In Figure 2.17 (b), the junction need not be formed physically at one point and instead could be distributed. Figures 2.17 (c) and (d) are respectively called the laws of intermediate metals and intermediate temperatures.

Two additional effects that influence thermocouple performance are Peltier and Thompson effects. The former is related to the heating of the junctions between dissimilar materials when a current flows through them and the latter to an emf production due to a temperature gradient along the length of the wires. Peltier effect is significant when the voltmeter measuring the thermo-emf does not have a large enough input impedance. However, thermo-emfs tend to be small and this effect is usually insignificant unless the sample whose temperature is being measured is physically small or the unknown temperature is considerably below the ambient. On the other hand, Thompson effect is an important source of error in thermocouple measurements. It can be kept small by using very short lengths of thermocouple wire and using high-conductivity copper wires to convey the thermo-emf to the measuring instrument. The thermocouple-copper junction will not result in any new emf if the laws of thermocouples described above are not violated. In particular, the copper-thermocouple and copper-instrument junctions must be at the same temperature.