At Uncommon Descent, “niwrad” has posted a link to a Sequences Probability Calculator. This webserver allows you to set a number of trials (“chemical reactions”) per second, the number of letters per position (20 for amino acids) and a sequence length, and then it calculates how long it will take for you to get exactly that sequence. Each trial assumes that you draw a sequence at random, and success is only when you exactly match the target sequence. This of course takes nearly forever.
So in effect the process is one of random mutation without natural selection present, or random mutation with natural selection that shows no increase in fitness when a sequence partially matches the target. This leads to many thoughts about evolution, such as:
- Do different species show different sequences for a given protein? Typically they do, so the above scheme implies that they can’t have evolved from common ancestors that had a different protein sequence. They each must have been the result of a separate special creation event.
- If an experimenter takes a gene from one species and puts it into another, so that the protein sequence is now that of the source species, does it still function? If not, why are people so concerned about making transgenic organisms (they’d all be dead anyway)?
- If we make a protein sequence by combining part of a sequence from one species and the rest of that protein sequence from another, will that show function in either of the parent species? (Typically yes, it will).
Does a consideration of the experimental evidence show that the SPC fails to take account of the function of nearby sequences?
The author of the Sequences Probability Calculator views evolution as basically impossible. The SPC assumes that any change in a protein makes it unable to function. Each species sits on a high fitness peak with no shoulders. In fact, experimental studies of protein function are usually frustrating, because it is hard to find noticeable difference of function, at least ones big enough to measure in the laboratory.
A point to emphasize is that the assumptions of the Sequences Probability Calculator are not that different groups of organisms exist on high peaks with unbridgeable gulfs between them, but that each individual species is on such a peak.
Another isolated islands argument. Alleles don’t exist. Gradients of function don’t exist. The current configuration was a target.
I’ve given Niwrad a heads-up, guys. Please play nice.
Happy New Year to all! Thanks to Joe for bringing us back to biology!
From the formulas given in the “Computed Results”, the “Sequences Probability Calculator” apparently also makes some implicit assumptions about those rates of interaction; namely, that individual interactions take place sequentially.
It is not clear what the point of the “number of chemical reactions per trial” is. What does the “number of random trials” mean? Number of trials to accomplish what?
It also seems to be directed at the origins of life issue.
However, in most processes with chemical reactions in complex environments, reactions take place in parallel; there are literally thousands of them going on at the same time.
Not only are these processes going on that the same time, but we can easily learn from an elementary scaling-up exercise that can be done with high school level physics, chemistry, and math, that the energies of interaction on the order of electron volts between atoms and molecules spaced at distances on the order of nanometers scale up to energies on the order of 1026 joules or 1010 megatons of TNT.
Charge-to-mass ratios have enormous consequences for rates of interaction; accelerations due to electrical forces are enormous.
When complex chains of molecules are interacting within themselves and are also immersed in liquids like water, there are literally thousands of things going on simultaneously as these chains and membranes fold and coil according to quantum mechanical rules and the overall structure of the molecules themselves.
All this calculator apparently does is to multiply the number of permutations to get from a random arrangement of a set non-interacting letters to a specified sequence by a “time-per-permutation.” That is NOT how chemistry works.
So the question one has to ask about the “Probability Sequence Calculator” is, “What does it have to do with the reactions among atoms and molecules in any situation that approaches a real environment?”
I always get the impression that ID/creationists simply do not know how chemists and physicists actually do these kinds of calculation.
Would it be possible to construct a sequence probability calculator capable of accepting as a “hit” ANY protein that either might conceivably serve some function, or might conceivably not be harmful? If so, since this is much more nearly how the Real World works, it might illustrate just how protean proteins can be!
I just registered.. first post. I didn’t see an introductory comment section so I’m jumping in here but not intending to make a splash. A commenter on another blog linked to here for info about the Granville Sewell issue. On looking around, I thought what a pleasant surprise, like an oasis in the Mojave. I really appreciate the tone.
Dr Felsenstein, I appreciate your posts and comment. I’ll try to find Theoretical Evolutionary Genetics.
Fixed that for you. 😉
Flint, making such an SPC would involve building in a judgement about which sequences were less fit and which were so much less fit (such as having none of this function at all) that they would not on the same fitness peak. “niwrad’ has built in an assumption that nothing but one target sequence has any function.
