last modified: Thursday, 03-Oct-2013 22:04:25 CEST

Document status: usable

1488x1088 247kb - 2976x2176 670kb
A gwindel (pronounced "gvin-dell") is a very odd growth form whose formation is still not fully understood. In a gwindel the crystal appears to have grown "sideways", roughly parallel to its c-axis, and along an a-axis, to form a platy crystal like a faden quartz, however not flat, but slightly twisted and with bent crystal faces. The term "Gwindel" is derived from the German "gewunden", which means helicoidal, twisted. Hence the name twisted quartz.

The first image shows a smoky quartz gwindel on matrix from the Furka pass area in Switzerland. Gwindels often show a peculiar and attractive "faceted" surface structure.




H264-movie, 256x256 px, 360 frames:
gwindel.mp4 556kb

Most gwindels look as if they were made of stacked crystals whose c-axes slowly rotate around a common a-axis, as shown in an idealized computer rendering (with a link to an H264-movie) of a so-called "open gwindel" (definition see below). The crystal faces usually indicate untwinned subindividuals of a equal handedness that determines the direction of the rotation: left-handed gwindels twist counter-clockwise, right-handed gwindels clockwise. Most gwindels show large triangular x-faces that can be used to determine the handedness of crystals. The angle of rotation is not fixed, some gwindels are twisted by an angle of a few degrees per centimeter, others are almost flat. There seems to be a fixed relationship between the twisting angle and the thickness of the crystal, with thin crystals being more twisted (Žorž, 1993).

1392x1136 191kb - 2784x2272 603kb
This image shows an almost ideally developed smoky quartz gwindel suitable to demonstrate some of the typical features. The crystal is from the Secunda Muota in Val Giuv, Graubünden, Switzerland, a classical gwindel location, and was found in October 2006.

The first thing to note is that the crystal is flattened and has an almost rectangular shape, and it could be mistaken for a faden quartz, but there is no "faden" in it. Not all gwindels look rectangular (the one in the first image on top of the page doesn't), but in an early stage of gwindel development they tend to. The first authors didn't know what to make of them, and some even took them for some other mineral.

The large central, roughly rectangular face is a prism face (m-face), to the left and to the right of it there are rhombohedral faces (r to the left and z to the right), and the shiny "roof" on top of the crystal is another prism face. The bottom of the gwindel has been attached to the host rock and shows no crystal faces. There is a triangular area with a mosaic-like structure at the top left corner of the large prism face, this is an array of x-faces. These x-faces can be found on most gwindels, and their presence is very helpful in understanding gwindel morphology.

Now that the faces have been identified, it should be obvious that this particular gwindel is a right-handed quartz, because the x-face is at the lower right corner of the large rhombohedral r-face (for the x-faces to be at the lower right corner, the crystal has to be turned by 90° so the r-face point upwards).

1072x1024 160kb - 2144x2048 468kb

1200x1024 180kb

This is an oblique view of the same gwindel, demonstrating the "twist" in it. If this was a car or a chair or some other familiar object that was twisted, one would immediately see that "something's wrong with it". But minerals are alien objects, so it is very difficult to capture the "twist" of a gwindel in a photograph, and the fact that the twist is very subtle in many cases doesn't help either.

In order to better visualize the twist, virtual crystallographic axes have been drawn into the photography in the next figure. The a-axis is shown in blue, while the c-axis is drawn red in the lower part of the crystal that was attached to the matrix.

If the crystal had just grown upwards along the a-axis, the c-axis in the upper part would be parallel to the c-axis in the lower part and run in the same direction, as depicted by the upper thin red axis. But although the c-axis maintains a right angle with a-axis, it is twisted around the a-axis in the upper part of the crystal. The actual direction of the c-axis in the upper part of the crystal is depicted by a yellow line, and the arrows indicate the direction of the twist. The c-axis does not change the direction all of a sudden, but gradually with increasing distance from the base.

