhibits
up at the volcano? "All" I've ever lived through are some earthquakes,
floods and fires here in Southern Cal. <g>
>
> >> >I've _seen_ lots of film
> >> >footage of volcanic eruptions. None of them can be described as
> >> >"thick and sludgy".
> >>
> >> it sure looks thick and sludgy to me...like hot fudge slowly
> >> advancing.
> >
> >I think the viscosity has a lot to do with what part of the lava flow
> >you're observing. A leading edge that is cooling rapidly can look pretty
> >sludgey, but I know I've seen pictures of it flowing fairly rapidly
> >under some circumstances. FWIU, though, the mixing and diffusion of
> >molecules would have been done before the lava ever reaches the surface
> >anyway. It is even less viscous while in the magma chamber, I would
> >think (no cooling yet).
>
> but will churning take place? Mixing? Shouldn't it just sit there,
> unmoving?
Yes, yes & no. As a thought experiment visualize oatmeal in a pressure
cooker with milk and brown sugar dumped on top. After you turn on the
heat, does it all 'just sit there'?
>
> >Here's a site I found that lists the viscosity of lavas as compared to
> >everyday liquids. The info appears to be provide by the U of Hawaii,
> >Institute of Geophysics and Planetology, so I guess they know what
> >they're talking about.
> >
> >http://www.spacegrant.hawaii.edu/class_acts/ViscosityTe.html
> >
> >From the table at the bottom of the page, it looks like pahoehoe lava
> >(the kind that Kiluea spills) is 100 to 1,000 times more viscous than
> >motor oil (Motor oil is listed at 1 Pascal second at room temp and the
> >lava at 100 to 1,000 Pascal seconds with no temps given). I'm not sure
> >how that compares to hot fudge, but I found a reference for honey at 20º
> >C at around 200 poise or 20 Pascal seconds.
>
> I would think that hot fudge might be the consistency of honey, yes,
> so maybe it's more the consistency of cooling fudge.
Ha! Found this table from a geology education site.
https://courseware.vt.edu/users/bekken/GS1104/HW/HW3/HW3.html#Table
It lists volcanic flows at 5 km/hr for andesitic lava (the really thick
stuff) and 5-30 km/hr for basaltic lava (like Kiluea). So some of it can
get going pretty fast!
Here's a reference to some lava flows traveling 50-60 mph!
http://www.cotf.edu/ete/modules/volcanoes/vhazards.html
>
> snip>
>
> >> what causes the mixing? Tornadic activity? Squeezing and pumping
> >> activity, like a stomach? What would cause this mixing, Jon?
> >
> >Convection currents; being squeezed through holes, tunnels, chambers in
> >the mantle on the way to the surface; expansion and reduction of
> >viscosity with reduced pressure as the magma rises; etc.
>
> isn't this rise to the surface a single instance in time? -- not
> sufficient time to thoroughly mix lava, is it?
No, magma chambers (with magma) can exist for tens of thousands of years
and may never erupt, just harden in place. Certainly many of the
intrusions of magma into crustal rocks (creating granite) never flow to
the surface. The movement of magma can also take thousands & thousands
of years to reach the crust/surface.
> While it sits trapped
> underground, why should there be a mixing motion?
Jon and others have pointed out that _most_ of the the mixing will be
through diffusion, not mechanical mixing. What this means, basically, is
that even just sitting still, the molecules will tend to distribute
themselves evenly within the mix by chemical diffusion.
>
> snip>
>
> >> yes. So any never-remelted, genuinely old rock should register at
> >> least 15 billion years, not 4.5 billion years, right?
> >
> >[blink, blink] Uh, no, Zoe, about the only element that existed 15 bya
> >was hydrogen. There wasn't anything to make rocks out of at that time.
> >It wasn't until several billions of years worth of fusion in stars that
> >there was enough heavier elements to make any rocks anywhere in the
> >universe.
> >
> >Furthermore, this planet did not exist until a bit less than 5 bya, so
> >no Earth rocks could possibly be older than that.
> >
> >Finally, there can't be 4.5 billion years worth of 'oldD down under'
> >because, as Jon explained somewhere else, P, D and Di were all created
> >by the supernova(s) that produced the gas cloud our sun and solar system
> >condensed out of. We have no way of knowing how old that cloud was and
> >how much of these elements were already in it and how much P decayed to
> >D before the solar system formed, cooled and Earth rocks were available
> >to start an isochron clock ticking.
