On the basis of morphology, Haeckel's Gastraea Hypothesis seems to provide a reasonable pictures of how multicellular organisms evolved. However, at the genetic level there are serious obstacles.

In order to survive as living cells, the early ancestral cells needed a genotype capable of producing all the relevant proteins required to fulfil their physiological and structural needs. These early cells would have had genes coding for all the essential enzymes required to maintain the physiological processes and genes coding for all the necessary proteins involved in the structure or morphology of the cells. The probablility of a cell being formed by chance is incredibly minute, but for the sake of this argument, we will assume that such a cell did in fact arise.

Furthermore, it is not too difficult to imagine that a situation could have arisen where cells remained stuck together after cell division, thus resulting in multicellular colonies with the cells embedded in a common matrix. Problems arise, however, when the evolution of cell differentiation and eventual specialization are considered. If the colony arose through cell division, then each of the original colonial cells would have had the same genetic composition, coding for the simplest of cells.

Live Mouse Skin Cell Imaging https://www.flickr.com/photos/jonathanpalero/171301223/in/photolist-g8XTn...

The evolution of specialized cells requires that the different cells also evolve different morphologies and specialized structures dictated by their function. New and diverse morphological and physiological features had to develop as the organisms became more and more complex. The simple colonies would thus eventually consist of more than one cell type.

In order to ensure continuity, the genetic changes would have to be passable to the next generations, which requires a far more complex gene arrangement than existed in the unicellular organism. All the variants would have to be located in each cell, with the possibility for selective activation of one or the other batteries of genes.

Assuming that the new genes somehow did evolve, and the organism was endowed with different sets of genes governing the different morphological expressions, there would then be an even greater obstacle to overcome, namely selection. The genes of cells in particular situations would have one set of genes activated and cells in another situation would have the alternative genes activated. As a comparison, in organisms living today, nerve cells have a set of genes activated that distinguish them morphologically and physiologically from liver cells, which have a different part of the genome activated, although both possess the full set of genes.

This differential activation of either the one battery of genes or the other requires a complex system of controlling genes, which would all have to come about by chance. The probability of just one function gene arising by random chance process is less than one in the number of particles in the entire universe. In fact, it is more probable for an explosion in a woodpile to construct a functional house by chance than it is for just one such new gene to come about by random chance processes. Moreover, one would have to postulate the same scenario thousands of times as cell differentiation increased. This requires a great deal of faith.

The complexity of the genetic requirements for just two different cell types to coexist within an organism is awesome, as can be illustrated by the following example.

If we look at the relationship between a muscle cell and a nerve cell, then it is obvious that there is a great deal of morphological and functional difference between the two. This requires different gene sets to be activated in the two cell types.

Of course, these two cell types would have to cooperate with each other in the living organism in order to be of any value to the organism. Also remember that at the level of the genotype, the processes occur by chance and natural selection can only come into play once thephenotype has been produced. We are not dealing with just a simple genetic variance to achieve these goals, but a host of new genes is required to allow just these two cell types to coexist, let alone the thousands of cell types present in complex multicellular organisms.

For just these two cells, the following genes are required at minimum:

Overall structure of a muscle cell and neuromuscular junction:
1. Axon
2. Neuromuscular junction
3....

1. Promoter genes enabling the selective activation of either the one or the other. In nerve cells, only those genes which are required for nerve cells will be activated. In muscle cells only those required by muscle cells will be activated.

2. Genes, or DNA sequences, which are sensitive to the environmental cues.

3. Genes to govern the cooperation between the two cell types. This is a very complex arrangement. The two cells would have to link up morphologically in order for the one to activate the other, and there would have to be receptors that enable transfer of information from one to the other.

Where did all these genes come from? The first simple organism required more of these genes which make cooperation between different cells possible. As natural selection does not operate at the level of thegenotype , and cannot create anything anyways (only sort out that which is already there), these genes had to come about by either chance or design. Considering the complexity of the system, design seems to be the only option.

Haeckel's Gastraea theory is based on a simple morphological sequence that looks good on paper, but is untenable in reality. Read about the ways variation is increased

All mechanisms that produce variation rely on existing genetic material. None of them were subject to selection, and each of them had to come about by chance or design. 

Updated March 2010.