As explained above, this quartz crystal is right-handed, and it twists clockwise. The rule is as follows:

With increasing distance from the base
- right-handed gwindels twist clockwise.
- left-handed gwindels twist counter-clockwise.

1200x1056 186kb

This is a view roughly along the c-axis, if this was an ordinary crystal, it would be a top-view, because we look at the crystal's tip.

Here one can clearly see the twist in the crystal, as the left prism face peeks around the corner at the base of the gwindel and gradually disappears behind the edge of the left rhombohedral face in the upper portions of the crystal. For this to work, the prism face must be bent. In fact, all the faces of a gwindel (at that developmental stage, as explained below) are bent to some degree, but this is most obvious on the large prism faces.

Like in this crystal, the upper end of a gwindel is always formed by the edge between two prism faces. This, of course, is so because the a-axes of a quartz crystal run through the edges between the m-faces.

In the figure to the right of the photography all crystal faces except the prism faces have been tainted by different colors depending on the type of face (see legend below image). There is a large rhombohedral r-face on the right side and a smaller rhombohedral z-face on the left side, and the same pattern can be found at the opposite end of the crystal (not shown). The gradual variations in the reflections on the z-face show nicely that these faces are also slightly bent.

1568x1360 285kb
Finally,  a top view of the gwindel is shown (corresponding to a side view on an ordinary crystal).

One can see the clockwise twist, as the bright prism face peeks around the upper right corner of the prism faces, and you will note that the large right r-face is bent, too. The most peculiar feature is the edge between the prism faces, though: it is rotated with respect to the c-axis in clockwise direction. So this edge is even more twisted around the a-axis than the c-axis, and both prism faces are distorted.

Below you see another version of that image with the faces being color-coded as in the preceding figure. In most gwindels there are no x-faces on the upper edge between the prism faces. There are exceptions, though, and one should keep in mind that if a gwindel developed as a "floater" one would expect to see x-faces along the opposite edge.

In this particular gwindel most x-faces are broken up into a mosaic of x- and m-faces, but in other gwindels the x-faces can merge into a large, slightly bent face. In rare cases the x-faces may be missing.

You might have noticed that the legend for the color-coding of the crystal faces includes an "s"-face (in green). There are two tiny crystal faces that are only well visible in the large version of the lower figure, and which are situated between a small r-face and an m-face at both ends of the crystal. Such faces can occasionally be found on gwindels and are an interesting subject by themselves. If this was an ordinary crystal, the position and the shape of the faces would clearly suggest that these are s-faces. s-faces are not as common as x-faces on crystals with macromosaic structure, but they can be seen occasionally as small elongated rhombs, usually neighboring to x-faces (or other trapezohedral crystal forms, like u or y). So why are there quotation marks around the "s" in the legend?

The position of the x-faces indicates a right-handed quartz, and if the crystal is untwinned, an s-face would be expected to be found at a different position, adjacent to the x-faces. On the other hand, if this face was truly an s-face, it would indicate some kind of twinning. In that case, there are two options: One, it could be a Dauphiné law twin. However, this could only be a very local twin domain around the "s"-face: the overall morphology clearly indicates the presence of well differentiated r- and z-faces, whereas in a Dauphiné law twin with roughly equal proportions of twin domains r- and z-faces are indistinguishable because r- and z-domains share the same crystal faces. The second option: it could be a Brazil law twin, which would be in line with the relative sizes of the rhombohedral faces, but, since the handedness of the crystal determines the direction of the twist and Brazil law twins are composed of right- and left-handed subindividuals, would interfere with twisting. So far, only small localized Dauphiné law twinning domains have been found in gwindels (Stalder, unpublished results mentioned in Rykart, 1995). There is yet another explanation for the faces: perhaps they are not s-faces that correspond to a positive trigonal bipyramidal form, but faces that correspond to the rare negative trigonal bipyramidal form. This would save one a lot of trouble around twinning, but admittedly sounds a bit exotic.