>
> I meant that there has to be at least 15 billion years worth of D in
> the rock. And if D were supposedly formed independently of P in
> supernovae, then there would be more than 15 billion years worth of D
> in the earth's rocks. Should not the isochrons show this
> disproportionate amount of D to P and D to Di?
1) Depending on what element you're designating as D, there would be
something less than 15 b.y. worth _in_the_universe_ because none of the
heavier elements existed for the first few tens or hundreds of millions
of years of the universe.
2) I don't think it's known if there is an even distribution of
elements, especially heavier elements, in the universe, or even in our
galaxy (and there probably ISN'T a even distribution, especially wrt
local conditions). So, regardless of how much of any element there was
in the universe or galaxy, that might not have a lot to do with the
concentration of that element in the _local_ cloud that our solar system
condensed out of. I don't see any way to correlate between any '15
billion years worth' of any element and how much of that element was or
is on the Earth.
3) What 'disproportionate amount'? How would you know if it was
disproportionate or not? You'd have to be able to determine how much of
all these elements there was in the original solar dust cloud & how much
ended up in the Earth. Does Jupiter have the same chemical composition
as the Earth? (A: no) Did more of certain elements end up on Saturn than
on the Earth? (A: yes). So how much is 'disproportionate'? You can't
tell from looking at P, D and Di in some minerals in some rocks.
4) Different minerals are made up of different elements. How could you
tell the overall abundance of any element on the Earch when you're ONLY
looking at minerals that are rich in that (or those) elements to do an
isochron? We wouldn't be looking at minerals that *didn't* have P, D and
Di to do an isochron (for any of various P, D & Di designations). Do
quartz crystals have the same composition as zircons? Olivine? Garnets?
Amphibole? Biotite? You can't tell anything about the age of the Earth
by looking at the abundance of some elements in some minerals in some
rocks.
5) Different minerals in the same igneous rock formation would take up
different proportions of P to oldD and Di as it cooled. There would be
different D/Di ratios in different magmas (and therefore in the
different rocks after cooling) due to differences in the composition of
the rocks that melted to make up the magma. From what I've read no two
magmas are the same composition of elements, especially in the trace
elements. You cannot determine an age for the *Earth*, even a fuzzy one,
by looking at the various abundances of P's, D's and Di's in some of the
minerals in some of the rocks.
You'd have to test _all_ rocks on the planet plus determine the exact
composition of the interior of the Earth to determine how much of any P,
D and Di combinations there are on the planet. And you still wouldn't be
able to know how much of each element was in the original dust cloud.
Again, you CANNOT determine the age of the Earth, not even a fuzzy one,
from the amount of any element on or in it.
>
> >So there is _NO WAY_ to get any idea of the age of the Earth just from
> >the amount of D found in any rocks or magma.
>
> you're probably right on that score, but should one expect to see
> evidence of 15-plus billion years worth of oldD in the isochrons?
No, see above on creation and distribution of elements.
>
> snip>
>
> >> so how do you determine when a rock is partially melted?
> >
> >Depends on the rock and what you mean by partial melt. My understanding
> >is that partial melt refers to deeply buried rocks that are heated to
> >the point where only some of their crystals melt. The liquid from these
> >partial melts rise and become magma, the rest of the rock - less the
> >melted crystals - stays where it is. Geologists don't generally see
> >these rocks.
> >
> >A partially *remelted* rock would be a solid rock that is exposed to
> >heat but not enough to become completely molten. This would be a
> >metamorphic rock, I think. Does that make sense?
> >
> >Here are some links to info on magma, lava, volcanoes and igneous rocks.
> >Maybe you can get some of your answers at one of these.
> >
> >http://www.tulane.edu/~sanelson/geol204/volcan&magma.htm
>
> was able to print this one.
>
> >
> >http://geoweb.tamu.edu/Faculty/Wiltschko/Geology104/assets/files/06b.%20
> >Origin%20of%20Magma.htm
>
> can't access this one.
Well, you could try to unwrap it, but this one was lecture notes and not
as informative as some of the others anyway.