Mike Elzinga, I think that simultaneous search by many agents would not solve the problem if niwrad’s assumptions are sensible. One could have every elementary particle in the universe be an organism doing the search, and have one search per zillionth of a second by each of them, and they would still not find the target. That’s the point of Dembski’s Universal Probability Bound. However if nearby sequences have some noticeable amount of function (and empirical results say that) then the whole isolated peaks argument collapses.
petrushka, I don’t think this is just the isolated islands argument we saw before. It is much more restrictive — each species must be on its own isolated island, at least if you ask the SPC search to find only the exact sequence that is in that species. In that sense it is closely connected to Special Creation, and even works against Michael Behe’s version of the argument.
Niwrad is not Bill Dembski. There is some internal evidence that niwrad is not a native English-speaker. Plus niwrad’s arguments are not the sort of stuff Dembski would engage in.
That is not the point of my comment. There is no “target;” what falls out falls out. If it survives in the environment, then it can become a template for something more complex; and that “something” could just as well be a structure that is energetically nearby.
Mathematical modeling of chemical processes on a computer involve some very sophisticated coding; not the sophomoric coding of this “Sequence Probability Calculator.” Good examples of such kinds of computer processing can be found in journals such as Computing in Science and Engineering.
In real physical/chemical processes – including those kinds of processes suspected to be involved in the origins of life – complicated mixtures of atoms and molecules are brought together in a heat bath of sufficient energy that unlikely reactions can occur. If such reactions are also taking place in a solvent such as water or other molten materials, there is a high probability that other dissolved elements get shuttled into these reactions as well. The “solvent” acts as a sort of catalyst.
The products then are “annealed” by being transported by convection or other processes into a less energetic environment where they become stable. Further evolution of these products can then occur on top of the template provided by the molecule formed at high energy; and usually these reactions occur at lower energies than those involved in forming the original template. It is in these kinds of energy cascade that the beginnings of natural selection occur. There is no target; just what survives.
And we don’t necessarily have to be talking about “living” organisms; most of the thinking based on physics and chemistry follow these known types of process. Once complex chains of “soft” matter are formed and are immersed in a heat bath that keeps them “soft and flexible,” these chains can bend, fold, coil, and crawl along themselves forming all kinds of complex structures depending on where exposed bonds occur. And they are able to do all this because they are continually “shaken” by the heat bath in which they are immersed. If the temperature of that heat bath is lowered sufficiently, these folding and crawling processes cease. If the temperature is raised too high, the structures come apart.
There is nothing strange about any of this; many industrial processes make use of this knowledge. Those of us who have studied the properties of materials in the laboratory often make our materials using these notions. It is done routinely; but modeling these processes on a computer demands some very sophisticated coding on parallel processors and super computers.
None of this kind of behavior is ever considered in any of the ID/creationist calculations. The only thing I have ever seen in any of ID/creationist calculations are probabilities of sequences that are simply some basic, a-priori probability raised to a power corresponding to the number of non-interacting characters in a one-dimensional sequence. This is so far from reality that it doesn’t even count as anything worthy of consideration in any realistic chemical and physical process.
I’m pretty sure that niwrad is Giuseppe Sermonti, author of Why is a Fly not a Horse?.
Click here for the evidence.
Welcome to TSZ.
Nice try, keiths, but I’m not clicking through to UD. 😉
I remember thinking that the name niwrad reminded me of Dembski’s type of wit (for lack of a better term). I will admit I have no other evidence to support that theory.
As a side note, when I posted that comment it showed “Niwrad” with a strike-through line, but doesn’t display it that way now.
The basic problem here is, there’s no actual utility in calculating the probability of matching something essentially impossible to match. The “utility” lies entirely in “showing” that evolution can’t happen by constructing a sufficiently inaccurate model of evolution. As such, it’s trivial and boring. We get no insight about how things happen by pounding on the notion that what doesn’t happen, doesn’t happen.
I suggest we not play games of trying to guess the real names of commenters. Unless there is some compelling reason, such as the person misrepresenting who they are or what their background is, this is bad internet form. People have a right to anonymity in these discussions, unless they can be argued to have abused it.
Well said, Flint.