To summarize, the face looks like an s-face at a first glimpse, but labeling it so might have confusing consequences. I do not yet have sufficient data to come to a satisfying conclusion myself.


Types of Gwindels

Three types of gwindels are distinguished:
In closed gwindels the rhombohedral faces have merged into large bent crystal faces at both ends. Usually the prism faces also merge into single bent crystal faces. Sometimes closed gwindels develop an almost rectangular shape. The French expression for them is sucre.
Open gwindels show clearly distinguishable subindividuals with well developed individual rhombohedral faces and visible steps separating individual prism faces. Because they resemble a comb, in French these are called peigne.
Half open gwindels are intermediates between open and closed forms.

Closed gwindels are usually small, open gwindels relatively large. Thus it has been suggested that these types correspond to different developmental stages, with the open gwindel being the final step. Laemmlein (1937) suggested two basic developmental stages, with the closed gwindels corresponding to the early stage and the open gwindel to the late stage. X-ray studies on gwindels confirmed that the initial stage is a closed, single twisted crystal with a smooth, bent surface (Kuzmina et al., 1987). The initial gwindel has a homogeneous interior structure as opposed to the macromosaic structure that is visible on the surface of the final crystal and that is also present in the normal habit crystals that accompany gwindels (Dolino and Bastie, 2006, Dolino and Bastie, 2009). Cathodoluminiscence images of gwindel slices seem to confirm this - they show a gradual transition from closed to open forms during growth (unpublished results by Vollenweider and Stalder, 1990).

1392x1136 193kb

1392x1136 191kb
Many smoky quartz gwindels show a zone of faint color running obliquely through the center of the crystal. It is difficult to see, and I have marked the bright zone blue in the upper image and added the original for comparison.

This zone likely corresponds to the initial twisted crystal that lacks macromosaic structure.

Apparently open gwindels can become more "closed" again when they continue to grow, assuming a rhomb-like shape, and finally they may even look like ordinary, just slightly bent elongated crystals. Often these crystals still show steps on the prism faces and mosaic-like x-faces.



Gwindels are quite rare and can be found only at a few locations, mostly in igneous rocks, gneisses and sometimes in shists, the best coming from the central Alps (Aare Massive) in Switzerland and the Mont Blanc Massive (France and Italy). Other locations include the Hohe Tauern in Austria, the Polar Urals, Nepal, the Gamsberg mountain in Namibia, and Brazil. Single finds have been reported from Greece, Macedonia, and Bosnia. Gwindels are -still- unknown from North America and Australia. For an overview of locations, see Moore, 2007 and Žorž, 2009.

At all these locations gwindels are found in alpine-type fissures. Slow growth of the crystals seems to be a prerequisite for gwindels to form. Gwindels are typical examples of crystals with a so-called macromosaic structure and are accompanied by crystals of normal habit that also show this type of internal structure. Gwindels apparently do not occur together with crystals of Tessin habit or transitional habit, even if these are macromosaic crystals. The x-faces commonly found on gwindels are also indicative of slow crystal growth.

Gwindels are always accompanied by "ordinarily grown" quartz crystals. Some people claim that while these quartz crystals point in more or less random directions, gwindels are oriented parallel to the host rock. This is not true, however, one can find gwindels that sit on the host rock obliquely. But of course a gwindel will be more easily recognized and will also have better growth conditions if it is oriented in parallel to the host rock. Gwindels can be found on all walls of a fissure, not only the roof, as sometimes is claimed. Gwindels can not only be found attached to host rock, but also to other quartz crystals, in the latter case apparently intergrown at random orientations, which makes their formation even more miraculous.

Gwindels are not generally larger than the other normal habit crystals that grew with them in a fissure, often they are a bit smaller.


