Latency is the time involved in any sort of process, usually a communication of some form.
Throughput is how much information any given system can process, usually measured according to set increments of time.
High latency does not have to mean low throughput and vice versa. And there a few examples we can use to illustrate this:
For instance, in an internet connection, latency is generally measured by how long it takes for a request to get from your computer to a given server. With Cable or DSL this number tends to be relatively low. With Satellite, this number tends to be very, very high. Throughput over that same connection is measured in Mbps or Megabits per second. The throughput numbers for these different connections may be very similar, even when the latency numbers are not. Satellite internet, for instance, tends to have quite a bit of bandwidth. They are are to transmit a very large amount of information from the satellite to you over a short period of time. It just takes a long time to make the round trip from your computer to the server, to the satellite, and back to you again.
Another way to think about this would be a station wagon rolling down the highway at 55mph (my grandmother is driving) full of SD memory cards or USB memory sticks. Hundreds of thousands of them, to be more precise. The throughput of this station wagon is insanely high. Think about it: there could be more than several hundred Petabytes inside that station wagon. So even if the latency of the trip from Chicago to Sacramento could be 3 whole days (25 hours or so of driving), that 25 hours divided by a few hundred Petabytes is still enormously faster than any internet connection money can buy.
The reason that latency and throughput are not necessarily correlated is because of data density. It's not so much a matter of how quickly information can make the trip. It's more a matter of how much information can make the trip each time.