The early research stages of learning how to make compounds and molecular structures consist of a lot of “shake-and-bake” processes followed by sorting. The sorting can be done by a variety of methods depending on the thing one is searching for; e.g., centrifuging, electrophoresis, fluorescence, magnetic properties, etc..
Many times this involves producing all sorts of things, of which the product you are looking for is a very tiny percentage of everything. Then you sort; in other words, it’s artificial selection which could just as well have taken place naturally.
As one learns more about the specific constraints that were involved in the product of interest, one can eventually tailor the processes to encompass just those constraints; thereby making enormous increases in the efficiency of producing the desired product.
This is very likely what we will see as research narrows down the constraints on the processes that went into the origins of life. At this stage, however, we are looking on other planets and moons as well as in the extreme environments on Earth for the conditions that produce the products leading to living organisms. That search has been extremely encouraging.
We are at the “shake-and-bake stage with only some very crude hints of possible constraints. But once we break through to a more precise understanding of the constraints, then progress will move very fast; and we will begin to see routine production of “living” organisms that take in energy and reproduce on their own. It may not occur within my lifetime; but I suspect there are enough clever people out there working on it right now that it will happen. If I had another lifetime to continue research, this is one of the areas that I would love to spend time on.
Barry Arrington’s variation of the tornado in a junkyard.
Every time I see an ID/creationist argument like that, I know exactly where their science education stopped.
Somehow, the concept of process never crosses the minds of the UD folks. Nothing is gradual, nothing is incremental, selection is not involved, feedback is unknown. It’s poof-style magic all the way down.
And to be fair to Barry, if process is disallowed, magic is all that’s left. It’s like he’s playing with a 1-card deck. No matter how he shuffles it, he always draws the same one.
Not a full deck, anyway.
Your words, “the concept of process – nothing is incremental – selection is not involved” is how it was for me from a YEC home, school and church life. Creationism was an essential part of our hope of salvation. Any data not supportive of that belief was a deception of satan and it was unthinkable to disbelieve. Recent creation was true and from that position reality was arranged around that non-negotiable point. The way I see it now, I was brainwashed. To learn secular science changed my world like falling through the looking glass – into reality.
I’m glad I came across this site. Cheers dl
Echoing Joe’s comment above, let’s keep the thread on topic. I don’t have the time or inclination to take on the rôle of tone police but I will move some off-topic comments to the sandbox if Joe would like me to.
You are invited to start a thread of your own if you think an off-topic could be a topic, otherwise please use the sandbox or do as I often do, start a comment and think better of it!
Hi to Dave L. I wonder if we shouldn’t have a welcome thread or page where new commenters can introduce themselves.
I despair – Hoyle’s Fallacy has been batted around for years. Without a handle on the density distribution of ‘targets’ in the space – including the subspace of shorter strings – calculations of impossibility based upon the size of the combinatorial space are utterly meaningless.
Completely random synthetic proteins contain function at a surprising density, even given the necessarily limited means by which function is assayed in the lab. And alternative splices from the same exon set are differently functional, despite shuffling or omission of the various exon stretches in the variants. You can split proteins in the middle and join their current N and C ends, to make something new but still posssessing ‘function’, without changing a single amino acid. And I recall (but cannot find) a paper where a mushroom gene replaced the function of a sequence-related gene involved in mouse brain development. So it certainly does not appear that there is any strength to the ‘too-impossible-for-this-universe’ argument, based simply upon there being v^n permutations of v variants in strings of length n.
There also appears to be a complete unawareness of the chemical similarities of the various amino acids. Most acids can be substituted by several others in most positions without effect. Only a few residues are critical for any given function. If there is always a functional peptide a step away from any current one, on what basis must we insist that this is merely permissible local movement, and we are ‘really’ in a box (or a Cave)?
Gpuccio appeared to concede that the modularity of proteins was real – he retreated behind a conviction that those modules themselves were unexplained/inexplicable. Nonetheless, those motifs appear over and over throughout the proteome, in very different proteins of very different functions. This is the real combinatorial level, not acid by acid. Iteratively duplicate a simple 4-acid turn, for example, and you get a long helix, which illusorily becomes exponentially more ‘impossible’ the longer it gets, and the more acids exist in the wider combinatorial library. Then stick it inside another peptide, and the gee-whiz index goes off the scale.
