can anyone help me out by explaning how 16 bit programs run on 32 bit machines or does any one know of a web address that will explane this in some depth thanks

For one thing Intel kept each generation of the 8086 microprocessor almost fully backwards- compatible. For example each 32- bit instruction has a prefix, and normal reigister use is permitted by using the lower 16- bit word of the EAX register (80386+) as the AX. There are a few exceptions to the backwards-compatibility in 286s and up, that may have rendered a few good programs candidates for replacement. An 8086 push sp assembler instruction will push a value onto the stack that is reflected in the instruction, in later 286s the pushed value became the SP before the push was executed. I really don't know any more on the subject (heck, my first machine was a Pentium!), but I hear Intel offers several IA32 manuals in downloadable .PDF format.

Windows retains the ability to run old applications by a process called Thunking (or something like that). Basically, they left the Win31, 16-bit code in windows. Ever wonder why it's so bloated?

16 bit code has a offset register, and the code is offset by that, so when code is place at different memory locations, the code wont't crash. it looks at the offset and and users the register as a pointer.