Richard Feynman once said that a scientific hypothesis starts with a guess, which should perhaps be taken with a pinch of salt. But nonetheless scientists like to use patterns when considering explanations for phenomena; at a first glance, this technique matches the principle of parsimony, or Occam's Razor, i.e. the simplest explanation is usually the correct one - excluding quantum mechanics, of course!
An example in which a potential pattern was widely publicised prior to confirmation via hard data was that of periodic mass extinction, the idea being that a single cause might be behind the five greatest extinction events. Four years after Luis Alvarez's team's suggestion that the 66 million year-old Chicxulub impactor could have caused the Cretaceous-Paleogene extinction, paleontologists David Raup and Jack Sepkoski published a 1984 paper hypothesising extinctions at regular intervals due to extraterrestrial impacts.
This necessitated the existance of an object that could cause a periodic gravitational perturbation, in order for asteroids and comets to be diverted into the inner solar system. The new hypothesis was that we live in binary star system, with a dwarf companion star in an highly elliptical, 26 million-year orbit. This would be responsible for the perturbation when it was at perihelion (i.e. closest approach to the sun).
What's interesting is that despite the lack of evidence, the hypothesis was widely publicised in popular science media, with the death-dealing star being appropriately named Nemesis after the Greek goddess of retribution. After all, the diversification of mammals was a direct result of the K-T extinction and so of no small importance to our species.
Unfortunately, further research has shown that mass extinctions don't fall into a neat 26 million-year cycle. In addition, orbiting and ground-based telescopes now have the ability to detect Nemesis and yet have failed to do so. It appears that the hypothesis has reached a dead end; our local corner of the universe probably just isn't as tidy as we would like it to be.
Now another hypothesis has appeared that might appear to belong in a similar category of neat pattern matching taking precedence over solid evidence. Bearing in mind the importance of the subject under scrutiny - the origin of complex life - are researchers jumping the gun in order to gain kudos if proven correct? A report on 565 million year-old minerals from Quebec, Canada, suggests that at that time the Earth's magnetic field was less than ten percent of what it is today. This is considerably lower than earlier estimate of forty percent. Also, the magnetic poles appear to have reversed far more frequently during this period than they have since.
As this is directly related to the composition of the Earth's core, it has led to speculation that the inner core was then in the final stage of solidification. This would have caused increased movement in the outer liquid, iron-rich core, and thus to the rapid generation of a much higher magnetic field. In turn, the larger the magnetic field dipole intensity, the lower the amount of high energy particles that reach the Earth's surface, both cosmic rays and from our own sun. What is particularly interesting about this time is that it is just (i.e. about twenty million years) prior to the so-called Cambrian explosion, following three billion years or so of only microbial life. So were these geophysical changes responsible for a paradigm shift in evolution? To confirm, we would need to confirm the accuracy of this apparently neat match.
It's well known that some forms of bacteria can survive in much higher radiation environments than us larger scale life forms; extremophiles such as Deinococcus radiodurans have even been found thriving inside nuclear reactors. Therefore it would seem obvious that more complex organisms couldn't evolve until the magnetic field was fairly high. But until circa 430 million years ago there was no life on land (there is now evidence that fungi may have been the first organisms to survive in this harsh environment). If all life was therefore in the sea, wouldn't the deep ocean have provided the necessary radiation protection for early plants and animals?
By 600 million years ago the atmospheric oxygen content was only about ten percent of today's value; clearly, those conditions would not have been much use to pretty much any air-breathing animals we know to have ever existed. In addition, the Ediacaran assemblage, albeit somewhat different from most subsequent higher animals, arose no later than this time - with chemical evidence suggesting their development stretched back a further 100 million years. Therefore the Canadian magnetic mineral evidence seems to be too late for the core solidification/higher magnetic field generation to have given the kick start to a more sophisticated biota.
In addition, we shouldn't forget that it is the ozone layer that acts as an ultraviolet shield; UVB is just as dangerous to many organisms, including near-surface marine life, as cosmic rays and high-energy solar particles. High-altitude ozone is thought to have reached current density by 600 million years ago, with blue-green algae as its primary source. O2 levels also increased at this time, perhaps driven by climate change at the end of a global glaciation.
Although the "Snowball Earth" hypothesis - that at least half of all ocean water was frozen solid during three or four periods of glaciation - is still controversial, there is something of a correlation in time between the geophysical evidence and the emergence of the Ediacaran fauna. As to the cause of this glacial period, it is thought to have been a concatenation of circumstances, with emergent plate tectonics as a primary factor.
How to conclude? Well, we would all like to find neat, obvious solutions, especially to key questions about our own origin. Unfortunately, the hypothesis based on the magnetic mineral evidence appears to selectively ignore the evolution of the Ediacaran life forms and the development of the ozone layer. The correlation between the end of "Snowball Earth" and the Ediacaran biota evolution is on slightly firmer ground, but the period is so long ago that even dating deposits cannot be accurate except to the nearest million years or so.
It's certainly a fascinating topic, so let's hope that one day the evidence will be solid enough for us to finally understand how and when life took on the complexity we take for granted. Meanwhile, I would take any speculation based on new evidence with a Feynman-esque pinch of salt; the universe frequently fails to match the nice, neat, parcels of explanations we would like it to. Isn't that one of the factors that makes science so interesting in the first place?