Theories on Gwindel Formation

There is no "faden" inside a gwindel that might explain its flat shape, and so far there is no other evidence of an external cause of gwindel formation. It seems to be an inherent property of the initial quartz crystal that causes material to accumulate primarily along an a-axis, and with a "twist".

Similar growth forms and twisting of crystals around an axis can also be seen in other minerals, so what is so enigmatic about gwindels? Let us compare gwindels with so called iron roses, for example. Hematite can sometimes be found as flower-like aggregates of platy crystals; these roses consist of intergrown crystals that are twisted with respect to each other. In a sense, iron roses are the "open gwindels" of hematite. One very irritating difference is that if you open a pocket with iron roses, all hematite in it will show that shape. Not all of the crystal aggregates will be developed perfectly, but they all show the same growth form, because that is the growth form hematite assumes at specific environmental conditions. But if you open a pocket that contains quartz gwindels, only a relatively low percentage of the quartz crystals in it will actually be gwindels, most of them will be "ordinary" prismatic quartz crystals. Even at famous gwindel locations, gwindels are a rare find. A very remarkable exception was a large "Gwindelkluft" at the Scheuchzerhorn in the Grimsel area that Swiss strahlers Beat Teige and Alexander Willener had been working on between 1993 and 1994. A piece of the fissure wall that was reconstructed from several pieces (2.78 m long, weighing 217kg) carried 157 "ordinary" smoky crystals, 138 open and 15 closed gwindels (Teige and Willener, 1996).

One might guess from this that the formation of a gwindel is a most delicate process that is easily disturbed, but this is probably not true either: many of the gwindels presented here contain large amounts of inclusions, mostly chlorite, but although this led to some interference with crystal growth, it didn't disturb the gwindel growth. One should also keep in mind that gwindels appear in environments of exceptionally slow crystal growth, and it took the crystals millions of years to grow.

It has been suggested (for example, by Žorž, 1993), that piezoelectricity might play a crucial role in gwindel formation. Upon mechanical deformation quartz gets electrically polarized along its a-axes and the surface of the crystals assumes opposite charges of several thousand volts at both ends of the a-axes. Even a charge difference of just a few millivolts might well promote or hinder the accumulation of silica during growth, either by directly attracting or repelling orthosilicic acid molecules, or by attracting other ions dissolved in the watery solution that attach to the surface, or by effecting the condensation reactions at the surface. The first mechanism (as proposed by Žorž) would explain the increased accumulation, the others might help understanding the induction of regular disturbances in crystal growth that lead to a twist.

So far, there is no direct empirical evidence that supports the piezoelectricity theory. In fact, it leads to more questions that need to be taken into account for a satisfactory explanation. For example, if some deformation (internally or externally caused) has lead to an electrical polarization of a crystal, it is hard to understand what maintains that charge distribution over a long period of time. Under ideal conditions, a polarized crystal will keep its charge for a few hours, but this is very unlikely to happen in the salty brine that the crystals grow in over a period of hundreds of thousands and even millions of years. So one needs a dynamo that recharges the crystals, and that acts specifically on the gwindels already present in the fissure and not on the other crystals. Another problem is the geometry of the charge distribution on the crystals: there is not just one a-axis, there are three of them and accordingly the charge distribution on the surface assumes a trigonal pattern, so the crystals should tend to have a trigonal or Muzo habit.

This is not to say that piezoelectricity does not play a role. But there is apparently still no conclusive and generally accepted theory of the mechanism that is responsible for the selective development of specific shapes of certain crystals, and gwindels remain puzzling.



To demonstrate that they do not necessarily grow perpendicular to the host rock, all gwindels in the following images have been placed in the orientation they presumably had when they were attached to the host rock.

1360x1004 191kb - 2720x2008 529kb
A  clear cinnamon-colored open gwindel, again from the Secunda Muota, Val Giuv. In open gwindels the crystal faces tend to develop a parquet-like structure that causes the gwindel to sparkle more than ordinary quartz crystals. The gwindel likely grew on the host rock at an oblique angle.

