The jellyfish Phialidium gregarium (=Clytia gregaria) is frequently found in the nearshore waters of the Pacific Northwest.  It's a small medusa (up to 20 mm) and rather insubstantial, with a transparent bell quartered by four radial canals.  All medusae are ephemeral, and so is Phialidium, but because it's a little less ephemeral than most, and because if well-fed it will spawn every morning, this species has contributed to an outsized share of the few studies of development in jellyfish. 

As in most other cnidarians, cell division in Phialidium is unipolar – the cleavage furrow ingresses from only one side of the cell.  This is more than just a curious way for a cell to divide: the site where the first cleavage furrow initiates, usually the animal pole where meiosis completed, will become the oral pole, which happens to be the posterior of the larva.*  In a classic paper (citation 1), Gary Freeman presented an elegant experimental proof that the first furrow initiation site isn't merely correlated with polarity of the embryo, but actually determines that polarization.  He suppressed first cleavage with the drug cytochalasin, then washed the drug out.  These eggs, which had missed first cleavage, had two nuclei, which each initiated a cell division furrow.  If they did so at different sites and more or less at the same time, resulting embryo developed two body axes. 

Cell division is equal, results in cells of roughly the same size, and occurs synchronously.  These embryos are so simple they're almost boring.  But because so little takes place, as one strains to find a point of interest occasionally one notices something that might pass unremarked in a flashier embryo.  Freeman, for instance, focused on the polarity of cells in the epithelium – that’s about the only trait this epithelium displays until metamorphosis – and through grafting experiments showed that the polarization initiated at first cleavage is propagated something like the polarization of dipoles in a magnet (citation 2). 

In this movie, a couple of bits of rubble draw attention to a mundane yet profound developmental transition.  First cleavage completes to the northwest, and a few bits of cytoplasm get left behind.  During the first few divisions, these fragments get slurped in and out through a space between cells.  Eventually, however, it so happens they're trapped inside.  This reveals one of the most fundamental, if little-noticed, events in the development of animals: at some point, the embryo ceases to be a cluster of loosely-associated cells, and those cells conspire to behave as a tissue.  Most often, as in this embryo, that first tissue is the single-layered epithelium, apical ends to the outside world, cell-cell junctions making seals around the sides, and basal ends facing a new, interior space.  It's no wonder that some early embryologists imagined the jellyfish embryo as the prototype for us all.

— text by George von Dassow and Katie Bennett

* Yes, that all sounds backwards if you know much about other animal embryos.  The hydrozoan planula larva swims with the oral pole behind; the aboral pole (the front end) will attach to the chosen substrate when the larva settles, and the tail end will form the hydranth, the anemone-like polyp.  The mouth will open at what is then the top-most point, which had previously been the hindmost point while the larva was swimming.  Thus the animal pole of the egg becomes the posterior pole of the larva which becomes the oral pole of polyp.  Doesn't the animal pole usually equate to the front end of the swimming larva?  Yes, but happily, such rules (like grammar) are descriptive, not pre- or proscriptive.

Citations:

1. Freeman, G. (1981) The cleavage initiation site establishes the posterior pole of the hydrozoan embryo.  Wilhelm Roux's Archives of Developmental Biology 190:123-5.

2. Freeman, G. (1981) The role of polarity in the development of the hydrozoan planula larva.  Wilhelm Roux's Archives of Developmental Biology 190:168-84.