I've alway thought that it was strange, but the "price" that gets quoted on a stock exchange is just the price of the last transaction. The irony of this definition of price is that there may not actually be any more shares available on the market at that price.
It's also strange to me that the price isn't adjusted at all for the size of the transaction. A transaction of just 1 share will post a new price even if just seconds earlier 100,000 shares traded for a different price. (Ok, unrealistic example, but you get my point.)
I've always believed this is an odd way to describe the price.
Anyway, my diatribe here is supposed to illustrate the point that the fluctuations you see in price don't really reflect changing valuations by the stock-owning public. Each post in the exchange maintains a book of orders, with unmatched buy orders on one side and unmatched sell orders on the other side.
Buy Sell
100 @ .20 50 @ .22
75 @ .19 80 @ .23
125 @ .15 100 @ .30
... ...
If you go to your broker and tell him, "fill my order for 50,000 shares at market price", then the broker won't fill you 50,000 shares at .20. Instead, he'll buy the 50 @ .22, then 80 @ .23, then 100 @ .30, etc. Because your order is so large compared to the unmatched orders, your market order will get matched a bunch of the unmatched orders on the sell side, and each match will notch the posted price up a bit.
If instead you asked the broker, "open a limit order to buy 50000 shares at .20", then the exchange will add your order to the book:
Buy Sell
100 @ .20 50 @ .22
50000 @ .20 80 @ .23
75 @ .19 100 @ .30
... ...
In this case, your order likely won't get filled at all, since nobody at the moment wants to sell at .20 and historically speaking it's unlikely that such a seller will suddenly appear.
Filling large orders is actually a common problem for institutional investors:
http://www.businessweek.com/magazine/content/05_16/b3929113_mz020.htm
http://www.cis.upenn.edu/~mkearns/papers/vwap.pdf (Written by a professor I had in school!)