>
> >http://geoweb.tamu.edu/courses/geol101/grossman/Ign.rocks.html
> >
> >http://www.science.ubc.ca/~geol202/rock_cycle/rockcycle.html
> >
> >http://www.uoregon.edu/~dogsci/dorsey/geo101/lect6.html
>
> printed these last three. Thanks for the links, June. I guess I've
> got some extra reading to do on my two weeks of vacation -- which
> starts this Sunday, btw :-).
Enjoy your vacation.
[sigsnip]
zoe_althrop
wrote:
>> >OT: We went to Oregon for vacation a couple of weeks ago and spent a day
>> >at Mt. St. Helens. Definitely worth the price of admission if you ever
>> >get up there. Went to Crater Lake, too, but all the smoke from forest
>> >fires obscured the view, bummer.
>>
>> now THERE, I have been. Lived there for about 11 years, outside of
>> Beaverton, across the border from Washington -- and went through the
>> Mt. St. Helens eruption -- ash all over every inch of our 31 acres.
>
>WOW. That must have been an experience. Did you ever go see the exhibits
>up at the volcano?
no. I left Oregon the next year, and any attempt to see the remains
before that would have been too soon for me. What if it exploded
again while I was there poking around the site?
>"All" I've ever lived through are some earthquakes,
>floods and fires here in Southern Cal. <g>
"All"? That's plenty.
>>
>> >> >I've _seen_ lots of film
>> >> >footage of volcanic eruptions. None of them can be described as
>> >> >"thick and sludgy".
>> >>
>> >> it sure looks thick and sludgy to me...like hot fudge slowly
>> >> advancing.
>> >
>> >I think the viscosity has a lot to do with what part of the lava flow
>> >you're observing. A leading edge that is cooling rapidly can look pretty
>> >sludgey, but I know I've seen pictures of it flowing fairly rapidly
>> >under some circumstances. FWIU, though, the mixing and diffusion of
>> >molecules would have been done before the lava ever reaches the surface
>> >anyway. It is even less viscous while in the magma chamber, I would
>> >think (no cooling yet).
>>
>> but will churning take place? Mixing? Shouldn't it just sit there,
>> unmoving?
>
>Yes, yes & no. As a thought experiment visualize oatmeal in a pressure
>cooker with milk and brown sugar dumped on top. After you turn on the
>heat, does it all 'just sit there'?
yes. I've used pressure cookers, in which I've dumped whatever the
ingredients were plus onions and garlic and whatever else pleased my
palate, and when it was done and I opened it, the stuff was still in
the same configurations, except they were now al dente. I had to
manually whiz or stir or mix in order to achieve whatever final dish
was desired.
>> >Here's a site I found that lists the viscosity of lavas as compared to
>> >everyday liquids. The info appears to be provide by the U of Hawaii,
>> >Institute of Geophysics and Planetology, so I guess they know what
>> >they're talking about.
>> >
>> >http://www.spacegrant.hawaii.edu/class_acts/ViscosityTe.html
>> >
>> >From the table at the bottom of the page, it looks like pahoehoe lava
>> >(the kind that Kiluea spills) is 100 to 1,000 times more viscous than
>> >motor oil (Motor oil is listed at 1 Pascal second at room temp and the
>> >lava at 100 to 1,000 Pascal seconds with no temps given). I'm not sure
>> >how that compares to hot fudge, but I found a reference for honey at 20º
>> >C at around 200 poise or 20 Pascal seconds.
>>
>> I would think that hot fudge might be the consistency of honey, yes,
>> so maybe it's more the consistency of cooling fudge.
>
>Ha! Found this table from a geology education site.
>
>https://courseware.vt.edu/users/bekken/GS1104/HW/HW3/HW3.html#Table
>
> It lists volcanic flows at 5 km/hr for andesitic lava (the really thick
>stuff) and 5-30 km/hr for basaltic lava (like Kiluea). So some of it can
>get going pretty fast!
was the speed tested at the height of the explosive force, while it
was shooting into the air, or after it landed and began to move
downwards and forwards? And how steep was the incline, and how much
did the steepness add to the speed? Hot fudge, or cooling fudge will
flow at different rates, depending on the angle of the spoon and/or
cookie sheet.
>Here's a reference to some lava flows traveling 50-60 mph!
