If you work your way backwards through the fossil record, for about 400 million years you would be working back through ages that included organisms living on land.  From about 400 million years back to 600 million years ago, all kinds of complex multicellular life would have been confined to the waters of the earth.  From about 600 million years ago back to about two billion years, you'd be looking at simple eukaryotes, and from two billion back to at least 3-and-a-half billion years, maybe much further, evidence of prokaryotes, including the stromatolites mentioned in a later section.  The earth itself is thought to have come together as a planet a bit more than 4-and-a-half billion years ago, but would have been in the finishing-up phase for perhaps three-quarters of a billion years.  So where did those prokaryotes come from?

One theory, panspermia or the space seed hypothesis, proposes that life can be found on meteors and other space debris, and such objects brought the first Life to Earth.  In fact, some very rigorous tests suggest that there may be bacteria in space, perhaps left from some long-destroyed planet and perhaps capable of surviving a leisurely trip across the universe.  If this is the case, then no one could be sure of the conditions under which that life would have first evolved.  This is a possibility, but a bit of a dead end explanations-wise.

Another theory is the special creation theory, that supernatural forces (pick a candidate from a long list) put the first prokaryotes together as part of the long building plans which would ultimately lead to us.  Science tends to be resistant to supernatural explanations of things, but when you're dealing with conditions so far in the past, there is not much less evidence for some aspects of this idea than many of the others.

What we are going to look at are the theories that assume that Earth's life is "home-grown," emerging from those earliest conditions to exist eventually as what we see around us. 

At their simplest beginnings, living "things" really only needed three characteristics:   the ability to self-organize, the ability to reproduce themselves and the ability to evolve.  

Eventually, to develop into the life we see on earth today, a few more traits were needed:  a DNA-based information coding system, a complex protein-based chemistry, and a cell membrane containment system.

When biologists first began to wrestle with these concepts, the ideas of uniformitarianism were still powerful, and people theorized in terms of a world not all that different from the world they knew.  Our world's ecosystems depend almost exclusively upon photosynthesis to construct the fuel that all life runs on;  in an ancient world with conditions similar to today's, it seemed that you would need plants (as organisms that can make complex molecules using simple building blocks and energy available from the environment, plants are known as one type of autotrophs, or "self-feeders") to evolve first, or there would be no bottom link to the food chain.  This was an insurmountable problem to theorists, because the processes of photosynthesis are much too complicated to have spontaneous formed from "nothing" under present-day conditions (or any conditions).

But through the first half of the 20th Century, studies from astronomy and geology suggested that the very early Earth was a dramatically different place than it was today.  It was a hotter and nastier place, heated by a warmer sun and a warmer interior.  This suggests more violent weather, with huge storms and lots of lightning.  The atmosphere had no or almost no oxygen, so no ozone layer to absorb the sun's ultraviolet rays.  Simple organic materials, which are common in the space dust and debris that the Earth formed from, would have filled the oceans with a kind of brownish gunk, potential building blocks of life.  This is known as primordial soup, and this concept of the early Earth led to something called the heterotroph hypothesis (heterotrophs require their complex fuel molecules already made, unlike autotrophs).  Some interplanetary missions are looking for similar circumstances elsewhere in the solar system, where life may also have arisen.

The strength of the heterotroph hypothesis is that it gives the first forms of life a source of "food" that doesn't itself come from living things:  the primordial soup.  This is how the rest is supposed to have happened...

The ability to self-organize.  This requires some already-formed building blocks, from the soup, and a source of energy that would serve to help drive them into increasingly complex forms.  Experiments in the early 1950s began to confirm that such processes could at least begin.  Those experiments used lightning, confined to a "primordial soup bottle," to stimulate production of complex materials.  Since that time, all of the various forms of energy available on the early earth have been tested, with varying results.  The current "leading contender" for life-organizer are hydrothermal vents, openings between Earth's surface plates at the bottom of the oceans.  There, water mixes with hot magma and releases a hot soup of materials even today.  They have an energy source - heat - a source of materials - once soup, now magma - and, perhaps most importantly, are a stable, long-lasting ecosystem and a place to "work the bugs out" of the earliest living systems.

