Buy a stepwedge.
Well, I think I'm opening a big can of worms here but...
The push-the-darkslide approach will work if you use the right interval, but I think it is a pain compared to other methods. First calculate a zone 1 (-4 stop) exposure from your meter reading. Assume the meter reads 1/60s @f/8 @ manufacturer's ISO. That would be zone V if the speed is correct. From that example, zone I would be 1/250s @f/16. Pull the darkslide until only about 1/2 inch of the film is covered. Then expose for Zone I (1/250s @ f/16). That's 1 exposure unit. Push the darkslide in 1/2 inch and expose again for 1 exposure unit at 1/250s f/16. That makes a 1-exposure unit stripe plus a 2-exposure unit stripe. Push the slide some more and add 2 more 1-exposure unit exposures (e.g., 2 at 1/250s f/16 or 1 at 1/125s @ f/16). Now you have 1, 2, and 4-exposure unit stripes. Push and add a 4-exposure unit exposure (e.g. 1/60s @ f/16). Push and add an 8-exposure unit exposure. You will now have stripes that represent zone O, I, II, III, IV & V with cumulative exposures of 0, 1, 2, 4, 8, & 16 exposure units, respectively.
Do the same thing on another sheet but use the zone V exposure as the starting point instead of zone I. Produce exposures of 16 (zone V), 32 (zone VI), 64 (zone VII), 128 (zone VIII), 256 (zone IX) and 512 (zone X) exposure units on the stripes and you'll have all the zones on two sheets of film. Assuming you guess and get the film speed and development close, you can make some judgement about tweaking things and then repeat the test for different developments and/or EIs.
Yuck. What a pain. The holes in darkslide method is easier and quicker.
Here's a much better way from John Schaeffer in his Ansel Adams Basic Guide to Photography book 2:
Schaeffer's method uses an inexpensive Stouffer's 21-step transmission density tablet (#T2115) taped over a piece of 4x5 film loaded into a regular holder. The step tablet increases by ~0.15 density units/step from a value of approximately 0.05 for step 1 through ~3.00 for step 21.
The target is a blank wall evenly illuminated, the light reflected from it measured, and the film given a single Zone X (+5 stops) exposure. The step tablet reduces the exposure on the film under the various steps about 0.15 density units/step. (That is equivalent to 1/2-stop per step). The film thus records all exposures simultaneously from around Zone X to Zone O. The film gets a Zone X exposure from the area not covered by the stepwedge. As the density increases on the wedge, the amount of light to the film underneath is reduced until it gets no exposure (Zone O). This method eliminates the effects of multiple, intermittent exposures. It's all there, all at once.
Start by exposing and developing a sheet using your best guess as to the EI of the film (usually 1/2 the manufacturer's recommended ISO for me) and the recommended development time. After seeing the results from that sheet, expose another 3 sheets (or 5 depending on how accurate you wish to get) of film similarly but making an adjustment for the EI of each, and then develop them for different amounts. I usually can get a pretty good handle on the +/- and N developments in 4 sheets.
The adjusted EI sheet (#2) gets what I believe will be normal development after looking at the first sheet, how it contact prints, and any characteristic curve I may construct (to get the contrast index).
The #3 sheet is exposed to zone X using a reduced EI (IOW overexposed -1/3 stop from #2), and that sheet is developed less than #2. (I use HC-110 and vary the dilution rather than time).
The #4 sheet is exposed to zone X using a increased EI (IOW underexposed +1/3 stop from #2), and that sheet is developed more than #2.
(If I use 6 sheets, I give the other two -2/3 stop in EI coupled with an even greater development and +2/3 stop coupled with even lesser development.)
I then determine the contrast index (or one could use Ilford's G-bar measure) from the film curves and create a contrast index vs. development graph.
Use the Stouffer tablet to also determine the paper's proper proof time and exposure scale. Once the proper proof time is known, proofing the sheets of film gives a visual record of the density shifts with exposure and development. From that, and the fact that each step is about 0.15 density units different than its neighbor, one can determine all the zone system effects visually. This eliminates the need to use a densitometer. (Though a densitometer makes the whole thing a lot easier the first time around.)
Hopefully, this makes some sense. It has been a lot harder to explain than actually do. Schaeffer's book leads you through it very clearly.
The following pic shows the stepwedge being used to determine the paper exposure scale and proper proof exposure. On the left is a T2115 wedge taped to an unexposed but developed sheet of film. The exposure on paper to the extreme right was deliberately overexposed. The dark tones are blocked until about step 6 (which you probably can't discern on the scan). To get the proper exposure, the first step should be black and step#2 near black (and around zone I). So, the second sheet needed to be exposed to shift the black step down 5 steps to step #1. Since each step is 1/2-stop, the exposure needed to be reduced 2 1/2 stops. The first sheet was exposed for 60s @f/8 with a #2 filter. To move the black tone to step 1, the second sheet was exposed for 10.5 seconds @f/8 with the #2 filter. I found the time by multiplying the original exposure by 17.5%. (x.70 moves it 1 step = 1/2 stop, x.50 moves it 2 steps = 1 stop, x.35 moves it 3 steps = 1 1/2 stops, x .25 moves it 4 steps = 2 stops, and .175 moves it 5 steps = 2 1/2 stops.)
On the second print, an actual gray card is slightly darker than step 7. That tells me the density needed to produce middle gray on that paper with that filter is about .90 density units above fbf if I print for maximum black at the shortest exposure. (Step 7 would have a density of ~ .95 density units.)
White occurs at step 12 with a little tone at the threshold step 11. Step 11 has a density of 1.55 so I know the paper goes white slightly above that value (and considering fbf in the equation as well). So now I have max black, near black, middle gray, threshold gray and paper white figured out in terms of the negative density need to produce those tones. I also can get the Exposure Scale of the paper/filter. That would be equivalent to step 3 (90% black with density around .35) subtracted from step 11 (threshold gray at 1.55) giving an ES of 1.20 for this combo.
Of course I don't do any of this now since I took up wetplate collodion a couple years ago. I just look at the light and guess.
Joe