Fertility vs. Color of Soil

Neither.

More detail required.

The dye itself may contain chemicals which are either advantageous or damaging to the crop - no comment.

if the dye affects the water holding effect of the soil it also may have an effect as to reduce or increase availability of water - no comment.

More detail please.

I am doing experiment about crystal soil.

I am choosing different colors of crystal soil growing the same plant.

Of course, they soaked with the same fertile solution first.

Anyone has conduct a similar experiment before?

Your experimental design is flawed. To learn something positive from an experiment, you should control all variables except one.

In your present experiment, the use of dye to test for the effects of soil colour/reflection is a confounding variable, because dye has chemical properties as well which could affect the plant. The experiment will not show whether differences are due to the colour or due to the chemical difference produced by the dye.

A better test of soil colour effects would be to use identical mulches of a different colour on the soil surface, which have a stable composition (not leaching free dye). It could be done with differently coloured natural (undyed) stone pebbles, for example.

for the most part,..... yes

The color, by itself, does not contribute to fertility vis-a-vis nutrients...Now it could be argued that a darker soil retains more heat and warms seeds so that germination takes place quicker than a similar seed in a lighter colored soil, but that's dubious...The amount of light a plant receives does contribute to growth, but too much and the plant can be cooked, and of course different plants react differently...so it would depend on the plant....

"which one is more fertile?"

Both the soil will exhibit equal fertility.

"Do the reflection and refraction affect growing of plant?"

Yes. It does affect growth.

As several have already noted, much more information is needed concerning where, when, in what weather, how and what is being grown.

.

There is another problem though. The inferred premise is flawed.

.

It isn't how dark or light something is colored that dictates if it is gaining heat or not. If you have ever walked barefoot on very hot white sand, or some white concrete, then you know this to be true.

.

Whether or not solar energy is adding heat to a surface (in the near term, and ignoring convective and conductive heat transfer) depends on the how the radiative emissivity of the surface compares to the radiative absorptivity of that surface.

.

There are surfaces that reflect a large percentage of the incident spectrum, but compared to the small portion that is absorbed, even less is emitted. Aluminum foil under a heat lamp is a good example of this.