The ability to reproduce.  As these early self-organizing molecules grew, only those which could make and spread copies of themselves had any real future.  Life on today's Earth uses DNA code to store all of the information it needs to make the proteins it actually runs on, but DNA has little activity beyond that, and proteins generally can't duplicate themselves.  There are theories that try to address those problems, but the leading current theory is that the first really complex systems were of RNA, a hypothesis usually called the RNA World hypothesis.  RNA has DNA's coding abilities and some protein-like activity, and it isn't difficult to see the evolution of a DNA-coded protein system growing quickly from an RNA-based ancestor.

The ability to evolve.  Once you have a planetwide ocean full of self-organizing molecules able to reproduce themselves, you have a competitive ecosystem where selection can take place.  This stage of molecular evolution would have favored those who could work most efficiently, or best accumulate building blocks, or reproduce the fastest, or work with other molecules in a cooperative fashion, perhaps linking RNA or DNA codes for particularly good proteins together to work as a unified system.  A combination of RNAs like those that work in cellular protein production (a little one holding amino acids and a bigger one to put those amino acids in a particular order) might have been especially useful and definitely on the track to Life-as-we-know-it.  And these unified systems might work even better with some confinement and protection...

That ancient primordial soup would have been coated with a mixture of oily-fatty lipid substances, materials that in a surf environment can form cell-like bubbles - not as tight a container as the membrane of actual cells, but a bit of an advantage for a contained chemical system.  From inside these loosely-sealed chambers, often called protocells,  raw materials could be accessed and kept away from poachers temporarily, and chemical systems could evolve with some room to move without everything floating away from each other.  At this point Life would start to have a noticeable resemblance to today's simplest organisms:  a membrane-enclosed soup of active proteins, made using some form of nucleic acids, interacting with the environment, pinching off offspring, and struggling to compete, winners continuing a family line and losers dying out.

LIFE.

Except...

This worldwide ocean of competing heterotrophs have two sources of fuel to run on;  the original primordial soup, quickly being consumed with no way to replenish it, and other early organisms.  This is not a formula for long-term success, and unless a system for making new fuel emerged, the limited resources would eventually run out.  One wonders on how many planets across the universe this course into a dead end may have happened.

But that didn't happen on Earth, because autotrophs, they who were eventually to become the whole basis of the food chain, evolved.  Most likely, the first autotrophs were able to use the heat from hydrothermal vents in especially efficient synthetic ways, a process called chemosynthesis.  Simple organisms performing this process can be found today in hydrothermal vents and some hot springs.

But how to get to photosynthesis, a system that also uses energy from the environment to construct fuel molecules, but one using light, a very different from of energy than heat?  One possibility is that, in order to stay near their vent homes (which are like oases in a desert, if displaced by the turbulent currents it would be good to find your way back), some organisms developed ways to detect the faint glow the vents produce as a homing beacon.  Once you have a chemosynthesis system and a light-reactive system in the same organism, it's easier to imagine development of a system that could use light as an energy source for synthesis.  This would be a huge advantage:  chemosynthesis-based ecosystems, even in the early earth, would have been few and far between, but the entire surface of the oceans would have access to light.

Recent research has indeed found a type of bacterium that lives in hydrothermal vents and uses the tiny amounts of light there for photosynthesis.

So what became of the various molecular complexes of the earlier primordial soup?  It is assumed that they went extinct, outcompeted by the more efficient cellular life now formed.  But do we know this for sure?  It's unclear whether anyone has actually searched for such entities in the oceans, which are filled with protein complexes, but such material is assumed to be a combination of proteins lost from living things, and viruses.

