What do you do when you're out in the field and you really really wish you could positively identify a particular mineral? There are ways of doing this which are no longer taught - involving powders and torches and things like that - I never learned these methods, these days we just take the rock home and make a thin section.
We were in northern Mozambique looking at what appear to be charnockitized gneisses - which should by definition contain pyroxene - normally orthopyroxene. However, the thin sections we made last year just had clumpy biotite - maybe replacing opx? But without finding any remnant opx we couldn't be sure if it had ever been there, or if all the patchiness we could see in the rock was just clumpy biotite. We crushed up some of the gneisses, which wasn't too hard to do because they have a nice "sugary", recrystallized annealed texture and readily fall into angular "sand". Some of the grains had a greenish smokey luster - possible orthopyroxene? Or just ugly looking feldspars? Or quartz with some micro inclusions? If only we could see the birefringence of the mineral grains we could readily pick out the opx - which has higher birefringence - from the feldspar and quartz, which have similar, low birefringence.
Birefringence is the effect when light passes through a crystal more quickly in one direction than in another, effectively splitting light like a prism. The strength of this effect varies with the wavelength of light as well as with the properties of the crystal. The birefringence of different minerals is a characteristic feature that can be used to identify them. To see this effect, geologic microscopes use two polarizing filters. Full-spectrum light is passed through a polarizing filter and then through a thin section of rock (30-microns is the usual thickness for this slide). Then the light passes through a second, polarizing filter perpendicular to the first one and through to your eye.
Two polarizing filters at right angles to each other will block all light - so if there is no rock slide in the path of the light, or if there is just glass or some other isotropic material in the path of the light, you will see nothing at all. However, a birefringent rock sample will take the oriented light and twist it - allowing some of it to pass through the second filter. The colour you see will change depending on how strongly the mineral "twists" the rays of light.
So what to do in the field with no slide, no microscope, no way to estimate the birefringence of these mineral grains? Kosuke came up with an answer - like geologist MacGyver. He used the polarizing filters from his camera lenses - mounted them on an incandescent head lamp (torch) and put the mineral grains in between. He then carefully rotated the top filter so that it blocked out all light except for what was passing through the mineral grains. Voila! A field petrographic microscope!
Did it work? Sadly, not really. The problem was, of course, that all the mineral grains had different diameters and therefore the path of light through the minerals wasn't constant. As the colour you see is a function of the length of path as well as the properties of the crystal, we couldn't keep this factor constant and were left with uncertain results. However, it was a pretty good exercise (and a lot of fun) and a reminder of the basic principles of geology which we sometimes take for granted.
You know what? Identifying rocks is pretty hard. Very few minerals come in only one color or shape, and a little bit of weathering or deformation can change the appearance of a rock pretty drastically. Geologists and geology students have to learn a whole range of criteria which change from place to place and rock to rock. It's a lifetime effort to learn to identify rocks, accelerated by seeing as many rocks as one possibly can but one can never hope to see them all.
In accretionary wedges - my "specialty" (if I have one, that is) - low temperature metamorphism and high strain can result in all the rocks looking pretty much the same - it can be very difficult to tell an igneous rock (eg. basaltic ash from the ocean floor) from a sedimentary one (e.g. volcaniclastic greywacke). It's hard to tell students that the first basic thing we tell them about rocks - that there are 3 types (igneous, metamorphic and sedimentary) might be the last thing you figure out about a particular rock sample. Some day when I have the time I'd like to teach myself some of those old techniques for identifying minerals without a microscope - I'm sure they would come in handy.