Added to SteveP's. Here is the rule:
1. Keep the classfull octets as they are. In this case (Class A). Octet 1 will stay at 77
2. Zero out all bits of the subnet-work. In this case bit 1 in 2nd octet up to bit 3 in 3rd octet (counting from the left). This will be your FIRST subnet.
3. Increase the subnet-work bits by 1. This is your SECOND subnet.
4. Increase the subnet-work bits by 1. This is your THIRD subnet.
......
And so on until I die.... (just kidding

). Host bits should be all zeros as SteveP did.
If the question asks for exact IP ranges. You would get the first and last HOST addresses for each subnet
First address: All zero for the host bits + 1
Last address: All ones for the host bits - 1
In such a case, Your final answer would be some thing like this:
first subnet is 77.0.0.1 to 77.0.31.254
second subnet is 77.0.32.1 to 77.0.63.254
third subnet is 77.0.64.1 to 77.0.95.254
fourth subnet is 77.0.96.0 to 77.0.127.254
Hope it's clear