Source code contains the instructions on how a program should function, written in a human-readable programming language.
The source code contains the design and building blocks that are used to create the program. Software developers can modify the source code to fix, modify or enhance the behaviour of a program created from it. For larger programs it is not uncommon for there to be thousands of files of source code, totalling hundreds of thousands or even millions of lines of code.
Almost all programs are distributed in a form that is ready to run. The way source code is converted into a runnable program can be quite complex, and can depend on the programming language that it has been written in. However, there will commonly be a stage where the program is compiled into object code or byte code. Once a program is in this form it is just a long list of numbers and so isn't human-readable.
With software that is proprietary or closed-source, you cannot see or change the source code it was created from. The person, team or organisation that created it maintains exclusive control over it. This means that it is impossible for anyone, except the authors, to enhance or modify the way it works. They are the only ones that can fix bugs in it, or check to ensure that it is working the way that was intended.
Open-source software is different. Its authors make the source code available to everyone, so they can view, copy, modify, and share it. This gives people the freedom to improve or alter the program in any way that they want to. It also provides transparency by letting people inspect and audit how a program works, and allows them to be sure that the program isn't going to do anything they didn't expect it to do. Also because the source code is freely available anyone can then build and run the program on whatever hardware they like.