So the challenge for the serious ID-er is to determine how ‘impossible’ the minimal functional peptide is (or rather was, in the organism in which it arose). He would no doubt insist that the burden of proof is with the ‘Darwinist’.
niwrad, in comment #21 at niwrad’s UD thread seems to acknowledge that the main use of the SPC is to model the Origin Of Life. I say “seems to”, since the original post left the matter totally unclear. The comment suggests this, but also fudges some.
Although the discussion here has concentrated on the OOL, my original post assumed that the SPC was modeling protein evolution after the OOL. If so, then its model builds in assumptions that argue for Special Creation, for total inability of proteins to evolve. And in that case experimental evolution studies of present proteins are very relevant.
Was I already off-topic in the original post? It depends on whether niwrad would apply the SPC to evolutionary change after the OOL. And on that niwrad is unclear.
It seems to be an article of faith that the OOL cannot have happened without protein – lots of it! In the extreme version, 20-acid protein synthesised in the modern manner, in a cell nearly as complex as the simplest modern cell. In that context, they must see the ‘core proteins’ of this system as essentially unevolvable.
Certainly, several commenters take the evolvability issue on, but I think there is a near-universal tendency to flip from one to the other in anti-evolutionary argumentation. niwrad’s comment #7 to timothya is curious. He appears not to appreciate the relevance to evolution of sequential shakes, returning everything to the bag as it were, vs taking a sequence and exploring its neighbourhood.
The simulation program seems to be a satisfactory incredulity generator. When the program is used as designed in attempting to build organic polymers, one can come away with an increased level of confidence that biopolymers can not be made this way. For a proponent of the concept of uncommon descent the program seems to support their argument in some way. For a proponent of common descent, the program is just an example of how biopolymers do not come to be. Or, an evolutionist is left with increased incredulity that biopolymers are produced by total random generation.
Nonetheless, random biopolymers can show a remarkable range of activity. I am fond, like a Watchtower-wielding JW, of shoving the following paper in the hands of passers-by: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0015364
The polymers were ‘designed’ for structure – an ability to fold in a fairly uniform way – but not function. The folding rule limits the greater space enormously. Nonetheless, the total space available to the patterning algorithm is still vast – enough to fill a sphere about 5 AU across – while the number they actually sampled would be less than a single bacterium’s worth of this space. Yet their sample contained functional sequences to replace 4 out of 27 genes tried in nutritionally deficient E coli mutants. There are 5000 other E coli genes they didn’t try, and 5-30,000 other proteins times 1-10 million species that may also have an analogue in this sample. If one can dip one’s thumb into random space in this way, even with a significant constraint on the algorithm, and pull plums out as a matter of routine, it argues against the belief that functional proteins are so widely scattered they are essentially ‘impossible’ by random means. They may well be widely scattered in places Life does not go – the universe of proteins that cannot consistently fold, for example.
I’ve asked KF many times what “biological process” that is happening when the “haystack the size of the cosmos” is being sampled.
So far he’s yet to acknowledge the question, never mind give an answer.
They dread actual biology.
In that same comment 21, darwiN spelled backwards also says:
As anybody who has actually observed the real world knows, this is just dead wrong.
For good reason; they have never taken a biology course. Nor have they ever taken physics or chemistry; despite their assertions.
Both that kairosfocus character and that bornagain77 character are pure bluff and bluster; and some of their followers over there at UD apparently think they are geniuses with vast knowledge of science. It’s all copy/paste without comprehension.
It’s a bizarre clown show over there at UD; but without the clown costumes, just scrawny nakedness. Funny in a sick sort of way.
Let’s turn down the temperature, folks.
It’s probably helpful to understand that within the creationist model, the origin of life and the origin of species was the exact same event, which means it was produced by the exact same method. Poof is poof is poof.
And as we’ve all seen, the origin of the universe itself is part of this same event. That’s why creationists always lump the big bang in with evolution. It’s all the same act of creation.
Well, we know that Wells got a PhD in biology, as did Kurt Wise. The last statistics I read said that of all creationists who enter and complete undergraduate biology degrees, 80% of them are STILL creationists at graduation. Education does not cure creationism, but this most emphatically doesn’t mean thay haven’t taken any biology courses.
Yeah, I am aware of that. It is indeed possible to get through a degree program and miss vital concepts.