>
>http://www.cotf.edu/ete/modules/volcanoes/vhazards.html
ah, I see here that the conditions do make a difference. It says,
"The speed of a flow depends on the viscosity of the lava and the
incline of the volcano's slope." So the 50-60 miles per hour speed
might come from a viscosity that is almost the consistency of water.
Not much rock could form from that kind of flow, could there?
>>
>> snip>
>>
>> >> what causes the mixing? Tornadic activity? Squeezing and pumping
>> >> activity, like a stomach? What would cause this mixing, Jon?
>> >
>> >Convection currents; being squeezed through holes, tunnels, chambers in
>> >the mantle on the way to the surface; expansion and reduction of
>> >viscosity with reduced pressure as the magma rises; etc.
>>
>> isn't this rise to the surface a single instance in time? -- not
>> sufficient time to thoroughly mix lava, is it?
>
>No, magma chambers (with magma) can exist for tens of thousands of years
>and may never erupt, just harden in place. Certainly many of the
>intrusions of magma into crustal rocks (creating granite) never flow to
>the surface. The movement of magma can also take thousands & thousands
>of years to reach the crust/surface.
well, I haven't read through your last links yet -- keeping them for
when I begin to relax and unwind next week. So will hold my fingers on
this one.
>
>> While it sits trapped
>> underground, why should there be a mixing motion?
>
>Jon and others have pointed out that _most_ of the the mixing will be
>through diffusion, not mechanical mixing. What this means, basically, is
>that even just sitting still, the molecules will tend to distribute
>themselves evenly within the mix by chemical diffusion.
yes, but depending on the viscosity. The thinner the material, the
more diffusion. But molten rock itself should be quite viscuous, I
would think. In which case, diffusion should be almost nil.
you say this with great authority. Why do you think it would be
millions of years and not billions of years or thousands of years?
>
>2) I don't think it's known if there is an even distribution of
>elements, especially heavier elements, in the universe, or even in our
>galaxy (and there probably ISN'T a even distribution, especially wrt
>local conditions). So, regardless of how much of any element there was
>in the universe or galaxy, that might not have a lot to do with the
>concentration of that element in the _local_ cloud that our solar system
>condensed out of. I don't see any way to correlate between any '15
>billion years worth' of any element and how much of that element was or
>is on the Earth.
are you saying that the age of 4.5 billion years for the Earth's
existence is a pure guess, based on nothing more than idle
speculation? Sounds like it to me. Quote: I don't see any way to
correlate...."
>3) What 'disproportionate amount'?
if Di remains constant for 4.5 billion years, but D has been
increasing over that same period of time, then you would expect to see
4.5 billion times more D than Di, or something like that, anyway.
> How would you know if it was
>disproportionate or not? You'd have to be able to determine how much of
>all these elements there was in the original solar dust cloud & how much
>ended up in the Earth. Does Jupiter have the same chemical composition
>as the Earth? (A: no) Did more of certain elements end up on Saturn than
>on the Earth? (A: yes). So how much is 'disproportionate'? You can't
>tell from looking at P, D and Di in some minerals in some rocks.
then you can't even estimate the Earth's age, either, then.
>4) Different minerals are made up of different elements. How could you
>tell the overall abundance of any element on the Earch when you're ONLY
>looking at minerals that are rich in that (or those) elements to do an
>isochron? We wouldn't be looking at minerals that *didn't* have P, D and
>Di to do an isochron (for any of various P, D & Di designations). Do
>quartz crystals have the same composition as zircons? Olivine? Garnets?
>Amphibole? Biotite? You can't tell anything about the age of the Earth
>by looking at the abundance of some elements in some minerals in some
>rocks.
you're making a great case against knowing the age of the earth and
even the universe -- at least based on using radioactive isotopes as
the means of determining age.
>5) Different minerals in the same igneous rock formation would take up
>different proportions of P to oldD and Di as it cooled. There would be
>different D/Di ratios in different magmas (and therefore in the
>different rocks after cooling) due to differences in the composition of
>the rocks that melted to make up the magma. From what I've read no two
>magmas are the same composition of elements, especially in the trace
>elements. You cannot determine an age for the *Earth*, even a fuzzy one,
okay, I'll accept that. We cannot determine an age for the Earth,
period. Not even a fuzzy one. Take back the 4.5 billion years and
the 15 billion years. Start over from square one.