1024x880 142kb - 2048x1760 474kb
A  small closed smoky quartz gwindel from the Secunda Muota, Val Giuv, Graubünden, Switzerland. A scar that runs vertically over the central crystal faces (probably caused by growth inhibiting calcite crystals) makes it look a bit like a Japanese twin. The gwindel shows a peculiar surface structure, as if it had been shattered and glued again. This reflects both the internal macromosaic structure of the crystal and the steps introduced by the adjacent crystal subindividuals that are slightly rotated relative to each other.

First it seems hard to name the different faces on the crystal. But the central face is striated, so it is an m-face, and as such a striation always runs parallel to the rhombohedral faces, the upper right shiny face and the lower left face must be either r- or z-faces. Which one is what? The dark, roughened trapezohedral face at the bottom is a large x-face, so the lower left face next to it must be an r-face, and the upper right shiny face is a z-face. The x-face is also indicating that this is a left-handed quartz. The rest is easy: the shiny upper left and the small bottom right faces are m-faces, and the dark horizontal top face and the dark face at the right end are r-faces.

880x800 126kb - 1760x1600 429kb
This  is another closed gwindel from the Secunda Muota with an almost square shape. The left side of it could already be considered "half-open", as one can see individual rhombohedral faces. The shiny right rhombohedral face shows the typical "shattered" surface structure. Inside the smoky crystal one can see irregular inclusions of pocket clay and greenish chlorite.

832x1000 157kb - 1664x2000 557kb
An  open smoky quartz gwindel from the Giuvstöckli, Val Giuv, that has grown in the "classical" upright position. The rhombohedral faces on both sides are covered with chlorite. The gwindel is almost 7 cm high and the orientation of the c-axes at the top and bottom differ by about 25°, which gives a twist of 3.6° per centimeter, an unusually large value.

1024x1000 208kb - 2048x2000 640kb
Another  closed smoky quartz gwindel from the Secunda Muota. The entire crystal is covered by green chlorite, giving it a silky shine. Note the very large x-face, the dark triangular face that almost reaches the r-face at the opposite crystal tip.

1168x1064 232kb - 2336x2128 675kb
This  clear but chlorite-covered gwindel is closed on the left and half-open on the right side. On the front there is a large triangular x-face that borders another x-face at its upper edge. Normally x-faces remain well separated and don't touch each other, but in gwindels adjacent x-faces can occasionally be found. It is from the Scheuchzerhorn in the Grimsel area of the Aar massive, Kanton Bern, Switzerland.

1360x1048 200kb - 2720x2096 593kb
Gwindels  may assume a rhomb-like shape. This is sometimes interpreted as a continuation from the open gwindel form, and usually they also show a mosaic of m- and x-faces, like this smoky quartz gwindel: there are numerous triangular and shiny x-faces and m-faces that can be recognized by their vertical striation. It is also a nice example for a crystal with macromosaic structure that is common to gwindels. Not only does the frontal m-face look a bit as if it had cracked and was glued together again, the large shiny top left rhombohedral face also has a "shattered" surface. The specimen comes from the Galenstock mountain north of the Furka pass in the Grimsel area.

1472x1024 246kb - 2944x2048 731kb
Another  rhomb-like gwindel from the Scheuchzerhorn in the Grimsel area. Note the mosaic of chlorite-covered x-faces in the upper left and of shiny m-faces in the lower right part of the crystal.

1224x1064 180kb - 2448x2128 516kb
A small closed gwindel from the Scheuchzerhorn in the Grimsel area in Switzerland. Although it is a flat crystal, it looks very unsuspicious, only the position of its x-face and the slightly bent m-face reveal its true nature. The Alpine-type cleft the gwindel came from is described in an article of the Schweizer Strahler (Teige and Willener, 1996).

Printer Friendly Version

Copyright © 2005-2013, A.C. A k h a v a n

Impressum    -    Source: