Multi‐residue gaps, a class of molecular characters with exceptional reliability for phylogenetic analyses

Qualitative methods of obtaining phylogenies seek distinct classes of characters that collectively are common, yet individually are insufficiently improbable that they provide reliable indications of monophyly. Multi‐residue gaps (insertions or deletions) in DNA and protein sequences are easily recognised and are available in diverse organisms and loci. They are likely to experience few homoplasies, because these require two separate events to be matched in both starting position and length. Two tests of the ability of multi‐residue gaps to recognise clades are conducted. (1) Among the 10.76 kb of noncoding DNA sequences from the ψη‐globin region of 5 primates, there are 7 analysable multi‐nucleotide gaps. These provide a corroborated and noncontradictory, fully resolved gene tree of hominoids. (2) Among the 35 amino acid sequences of globin chains available in 1978, there are 9 nonoverlapping multi‐amino acid gaps. These indicate 8 noncontradictory divisions that agree with recognised taxa and gene categories. In the 412 realigned globin sequences available in 1989, 8 of the gaps are still analysable. These show a maximum of two homoplasies among the 8 × 412 = 3296 species‐characters. Multi‐residue gaps have a high potential to indicate clades. They could be used to reduce the computing labour required in a quantitative study or pooled from different data sets to examine particularly important or difficult phylogenetic questions.