In vitro clonal propagation of plants should generate identical copies of the selected genotype. However, associated stress might result in a breakdown of control mechanisms and consequent instability of the genome. We have used several molecular methods to assess the genetic stability of long-term propagated (24 years) multiple shoot in vitro culture of pea (Pisum sativum L.). We focused on assessing the stability of repetitive sequences, such as simple sequence repeats (SSR) and retrotransposons, both comprising a large part of genome. No differences were found when seedlings (Co-2004) or original seed (Co-1982) controls and long-term or newly established in vitro (one subculture cycle) samples were investigated by the SSR, inter-repeats (ISSR) or inter-retrotransposon amplified polymorphism (IRAP) method. However, the more global amplified fragment length polymorphism (AFLP) and particularly the methylation sensitive MSAP methods detected 11 and 18% polymorphism among samples, respectively. Interestingly, investigation of the global cytosine methylation status by HPCE measurement revealed no statistically significant differences. Some evidence of retrotransposon re-arrangement was observed by sequence-specific amplification polymorphism. This occurred mostly in the abundant Ty3-gypsy type Cyclop element and to a smaller extent in the Ogre element. Alternatively, no polymorphism was detected among the PDR-1 element of the Ty1-copia type retrotransposon. Based on these results, multiple shoot culture of pea maintained over a long period may be considered as a true to type multiplication method of the original genotype.