Looking over your work, it's very clear that you've put a great deal of time into this, and you've directed your focus in the right general direction - that is, you've spent a lot of your energy focusing on using cast shadows to construct your gradients, keeping the focus largely on how your textures exist in 3D space, and deriving shadow shapes from the relationships between the different forms and surfaces that are present.

While I think that the way you've gone about it applies effectively to the exercise itself and its goals, I do think that there are some areas where you may have interpreted more complexity than is necessarily present or necessary from your reference images, and in a sense have made some of these textures considerably more difficult or time consuming than was necessary, and in so doing, didn't always capture the textures for what they were. Again - for the purposes of the exercise, it really isn't a problem (assuming your arm hasn't fallen off in the process, or you haven't been put on an involuntary hold at a psychiatric ward, 'cause really.. good lord no one can ever accuse you of not being patient, disciplined, and above all resilient) - but it is something I want to talk a bit more about for the purposes of your understanding of texture as a whole.

Ultimately textures will range naturally from those that are rough, consisting of many smaller individual forms, to those that are smooth, where textural forms may be entirely absent at the greatest extreme, or are simply larger and less numerous across the surface of the object. When it comes down to capturing a texture on the page, it is important that we consider where along that spectrum the texture lies, so that we do not accidentally communicate to the viewer that a surface is very rough, when in fact the intent is for it to be smooth.

A good example of this is the turtle shell texture from early on in the set. Turtle shells are quite smooth, with the major forms being the big interlocking pieces. While these pieces may have ridges to them as well, the surfaces between them are still very close to being co-planar, resulting in no substantial cast shadow between them, so our cast shadows are still going to come from the major forms. Here in this diagram I've identified the major shells on top of the reference image, and then analyzed it from a side view below, the top analysis being with the curve of the shell (which we'd remove for the purposes of our gradient since we're applying everything to a flat surface), and the bottom analysis being without that structural curve. As we can see here, we still don't end up with a lot of jumping between light and dark, but we do see a steady progression as we move from right to left away from the light source, with the areas of light getting smaller and the areas of shadow getting larger.

The thing to keep in mind here is that for the purposes of our gradient, this is still enough. We don't have to make everything very noisy with a lot of juxtaposition of black/white. As long as the progression is consistent - even if the groupings of light and shadow are larger, it is still valid. It is also worth noting that we would get a little more complexity from the fact that these textural forms are staggered when viewed from the top, but as a whole this texture would have to be considerably less complex than how you approached it in order to really convey that it is appropriately smooth to the touch.

To that end, starting by imagining yourself running a hand over that surface is a good way to determine whether you should be looking for more, smaller textural forms (in the case that the reference seems like it'd be rough), or whether you should really be focusing on larger groupings instead (in the case that it seems like it'd be smooth).

An example where you approached this more correctly would be your corn texture, although I can see that you did attempt to sneak in some extra complexity by relying on form shading in this area, which falls outside of what this exercise uses to derive its areas of solid black.

Anyway, as a whole I think you've definitely demonstrated a very high degree of patience and focus throughout the challenge, and where you went a bit awry was where you put far more labour into a given texture than was required. That is an issue that is much more easily remedied than the opposite, so I'm not at all worried about those issues.

I'll go ahead and mark this challenge as complete.