Crystallography of any substance is still a very mysterious subject in the chemistry / engineering community. We know many of the factors that DO affect crystallography, The material itself being the most important in determining crystal structure. Isothermal Compressibility, Heat Capacity, and thermal expansion coefficients are the "restrictors" on crystallization process.
The next biggest factor in determining solids packing is the solvent - different solvents provide different neighborhoods for crystals to agglomerate. Impurities in the crystal and solvent will also "force" a crystal in one particular structure.
The third most important factor in crystallization is heat (and mass) flux, or the rate the solution is cooling and evaporating. A good way to think about crystallization is consider the molecules phase change from aqueous / solid as a probability function. At certain temperatures and pressures (and solvent, solution concentration, impurity, etc) certain crystal structures are more "probable" than others.
If it was possible to create a perfectly valid equation of state for DMT (or any) liquid/solid, the necessary inputs such as temperature, pressure, heat flux, mass flux, concentration, could be inputted into the differential equations that govern our universe. Interestingly enough, the equations that govern mechanics, acoustics, thermodynamics are all in fact the same ones with different coefficients - solving the evaporation rate for a solution, one can simply apply a dimensional analysis to scale for the heat transfer without solving the equation again. If someone were able to do this, they could make a plot in the number of dimensions of uncertainty (P, T, X, Q, etc) and determine "regions" that converge to a certain crystal structure. Some of these regions overlap, it what is called an unstable solution of the EOS: at exactly that point, the system is stable: any fluctuation can send it toward a different, stable region. Because we can't hold a freeze precipitation completely constant, we may be operating in one of these regions, in which case you can see two different crystal structures near each other. I find this fascinating.
Sadly, the math and theory behind this is beyond our understanding and measurability. Some things, however, can be tested. If I had the equipment, I would do a massive extraction, spilt the solution into many different tanks, and precipitate them at different temperatures, pressures, solvents, and concentrations, and timeframes. These 5 properties are the biggest in determining crystal structure. It would be rather riveting to see a 5D phase diagram of DMT and its crystallography.
However, only a devoted community like us could manage this endeavor - modern science is still doing this for water, and I believe that have found 14 or so different packing structures for ice, and new ones are being discovered every year. We have still come a long way to determine what we know about forced airflow, heating, concentration, and temperature.
I believe that Crystallization is a form of art, and I applaud and thank you for your pics! Someday in our future we will see these hanging on walls in galleries, coffee shops, and malls; whether people know that they are looking at spice or not, isn't important, because those enlightened will
Edit: I forgot to say, this is my first post on the nexus; I am happy to be here and am thankful for the welcoming energy you all provide, an oasis in the deserts of the internet, as I view it! I have about a half dozen or so textbooks PDFs on Crystallization (applied and theory) if anyone is interested, as well as a library of about 2.5 GB of chemistry and engineering textbooks and .5 GB on psychedelics and entheogens. PM me if you have any interests!