FMF - Friends of Minerals Forum, discussion and message board

[Ru Smith]

Contemplating the attached fluorite twin, I think it's starting to make sense. All the interfingering domains with angular differences are confusing at first. I see now though that these are cube and octahedral faces of the two cube-octahedral members of the twin and, whereas the symmetry relationship between the two members is roughly a reflection across an octahedral face, there is no simple contact plane (as seen in spinel twins of spinel). The two members of the twin are interpenetrating in a complex way, so this seems to be a penetration twin rather than a simple contact twin. Is that right?

The second and fourth images show the interfingering twin domains on one of the two big octahedral faces (roughly hexagonal in shape due to the adjacent presence of three cube faces). You can also see the difference between the dihedral angles of a) two adjacent octahedral faces (109.47 degrees?) and b) adjacent octahedron and cube faces (125.264 degrees?).

I'd like to see some more images of spinel twins in fluorite. Do any of them have simple contact planes? Any ideas on what conditions promote the formation of spinel twins?

[mmauthner]

The twin composition surface is still parallel to the octahedral face and a plane, making this still a contact twin. Twin laws are by definition either interpenetrant or contact depending on the law's twin element which is either an axis or plane ( though 3D merohedral composition surfaces also exist).
Hope that helps.
Mark

[Ru Smith]

Thanks very much, Mark.

I was remembering Hurlbut's comments on mica twins (https://www.mineral-forum.com/message-board/files/mica_penetration_twins_203.jpeg) where he wrote, "However, these twins do not conform to that part of the law that states that the composition plane is (), if we consider composition plane in the usual sense, namely the plane on which the two individuals are united. The contact surface is highly irregular but roughly at right angles to (). It is a penetration rather than a contact twin."

This fluorite seems to show a very similar configuration. Indeed, in three domains shown in the last image the far element of the twin has extended completely through the element of the twin in the foreground. But, as you say, wherever there are flat portions of that highly irregular contact surface they are parallel to the octahedral face.

The wikipedia sentence on penetration twinning is as follows: "In penetration twins the individual crystals have the appearance of passing through each other in a symmetrical manner." In the case of this fluorite and Hurlbut's mica twins, the individual crystals do "pass through each other", but they don't do so "in a symmetrical manner", so I'm happy to accept your definition.

[Enrique Llorens]

Spinel twin of fluorite from Mexico.

[Ru Smith]

A very beautiful specimen, Enrique.

Can you see the contact between the two individuals in each twin? Is it a single plane, or is it a highly irregular surface with one individual "passing through" the other (as in the Erongo specimen)?

[Ru Smith]

Here's another complex fluorite to decipher (from the Houston show just now). What do you see?

[Ru Smith]

After looking at this for a while I can see rhombic faces of the dodecahedron, and bigger trapezohedral faces of the "trapezohedron" (icositetrahedron) which are truncated by cube faces (there's one abutting against the smoky quartz).

The contact between the two individual components of the upper twin (look for the re-entrants) is not a simple plane and "wraps around". Looking along one of the axes of 3-fold rotational symmetry it seems to me that a rotation of 60 degrees would overlay the faces.

So, this one is a penetration twin? Just like the well-known English penetration twins, but almost unrecognizable (to my eyes) as a result of the dominance of the "trapezohedron" (trapezoidal icositetrahedron) and dodecahedron.

I had never seen this before. Anyone else?

[Ru Smith]

Back to spinel twins, here's a galena showing the same juxtaposition of octahedron and cube faces as seen in the Erongo twin.

[Roger Warin]

Spinel twin in diamond.
Here is a diamond crystal twinned on {111} (spinel law).
Roger.

[John S. White]

Here is an interesting fluorite spinel twin from Dalnegorsk, Russia. It is 4 cm in the widest dimension. The bottom member of the twin is very transparent but the top member is not, which I find very unusual.

[Ru Smith]

Thanks, Roger. Your diamond is a nice example of a simple spinel twin where one individual is spun 180 degrees (just 60 also works) around [111] relative to the other on a (111) twin plane (or we just reflect across that plane). There's a side view of such a twin in spinel attached, which I like because you can see the reflection, but also easily imagine rotating the top part 60 degrees to reform the octahedron. I read somewhere that Be ions substitute for Mg ions along the (111) twin boundaries in these Mogok spinels, promoting exaggerated growth along the twin boundary.

So, are such twins, with a discrete twin plane between the individuals, ever seen in fluorite? or do fluorite spinel twins always show the complex intergrowth of the two individuals seen in the Erongo specimen above?

[Ru Smith]

That's a really interesting twin, John. Just octahedra this time, so a bit less mind-scrambling? What do you think is going on in this one? The two members were not (fully) coeval?

[Ru Smith]

.. though I see some cube faces too. Are the angles right? - are we sure this is a twin?

[Jesse Fisher]

I can not recall ever seeing a spinel-twinned fluorite having a simple contact plane as seen in twinned spinels. I always assumed that it was due to the presence of cube and/or dodecahedral faces on the fluorite, but we all know the old saying about assumptions. Would be interesting to know the cause.

Below are photos of a spinel twin fluorite from Pakistan, which shows this fairly nicely.

[Ru Smith]

Wonderful crystal, Jesse.

- and quite like the Erongo example in clearly showing the complex boundary between the two twin members.