ISCR1, first identified as a "common region" (CR) in some class 1 integrons, and other related ISCR elements have been responsible for capturing and mobilising certain antibiotic resistance genes. They are unusual insertion sequences, related to the IS91 family. Members of this family lack the terminal inverted repeats found in many other IS families and they move by rolling circle replication, catalysed by the Rcr protein encoded within the element. Replication starts at the oriIS end, downstream of the rcr gene, and continues through the element to terIS motifs defining the other end. Insertions of IS91 family elements in different locations allowed their ends and target sites to be defined, as well as identifying cases where replication has continued beyond the terIS end, resulting in capture of adjacent DNA segments. ISCR1 has always been found adjacent to the same position of the 3'-conserved segment of class 1 integrons, making it hard to define the exact terIS end. It has been suggested that this structure could be explained by a deletion encompassing part of an ancestral ISCR element and part of the 3'-CS. However, at 2,154 bp the element defined as ISCR1 appears longer than other ISCR elements (e.g. ISCR2 is ~67% identical to ISCR1 over 1,393 bp) and could already contain captured segment(s) adjacent to the original terIS. Searches with the Rcr1 protein identified an ISCR1-like element in a plasmid from an uncultured bacterium, also recently identified in association with the aac(6')-Ian amikacin resistance gene on a plasmid from Serratia marcescens and found with this gene in a plasmid from Escherichia coli and in two Acinetobacter baumannii genomes. An alignment of the nucleotide sequence of this element with ISCR1 revealed 79% identify over 1,811 bp and a pair of short inverted repeats, a characteristic of terIS of IS91-like elements, is present near to the end of this match. Identification of an element related to the region currently defined as ISCR1 appears to support the hypothesis that this region may contain an ancestral ISCR plus additional segment(s) adjacent to terIS.