Ah, OK. It sounds like what you really mean is that you can't see how stepwise evolution could give rise to creatures that need several parts to function. If you'd have asked about this at the start, we could have avoided a lot of wasted time and posts.
The key concept here is the evolution of multicellularity - this occurred in conditions where undifferentiated clumps of cells were more successful in survival and reproduction than single cells (clumping could be a result of a change to the cell membrane that made cells more likely to stick together after division).
The major advantage of multicellularity is functional specialisation through differentiation - cells becoming specialised for certain roles makes for more efficiency overall.
Here's a hypothetical: suppose a cell in the clump of cells has a mutation that causes it to produce a protein that makes food particles stick to its surface, or an enzyme that makes it easier to absorb food particles; that would be an advantage, so that trait would be likely to spread - a clump whose cells all had the new gene would have a huge advantage. Such a clump could absorb more food and grow bigger than the others, but its size would be limited because cells too far from the surface would be starved of energy from nutrients, especially as they would also be expending energy producing this protein that was only effective at the surface.
So there would be a selection pressure for the cells to develop differently at or near the surface vs inside the clump, so the cells inside the clump didn't waste energy producing this protein. Suppose a mechanism evolves (a regulator gene) that can suppress a gene if the concentration of some other chemical in the vicinity is high. Inside the clump, where all the cells are surrounded on all sides by other cells, the concentration of the various chemicals they produce is higher than on the surface, where those chemicals are constantly washed away; so it won't be long (in evolutionary terms) before the novel food gene in the cells inside the clump is suppressed by the regulator gene, which gives them an advantage - more energy, so they can grow larger, or whatever.
Now the inside cells have become functionally different from the surface cells, and this difference continues to increase, as the regulator evolves to become able to effectively switch whole segments of the genome on or off - the outside cells become specialised for food absorption and protection, and the inside cells specialise in converting the nutrients to products that all the cells need. The surface cells stop wasting energy making products the inside cells can make for them, and the inside cells, protected and supplied with nutrients, become increasingly fragile and unable to engulf food for themselves. The two types of cells work together far more effectively than the original single cells, but neither can survive alone.
That is a simple form of stepwise evolution of functionally separate but co-dependent parts.
The gene regulator mechanism which activates or suppresses genes according to chemical gradients that indicate where the cell is in the organism, is a key development - it gets copied, modified, extended, and eventually becomes part of a complex cascade of regulation that controls gene expression to enable the organisms to evolve advantageous shapes by distinguishing axial (anterior-posterior), lateral (left-right), and dorsal-ventral axes, and to produce consistent proximal-distal limb development.
That was just a hypothetical example, far too simple to be real; but it's an echo of what did happen early in evolution - similar gene regulators have been found in all
bilaterians, they're known as
Hox genes, and they are almost identical in all creatures - there's no significant difference between fruit fly Hox genes and human Hox genes, and if you replace a fly Hox gene with a chicken Hox gene, it will develop normally. Mess with the Hox genes or the chemicals that trigger them, and you'll mess up the creature's body plan, cause it to grow limbs in the wrong place, etc.