>by looking at the various abundances of P's, D's and Di's in some of the
>minerals in some of the rocks.
so what other method is used to determine the age of the earth besides
radioactive isotopes?
>You'd have to test _all_ rocks on the planet plus determine the exact
>composition of the interior of the Earth to determine how much of any P,
>D and Di combinations there are on the planet. And you still wouldn't be
>able to know how much of each element was in the original dust cloud.
>Again, you CANNOT determine the age of the Earth, not even a fuzzy one,
>from the amount of any element on or in it.
so how did you reach the 4.5 billion year figure, pray tell?
>> >So there is _NO WAY_ to get any idea of the age of the Earth just from
>> >the amount of D found in any rocks or magma.
>>
>> you're probably right on that score, but should one expect to see
>> evidence of 15-plus billion years worth of oldD in the isochrons?
>
>No, see above on creation and distribution of elements.
I saw, and now am of the opinion that you guys really have no good
foundation to state the age of the earth or universe.
snip>
----
zoe
zoe_althrop
then on what basis do you say that the Earth is 4.6 billion years old?
You've stated the problems very clearly. How have these problems been
overcome in order to come to your 4.6 billion years of age? Using
meteors or moon rocks would run you into the same problems. So tell
me again, how do you know the Earth's age without knowing for sure
what the starting P and D were?
----
zoe
Jon Fleming
wrote:
>On Thu, 22 Aug 2002 00:53:24 +0000 (UTC), Andrew Arensburger
Nope, you've changed his statement significantly. He stated that you
cannot predict the quantity of P and D today unless you know how much
the Earth started with. This is totally independent of and different
from the question of how we arrive at an age of the Earth. The short
answer is that we arrive at the age of the Earth in ways that don't
require us to know or assume how much P and D were incorporated in
the Earth when it formed.
>How have these problems been
>overcome in order to come to your 4.6 billion years of age?
There are no problems. We use methods that don't depend on how much P
and D were present when the rock formed.
>Using
>meteors or moon rocks would run you into the same problems. So tell
>me again, how do you know the Earth's age without knowing for sure
>what the starting P and D were?
Good Lord!
After all this time, you don't even realize that the amount of P at
solidification (as long as there is _some_) makes no difference to
_ANY_ radiometric dating method. Even the simplest generic method
isn't affected in the slightest by the amount of P at solidification
(as long as there's some P, the amount doesn't matter). We _NEVER_
need to know the amount of P at solidification to date a rock
As for the amount of D, there are several possibilities. We can use a
method (such as the isochron method) that _DOES_ _NOT_ DEPEND_ on the
amount of D at solidification (so we don't need to know the amount of
D at solidification). We can use a method that presumes there was no
D in the rock at solidification, and use various methods of checking
that assumption (such as concordia dating or concordia-discordia
dating).
Richard Uhrich
> On Thu, 22 Aug 2002 06:35:28 +0000 (UTC), jun...@adelphia.net (June)
> wrote:
<snip>
>
> so what other method is used to determine the age of the earth besides
> radioactive isotopes?
>
>
>>You'd have to test _all_ rocks on the planet plus determine the exact
>>composition of the interior of the Earth to determine how much of any P,
>>D and Di combinations there are on the planet. And you still wouldn't be
>>able to know how much of each element was in the original dust cloud.
>>Again, you CANNOT determine the age of the Earth, not even a fuzzy one,
>>
>>from the amount of any element on or in it.
>
> so how did you reach the 4.5 billion year figure, pray tell?
>
>
>>>>So there is _NO WAY_ to get any idea of the age of the Earth just from
>>>>the amount of D found in any rocks or magma.
>>>>
>>>you're probably right on that score, but should one expect to see
>>>evidence of 15-plus billion years worth of oldD in the isochrons?
>>>
>>No, see above on creation and distribution of elements.
>>
>
> I saw, and now am of the opinion that you guys really have no good
> foundation to state the age of the earth or universe.
>
> snip>
>
> ----
> zoe
>
>
You form grandiose opinions so easily!
Age of universe is extrapolated from rate of expansion of Big Bang and
from computed age of oldest stars. The two independent methods correlate
fairly well. The age of the earth is given by the computed age of the
sun. The evolution of stars is known from calculations of nuclear
reactions within as mass and chemical elements evolve. The required data
was are present intrinsic brightness and spectra. No isochron. Irrelevant.