If one’s subsequent work in the secular world doesn’t press one’s fundamental understanding of concepts to the point of having to really think them through and understand them, it is quite possible to engage in routine applications and technology without ever correcting one’s fundamental misconceptions.
Even more bothersome is their lack of awareness of the need to check their understanding of scientific concepts before going public – i.e., going outside their subculture – and declaring that a scientific concept (e.g., the second law of thermodynamics) conflicts with evolution. Even negative feedback doesn’t appear to trigger this sort of self checking.
Most of the people in research that I know – and have known over the years – feel a very strong obligation to submit their ideas and claims to testing and crosschecking; and they will do the work necessary to hone their understanding of a topic before making fools of themselves. In my own case, from a very early age, I was not allowed to get away with bullshitting; and many of the situations I worked in would never allow self-deception without serious consequences.
I suspect that the internal social pressures and dynamics of some of these sectarian movements, especially the YECs, are responsible for this lack of self-awareness and caution.
I have to admit however that I don’t understand their lack of embarrassment. There is at least one thing I have noticed that seems to keep them on this track; and that appears to be some intense need they have to be some kind of authority figure in their subculture. Nobody in that subculture ever questions their knowledge, so the more letters after their names they have, the more authoritative they are.
Anyway; enough of this. Back to the topic.
But not too much. That would make the evolution of proteins impossible.
The title of your OP poses the following question:
Is evolution of proteins impossible?
But you never define the essence of a protein.
So how can anyone say?
gpuccio seems to acknowledge that some proteins could have evolved from other proteins.
But how do we decide the question of which protein could have evolved from another protein?
Yet that is not a problem when ID supporters claim that the evolution of proteins *is* impossible.
Double standard much?
Actual work. Something that, typically, ID supporters don’t know anything about.
“Nonetheless, random biopolymers can show a remarkable range of activity.”
Yes, they can. 🙂
My point about the Sequences Probability Calculator is that it seems useful in generating doubt/incredulity about naturalistic process even though it doesn’t show biopolymers cant evolve. It is doubt that can be used as a tool try to demonstrate there is a ‘controversy’… as in the “teach the controversy wedge strategy” That’s why I labeled it an Incredulity Generator.
I know it’s not the same thing but it reminds me of the video clip of D. Berlinski in a wing-back chair pontificating on his presumed 50,000 changes it would take to convert a cow into a whale and the near infinite improbability required to do so – an Incredulity Generator.
One can simply note that the questions Mung has raised — what the essence of a protein is, and how we decide which proteins could have evolved from other proteins, are not discussed by “niwrad”; when niward made the Sequences Probability Calculator the first question was not addressed at all. The second was, but only implicitly: in setting up the SPC it was assumed that no protein can change in any way without loss of its function. The experimental evidence on protein experimental evolution says otherwise.
I am going to police this discussion to avoid troll-like diversions: they will go to Guano or Sandbox.
This is not an easy topic to stay on. After all, either proteins evolved, or they happened by magic. The SPC’s purpose is to stack the deck so that magic is the only possibility. This can be either accepted or rejected a priori, based on personal preference, or it can be subjected to experimental investigation. Investigation indicates that proteins evolve, so resorting to magical explanations is not required.
Of course if magic is assumed and biological alternatives are disallowed (what the SPC was written to do), no experimental evidence means anything anyway, because its relevance has been assumed away. So both sides can start with mutually exclusive non-negotiable assumptions and talk past one another.
In a sense, what the “essence” of a protein is really is the question.
A typical enzyme has a small active site region, which often is highly conserved in evolution. In the three-dimensional structure, a bunch of the amino acids sick into interior of the protein and interact with each other to hold the sttucture together. These are somewhat conserved. A modest number of other amino acids are near the active site and interact with it and with chemicals that encounter the active site.
Beyond these there are general external amino acids that stick out into the nearby water. And there are quite a few regions of the protein that make alpha helices and beta sheets. These latter two types of secondary structures have amino acids that are less conserved. There are lots of molecular evolution data that document these, going back into the 1960s. The conservation is of changes between species as well as variations that occur within species.
More recent experimental evolution evidence shows that proteins from the same gene that are not identical in sequence can often be moved from one species to another without their function being noticeably changed. The fitness of the resulting organism may be a bit lower, but not enough to be easily measurable in the lab. The organisms are certainly not killed by this.