And what of viruses?  If some primordial molecular complexes survived by becoming parasites in the cells that had evolved around them (there may have already been complexes that existed by using other complexes' abilities against them). They might have eventually evolved (as parasites do) to have only hints of their ancestry, to become so efficient in using the cells' machinery (with its efficient DNA-to-protein equipment), even to surrounding themselves with bits of host membrane (many viruses have non-membrane coatings, like one might expect from those primordial complexes) so as to seem more like degenerate cells than a continuation of the very first living systems.

The basic process of photosynthesis combines water molecules and carbon dioxide molecules to make simple sugar molecules, usable as fuel and as a structural building-block in plants.  The process needs light to work and releases oxygen as a by-product.  The early earth would have had little oxygen, or at least free oxygen, because the oxygen present would have been "tied up" in molecules such as water, carbon dioxide, and a long list of mineral compounds in rock.  Oxygen is good at combining with other materials, which would have made it a potential poison to early complex living systems.  As photosynthetic organisms flourished, systems to resist the damaging effects of free oxygen had to evolve as well (our own bodies have multiple systems in place to resist oxygen damage).  The environment itself changed:  a layer of sedimentary rock from this period shows that the oxygen combined with an ocean full of dissolved iron, settling a layer of iron oxide (rust) into the sediments, an indicator that the chemistry of the oceans themselves were changing. Oxygen left the oceans (it doesn't dissolve particularly well in water) and built up in the atmosphere, eventually rising to much higher levels than can be found in the waters.

Respiration is a system by which fuel molecules from food are converted to the fuel molecules that power cells, mostly in the form of a molecule called ATP (Adenosine triphosphate).  Various approaches would have existed long before the rise of photosynthesis, but the buildup of oxygen and wide availability of the simple sugar (glucose) produced by photosynthesis favored the rise of a particularly efficient respiration system that was almost a mirror image of photosynthesis:  aerobic respiration.

So at this stage of Earth's history, the oceans would have been full of photosynthesizing prokaryotes, aerobic prokaryotes, and predatory prokaryotes feeding off the others.  And a truly diversified world was about to get even moreso...

As some cells got more complicated, subdividing the cell into smaller specialized chambers with their own particular chemistry could increase the efficiency of certain processes, especially the process by which DNA code was converted into a form from which proteins could be made.  At some point, this DNA processing was put into a separate room, and eukaryotes evolved, large, efficient, able to gobble up smaller prokaryotes but probably not able to generate the same numbers.

An ingested prokaryote can be digested and absorbed, but sometimes they might be more useful confined and alive.  A photosynthetic prokaryote could make food in the sun for its captor, and an aerobic prokaryote could help it better utilize the food it took in.  The prokaryotes benefited, too:  beyond not being digested was a good thing, and becoming part of the larger predatory cells certainly reduced their potential as prey.  This type of mutual-benefit relationship is called a symbiosis, and that absorbed prey might be used rather than digested is reasonable - it can be found in a few of today's protistans.  The vast majority of present-day eukaryote cells contain aerobic respiration chambers called mitochondria that structurally resemble aerobic bacteria, even down to having the remnants of bacterial chromosomes in them, and eukaryote plant cells contain photosynthesizing chloroplasts with similar resemblances to a type of photosynthetic prokaryote.  That these structures began in the way described here is known as the endosymbiont theory,  proposed in the late 1960s by Lynn Margulis, then at Boston University, and eventually widely accepted.  Other eukaryote structures might have endosymbiont origins, but the evidence for those is more controversial.

So the world's oceans became a mixture of prokaryotes and simple eukaryotes (and viruses, maybe) for a very long time.  Eventually, something that eukaryotes can sometimes do but prokaryotes almost never do led to the next major step of Life on Earth...

Most of the noticeable life in our world is big, multicellular.  The trip to multicellularity undoubtedly went through the colonialism stage:  colonial organisms are made up of individuals that are capable of living independently, but join together and specialize in different jobs.  When unicellular organisms do this, you have a multicellular organism that can be split up into surviving parts.  An evolutionary progression to cells that can live together, specialize, and become so dependent on each other that they can no longer live apart is not much of a leap.