--
Richard Uhrich
---
Ignorance more frequently begets confidence than does knowledge. --
Charles Darwin
Jon Fleming
wrote:
>On Thu, 22 Aug 2002 06:35:28 +0000 (UTC), jun...@adelphia.net (June)
>wrote:
<snip>
>>Yes, yes & no. As a thought experiment visualize oatmeal in a pressure
>>cooker with milk and brown sugar dumped on top. After you turn on the
>>heat, does it all 'just sit there'?
>
>yes. I've used pressure cookers, in which I've dumped whatever the
>ingredients were plus onions and garlic and whatever else pleased my
>palate, and when it was done and I opened it, the stuff was still in
>the same configurations, except they were now al dente.
All solids. Was the liquid unmixed?
<snip>
>>Jon and others have pointed out that _most_ of the the mixing will be
>>through diffusion, not mechanical mixing. What this means, basically, is
>>that even just sitting still, the molecules will tend to distribute
>>themselves evenly within the mix by chemical diffusion.
>
>yes, but depending on the viscosity.
Not much. It depends much more on temperature.
>The thinner the material, the
>more diffusion.
And if the material is slightly hotter, the _much_ _more_ the
diffusion.
> But molten rock itself should be quite viscuous, I
>would think. In which case, diffusion should be almost nil.
Wrong. Because temperature has a much higher effect than viscosity,
there's a lot of diffusion in hot magma.
<snip>
>>1) Depending on what element you're designating as D, there would be
>>something less than 15 b.y. worth _in_the_universe_ because none of the
>>heavier elements existed for the first few tens or hundreds of millions
>>of years of the universe.
>
>you say this with great authority. Why do you think it would be
>millions of years and not billions of years or thousands of years?
Because many models have been developed for how elements formed in the
early universe, and only those that predict what June said are
compatible with the observed composition of the Universe today.
>>2) I don't think it's known if there is an even distribution of
>>elements, especially heavier elements, in the universe, or even in our
>>galaxy (and there probably ISN'T a even distribution, especially wrt
>>local conditions). So, regardless of how much of any element there was
>>in the universe or galaxy, that might not have a lot to do with the
>>concentration of that element in the _local_ cloud that our solar system
>>condensed out of. I don't see any way to correlate between any '15
>>billion years worth' of any element and how much of that element was or
>>is on the Earth.
>
>are you saying that the age of 4.5 billion years for the Earth's
>existence is a pure guess, based on nothing more than idle
>speculation? Sounds like it to me. Quote: I don't see any way to
>correlate...."
What she said is totally unrelated to how we determine the age of the
Earth. You don't need to correlate how much of an element was or is
on the Earth to date rocks; all you need is the measurements of that
particular rock (_not_ the entire Earth) today and the known physics
of solidification and radioactive decay.
>
>3) What 'disproportionate amount'?
>
>if Di remains constant for 4.5 billion years, but D has been
>increasing over that same period of time, then you would expect to see
>4.5 billion times more D than Di, or something like that, anyway.
It depends on how much D there was to start with (which we don't know)
and how fast P decays to D (which we do know). "4.5 billion times
more" is a lot, that seems to be possible only if the Earth started
with a gigantic amount of P and that P decays much faster than the
elements used in geological radiometric dating. (C14 is a special
case, because it's being produced as we speak; that's not true for
most radioactive parent elements).
>
>> How would you know if it was
>>disproportionate or not? You'd have to be able to determine how much of
>>all these elements there was in the original solar dust cloud & how much
>>ended up in the Earth. Does Jupiter have the same chemical composition
>>as the Earth? (A: no) Did more of certain elements end up on Saturn than
>>on the Earth? (A: yes). So how much is 'disproportionate'? You can't
>>tell from looking at P, D and Di in some minerals in some rocks.
>
>then you can't even estimate the Earth's age, either, then.
Totally wrong. All you need to date a rock is the measurement of the
rock today. You date a rock, it turns out to be 4.5 billion years
old.
In the many times we've applied the isochron method in the last eight
months or so, where did we need the amount of P and D that were on
the Earth when it formed?
(Hint: we never needed it, it's unnecessary and extraneous
information).
