My friend told me that there is mercury in seafood because Japan dumped tuns of the stuff in the ocean years ago and that it spread everywhere to pollute all sea life. This kind of sounded a little fishy. What’s the real dope?
Can’t really take Japan off the hook completely, but the mercury comes from pollution for which many countries including the U.S. are responsible. 
The OP reminded me of something I read or saw on TV ages ago, so after a web search I found.
http://www.guinnessworldrecords.com/index.asp?id=49231
Which may be the source of the OP’s friends statement, though certainly doesn’t blame Japan for all or even the majority of mercury in the sea.
What I gather from www.epa.gov is that it is difficult to trace mercury to specific incidents. Coal powerplant emissions and industrial/municiple waste disposal are listed as the two major contributers. Mercury can make its way into fish no matter whether it is dumped directly into the ocean by the Japanese, or sent into the air by a powerplant in Texas.
Mercury is one of those nasty chemicals that persists in the environment for far longer than we’d like. There are lots of different sources of it - both anthropogenic and natural. Human-related sources include combustion in oil and coal burning power plants, industrial boilers, and incinerators. In the area where I live, the main environmental source of mercury is from abandoned gold and silver processing mills.
I don’t know a lot about mercury dynamics in the oceans, but I’ve been working on a mercury-contaminated creek for a couple of years. The mercury in these creeks comes from the stockpiles of mine tailings - mercury was used to extract gold and silver from the ores that were mined. Over a period of 100 years or so, the mercury leaches into the soil and is eventually transported by rainfall to the banks of a small lake. This lake has a dam on it that controls the outlet; the creek begins at this dam and flows north to a larger river. Because the dam releases water from beneath a flood gate rather than over the top of one, the released water tends to scour sediment from the bottom of the lake.
In the creek, mercury is found mainly in an inert elemental form, usually adsorbed (attached) onto soil particles. In this form, it doesn’t really have much potential to affect the wildlife in the creek or to enter the food chain. Under some conditions, though, the elemental mercury undergoes a process called methylization. Methyl mercury is nasty stuff - organisms that live in a contaminated creek accumulate a buildup of methyl mercury in their fatty tissues. When you start to talk about seafood, you’re really looking at top-order predators in the environment. Take a trout, for example, that’s spent its entire life in this water: it’s got some mercury buildup just from that, but it has also spent its life eating smaller fish that live in the same water. Those fish have spent their lives eating caddis flies, crawfish, and other little goodies in the same contaminated water; all those organisms live their lives with mercury contamination too. So the seafood that we eat has elevated levels of mercury in its tissue due to the process of biomagnification - that one trout has a trout, 100 chubs, 1000 caddis flies, and 1000 worms worth of mercury in it.
The science of mercury methylization is still the subject of much research and debate. The general indication seems to be that anaerobic conditions and sulfur reducing bacteria produce methyl mercury. If anybody’s interested in the chemistry, I can dig it out and post it, but I’m not familiar enough to do it off the top of my head. What I can say, though, is that natural wetlands, which we spend a lot of time and money trying to preserve or restore, are traditionally areas that produce just those kinds of conditions. The preservation/restoration of riparian areas in mercury contaminated watersheds can be a real balancing act between these interests.