Prokaryotes can and often do live in groupings - a type of mineralized bacterial surf structure, stromatolites, shows up in very ancient fossils and can be found living today - but they rarely specialize.  It's more like they "hang out" together.  Eukaryotes, perhaps because with nuclei they are better able to control what genes get expressed in a particular cell, seem to have a talent for specialization.

For a very long time, the only multicellular forms in the fossil record were algae, barely above the colonial stage.  Multicellularity was an advantage for plants, but the potentials seemed limited to floating mats or attached strands in the shallows.  But when animals evolved multicellularity, it was like the evolutionary floodgates opened up.  Over an incredibly brief time, the world became full of swimming and crawling eating machines.  In a world where life had existed for 3 billion years, many types of complex multicelled animals "appear" in the fossil record over a period that might be as short as 40 million years, a period called the Cambrian Explosion.  With maybe one exception, every known phylum of animals evolved during this period, including early members of our lineage.

The war was on.  Plants seemed to be able to deal with larger plant-eaters with few obvious adaptations:  algae remained relatively simple.  But animals got bigger and nastier, smaller and quicker, more protected;  think of an adaptation that provides an advantage in a world of animals, and it appeared during the Cambrian Explosion.  Except one...

 
The water is a great place for a living thing, since they depend on water to float their molecules and support their chemistry.  Life evolved in the oceans and filled them wherever there was enough light to photosynthesize or food to eat.  But there were niches going unused, up out of the water, on the bare land of the continents.  How could it be exploited?

The land environment would have had several significant differences from the water environment that would need to be adapted to:

- NOTHING WAS ALIVE UP THERE.  For the first, pioneer organisms (this term is applied to the newcomers in any "new" environment), they needed to deal with an environment devoid of life and nutrient-poor.  Animals might use it as a place to avoid predators, but would need to return to the water to feed.  Plants had a trickier obstacle:  the light, water, and carbon dioxide they needed for photosynthesis might be available, but other nutrients for making molecules such as proteins would not be.  It is quite likely that plants would not have been able to move onto the land without symbioses established with fungi and bacteria to help them get the materials they would ordinarily get from the water around them.

- WATER EVAPORATES IN THE AIR.  The water content of cells is critical to the function of cells - if too much is lost or gained, the cells cease to function.  A land organism cannot lose too much water to the air or it won't survive.  But there are transitional ecosystems that might have required adaptations usable against evaporation:  tidal zones, where organisms are sometimes left "high and dry," as well as in pools that might fill with rain runoff or evaporate, where resisting a similar dilution change in the cells would be necessary;  fresh water systems, where a resistance to the inflow from very dilute surroundings would be necessary.  Our distant ancestors, the bony fish, apparently evolved in fresh water and developed an efficient waterproofing system to keep water from rushing in;  that barrier could also prevent water loss in the air.  It is quite likely that, for this and other reasons, all life on land evolved from tidal and/or fresh water ancestors.  Of the three multicelled Kingdoms, the fungi seem to have had the hardest time with drying, perhaps because of the way nutrients get absorbed - it's almost impossible to move materials effectively across a waterproofed surface - but they've gotten by in moister environments, in soils and in the wetness of other living things.

- YOU CAN'T FLOAT IN THE AIR.  The buoyancy of water reduces the need for strong support structures.  This was especially a problem for plants, which didn't undergo much dramatic evolution until they moved on to land, where the need for complex support structures and then structures to move materials around against the force of gravity led to an explosion of different forms.  Animals had some adaptations ready to go:  muscle systems for moving quickly through the water or across the bottom needed modification to work on land (fins needed to be more leg-like in our ancestors;  insect and spider ancestors had to lighten their outer covering just to hold themselves up), but structures used for moving across tidal flats or in very shallow water became usable away from the water as well.

- TEMPERATURE FLUCTUATIONS.  A body of water gains and loses heat more slowly than the air does, so temperature changes are slower there.  Temperature has a huge effect on cellular chemistry, and only chemistry that can somehow deal with rapid changes can be used in a land organism.  Again, tidal areas and shallow fresh water ecosystems would have been good staging areas for developing some flexibility.  Plants, not being able to move from place to place to adjust their temperature, had a more critical problem, and may have taken some time to adapt to non-tropical areas.

