Results are reported for several international collaborative experiments which examined methods of assessing degrees of partial resistance in potato and virulence in potato cyst nematode (PCN,Globodera pallida).It was demonstrated that absolute rates of multiplication can be extremely variable on both susceptible and partially resistant clones, even when the same population and test procedures are employed. It was therefore concluded that, on clones with quantitatively inherited resistance (i.e. partially resistant), absolute rates of multiplication cannot be used to separate pathotypes. Expression of these rates as percentages of those on the susceptible controls reduced the absolute differences between tests, but the values obtained were still too variable for statutory use. However, whatever the environment or nematode population used, it was observed that the resistance of the test clones and virulence of the nematode populations were generally ranked in a similar order. The main exceptions to this were: (1) a Petri‐dish test where cv. Vantage was less resistant than in canisters or pots and (2) a pot test where cv. Darwina tended to be more interactive with environmental factors than the other test clones. It was also observed that with some populations of PCN in pots the susceptible cv. Bintje was a less good host than cv. Désirée. On the basis of these results it is suggested that certain partially resistant clones should be used as internal references in statutory, recommended list and breeders' assessment tests against which the resistance of the test clones are compared. For international comparability it is necessary that the different centres conducting such tests use the same reference clones and nematode populations, and similar test meth