>>4) Different minerals are made up of different elements. How could you
>>tell the overall abundance of any element on the Earch when you're ONLY
>>looking at minerals that are rich in that (or those) elements to do an
>>isochron? We wouldn't be looking at minerals that *didn't* have P, D and
>>Di to do an isochron (for any of various P, D & Di designations). Do
>>quartz crystals have the same composition as zircons? Olivine? Garnets?
>>Amphibole? Biotite? You can't tell anything about the age of the Earth
>>by looking at the abundance of some elements in some minerals in some
>>rocks.
>
>you're making a great case against knowing the age of the earth and
>even the universe -- at least based on using radioactive isotopes as
>the means of determining age.
Nope, she's making a great case for not knowing how much P and D were
on the Earth when it formed. All radiometric dating methods are
independent of how much P and D were on the Earth when it formed.
>>5) Different minerals in the same igneous rock formation would take up
>>different proportions of P to oldD and Di as it cooled. There would be
>>different D/Di ratios in different magmas (and therefore in the
>>different rocks after cooling) due to differences in the composition of
>>the rocks that melted to make up the magma. From what I've read no two
>>magmas are the same composition of elements, especially in the trace
>>elements. You cannot determine an age for the *Earth*, even a fuzzy one,
>
>okay, I'll accept that. We cannot determine an age for the Earth,
>period. Not even a fuzzy one. Take back the 4.5 billion years and
>the 15 billion years. Start over from square one.
Wrong. Radiometric darting methods work, given the measurements of a
particular rock today and the known physics of solidification and
radioactive decay.
>>by looking at the various abundances of P's, D's and Di's in some of the
>>minerals in some of the rocks.
>
>so what other method is used to determine the age of the earth besides
>radioactive isotopes?
For the overall age of the Earth, there's not much else. Of course,
the amount of P and D present when the Earth formed don't affect the
validity or reliability of radiometric dating. Good thing, since we
don't really know the amount of P and D present when the Earth formed.
>>You'd have to test _all_ rocks on the planet plus determine the exact
>>composition of the interior of the Earth to determine how much of any P,
>>D and Di combinations there are on the planet. And you still wouldn't be
>>able to know how much of each element was in the original dust cloud.
>>Again, you CANNOT determine the age of the Earth, not even a fuzzy one,
>>from the amount of any element on or in it.
>
>so how did you reach the 4.5 billion year figure, pray tell?
By dating rocks that turn out to be 4.5 billion years old.
>>> >So there is _NO WAY_ to get any idea of the age of the Earth just from
>>> >the amount of D found in any rocks or magma.
>>>
>>> you're probably right on that score, but should one expect to see
>>> evidence of 15-plus billion years worth of oldD in the isochrons?
>>
>>No, see above on creation and distribution of elements.
>
>I saw, and now am of the opinion that you guys really have no good
>foundation to state the age of the earth or universe.
Why not? What methods do we use that depend on how much P and D were
present when the Earth or the Universe formed?
Jon Fleming
Er, isochron analysis of meteoric rocks that we have reason to believe
are about the same age as the Solar System give the same answer.
And isochron analysis of the oldest rocks on Earth set a lower bound
of 3.8 billion years.
Jon Fleming
wrote:
>so how did you reach the 4.5 billion year figure, pray tell?
Dalrymple wrote an entire book on the subject, and he didn't cover all
the evidence; just that which we get from rocks. "The Age of the
Earth".
Richard Uhrich
OK, thanks. Similar results. Point being that dating of volcanic rocks
by isochron method does not require age of earth or universe, nor vice
versa.
John Harshman
(zoe_althrop) wrote:
> On Thu, 22 Aug 2002 06:35:28 +0000 (UTC), jun...@adelphia.net (June)
> wrote:
[snip]
> you're making a great case against knowing the age of the earth and
> even the universe -- at least based on using radioactive isotopes as
> the means of determining age.
Right. The age of the universe is not determined using radioactive
isotopes. Those estimates are something else entirely.
The age of the earth is determined only indirectly, by dating rocks. A
rock formed on earth must be as young as or younger than the earth. The
oldest known rocks on earth are just a bit under 4.5 by. The earth is also
dated by reference to moon rocks, under the theory that the moon and earth
are about the same age.
[more snip]
--
*Note the obvious spam-defeating modification
to my address if you reply by email.