- DIRECT SUNLIGHT.  The frequencies of energy in sunlight can cause molecules in living systems to become unstable, as happens in the mutations that lead to skin cancer.  Water reflects several frequencies and quickly absorbs many more, making the problem much reduced for organisms that live below the surface.  Most land organisms have protective pigments to keep the sunlight from penetrating and harming them.  The adaptations would also have been required for life in tidal areas and shallow fresh water.

- MUCH MORE OXYGEN.  As mentioned earlier, the air can hold much more oxygen than water can, and oxygen is a very reactive material (you can be poisoned with too much of it!).  An organism can't live in the air if it can't handle the increase in oxygen.  Long-term, the higher oxygen levels allow for much more energetic metabolisms in aerobic animals.  Even an animal like a crocodile gets such an energy advantage from breathing air that it would never evolve a water-breathing system again, and it's difficult to understand how anyone could ever develop a system by which a human could breathe underwater - there just isn't enough oxygen.

- SPERM NEED WATER.  Sexually- reproducing animals and plants had for the most part evolved systems where the swimming sperm were released and had to get themselves to the waiting egg cell.  This doesn't seem like much, but for a couple of the major land groups it was the most difficult problem to solve - long after the difficulties of water loss, and support, and other land challenges were met, amphibians and ferns still required open water for reproduction.

Virtually every phylum of organisms was able to get a least a few species up onto land, (some were there long before the others, though) although they all still have some water-living species as well.

 
Through the course of Life's History, some interesting things have happened:

Continental drift.  The rocks that make up the continents mostly float like corks on heavier rocks beneath.  These huge corks, called continental plates, move slowly but with huge momentum pulling away from each other to make things like the Atlantic Ocean or colliding in huge "fender benders" that ripple up things like the Himalayan Mountains or the Panama bridge between the Americas.  These movement have huge effects on climate around the world, driving evolutionary change as areas get wetter or warmer or whatever.

Catastrophes.  Sometimes the world can change in an instant.  Huge chunks of rock fly in from space and smack into us, changing the weather for months or years and flash-frying whole continents.  A low-lying basin connects to the ocean, and the Mediterranean Sea fills in seemingly overnight.  A huge volcanic eruption covers almost a third of India with lava and spews huge amounts of climate-changing gas and dust into the atmosphere.  Continental drift allows an invasion of new competitors into a stagnant ecosystem over a matter of decades.  Major changes in climate, either warming or cooling, can affect almost all multicellular life.  These can cause the major transitions found in the fossil record, including a few so large that they are known as mass extinction events; the asteroid impact that wiped out the dinosaurs and left our tiny scavenging ancestors to take over is a well-know one, but there have been several, and the causes are not always known.  According to recent research, there may be a regular rise-and-fall of diversity (lots of different species, then a drop to very few, then a rise again) on a 62-million-year cycle;  this may renew interest in an older theory about Nemesis, a proposed "dark star" companion of the Sun on a very long elliptical orbit - in the original hypothesis, it was supposed to visit every 26 million years, do bad things about the solar system with its gravity effects, and move away, but the theory could be amended to a 62-million-year cycle.   Recently, the real nemesis may be closer to home:  humans adjust the environment on a huge scale to fit their preferences, and this can be a problem for the organisms that share our environment. 

One cause sometimes proposed is disease, but this is very very unlikely as a major player for a couple of reasons.  For one, diseases tend to adapt to particular hosts - a disease that can affect a wide variety of organisms is extremely rare, and even then the effects vary because each type of host is a unique ecosystem.  More importantly, however, is that diseases are caused by evolving organisms, and the more successful individuals are not the nasty ones, but the ones that keep the host semi-well and moving around to spread its offspring.  Except in tiny systems or small populations, diseases get less damaging as they spread and so are unlikely candidates for causing widespread carnage.