There are three basic configurations of bipolar transistors: common base, common emitter and common collector. The common emitter is most popular and is usually drawn with the emitter grounded. For an NPN, the base must be approximately 0.7V more positive than the emitter to turn on the transistor.
For a PNP, the base must be approximately 0.7V more negative than the emitter to turn on the transistor.
Let us say that we want to use an NPN transistor to switch an light bulb on and off. We could connect the emitter to ground, the collector to one side of the light bulb and the other side of the light bulb to the positive power supply rail. Putting a small positive voltage on the base with a dropping resistor will turn on the bulb. Remove this small positive voltage and the bulb turns off.
(I am over simplifying here. If is actually the base current that turns on the transistor.)
How do we accomplish the same thing with a PNP? We hook the emitter to the positive voltage rail and connect one side of the light bulb to the collector. The other side of the light bulb goes to ground. Now we turn on the transistor by pulling the base low with a dropping resistor.
Do a google search for BJT or bipolar transistor and you will find loads of examples.