All of which shows that the Sequences Probability Calculator is not modeling actual evolution of proteins after the Origin Of Life, whatever else it may be doing. And “niwrad” has left it unclear what it was intended to model.
The essence of a protein is that it is a polymer comprised of amino acids. These have a common structure, comprising a central carbon atom, a NH2 group (the amino bit) and a carboxyl group (the acid bit), plus a side chain that confers different properties upon the 20 acids commonly found in nature. The amino group of one can be peptide bonded to the acid group of another, in chains of a few to hundreds of acids in length. They fold into structures that frequently have catalytic activity (though that is by no means their only role).
Since proteins frequently share a common sequence, with or withour minor variants, and proteins are (in modern organisms) produced by DNA genes that are replicated with reasonable fidelity, the hypothesis would be that (precisely because of the increasing ‘unlikelihood’ of randomly happening upon the same sequence) their commonality is due to common descent, and their differences due to that ‘reasonable’ caveat.
It could be due to common design, but then anything could. Common descent is less ad hoc, in that it relates to a process observable now and reasonably inferred to be in operation throughout the history of the DNA-protein system. Protein phylogenies map very well upon phylogenies constructed by all manner of independent variables, lending support. But ultimately, over ‘deep time’, the commonality degrades, and it becomes harder to tell.
Some of the ID/creationists are actually articulating their misconceptions. Here is tjguy over at UD.
He clearly doesn’t understand the implications of what he is saying.
I guess I should expect to dissolve any minute.
Sure, don’t you know what happened to the Wicked Witch of the West? Beware of buckets of water …
Also incorrect. He just boiled away for a few days. He had no apparatus for ‘isolating’ the acids as they formed. His samples have also been re-examined, and several more acids detected by the more sophisticated techniques of the day. They didn’t just ‘react with other things and disintegrate’ in the interim.
You also find amino acids on meteorites – and not as earthly contaminants.
The ‘chirality problem’ isn’t a fundamental problem – unless you believe that
a) protein is essential for life
b) the only viable proteins in protein space are ones composed of left-handed acids.
The space of all biological proteins has an equal-sized ‘mirror space’ of all amino acids formed from D isomers, and a whole bunch of other spaces composed of mixtures. It so happens, in the modern world, that enzymes, and ribozymes, are ‘selectors’ of L amino acids, because enzymes and ribozymes are themselves stereospecific.
It is true that non-biogenic amino acids form racemic mixtures. But carbon typically forms asymmetric molecules, and the isomer, in many instances, is too different, merely through shape, to react. It takes too much retooling of binding and active sites to make the same catalyst work with both. Once processes fix on an isomer, and everything one consumes has fixed that way too, derivative pathways will inherit that ‘preference’.
Here are the California science standards.
Searching for the word convection shows a first appearance on Page 25; in other words, in the fifth grade vocabulary of how materials are translated and move around. Further examples occur throughout the entire document.
The people posting over at UD don’t even know about scientific concepts introduced in elementary school.
Energy cascades occur in nearly every environment as long as energy can flow out of the system. That guarantees that there are temperature gradients, convection, and annealing processes going on. A lightning flash produces a plasma column; but that plasma doesn’t continue existing forever. Energy flows out of the volume and new compounds – e.g., ozone – form.
The magma from volcanoes doesn’t remain a liquid forever. The water heated near thermal vents in the ocean moves by convection into cooler environments.
All this is summarize by a concept that ID/creationists have continually sneered at; namely that the Earth is an open system. They have never understood what that really means.
William J Murray over at UD is making the following assertion:
Not only is this a straw man argument; it displays yet again that fundamental ID/creationist lack of knowledge of basic high school level science.
Complex molecules actually do form in avalanches (energy cascades) but battleships don’t. I doubt that Murray can tell us the reason for the difference.
A battleship forming in an avalanche is unlikely for a good reason that can be determined scientifically. Where did the steel plates come from? Describe the processes and the energies involved. And once steel plates came to be, what are the energies of interaction between them?
It’s curious that some of those UD people actually come over to this site merely to throw feces yet never read anything that anyone here says.
That poor Mung character hasn’t yet asked any of the big gurus over there to do that scaling-up exercise of the energies of interaction at the atomic level to the corresponding energies of interaction between kilogram masses at the macroscopic level.
Oh; and what are the quantum rules for steel plates snapping together?