The phylogenetic analysis was performed using PAUP version 3.1.1 (Swofford, 1993), utilizing a data matrix originally set up in MacClade version 3 (Maddison & Maddison, 1992). Heuristic search analyses were performed with the following options applied. Addition sequence: simple; one tree held at each step during stepwise addition; tree-bisection-reconnection (TBR) branch-swapping performed; MULPARS option in effect; steepest descent option not in effect; branches having maximum length zero collapsed to yield polytomies; topological constraints not enforced; trees unrooted; multi-state taxa interpreted as polymorphism; character-state optimization: accelerated transformation (ACCTRAN). All characters unordered and unscaled.
As suggested by Hillis (1991) and demonstrated in Hillis & Huelsenbeck (1992) , the strength of the phylogenetic signal (measure of information content) for a given data set can be interpreted from the skewness (or otherwise) of the distribution of all the given trees. A symmetrical (bell-curve) distribution indicates that the data are random, and the most parsimonious tree obtained from these data would not necessarily reflect the phylogenetic history of the taxa used. Notwithstanding recent criticism of the skewness criterion as a measure of phylogenetic signal (Källersjö et al., 1992) it is applied in the present study.
Using the exhaustive search setting the distribution of all trees can be obtained from a given data set. An alternative method of estimating the skewness of tree distribution is to use the random trees option. A sample size of 1000 random trees was chosen as it is the minimum number needed to obtain a reasonably accurate estimate of skewness (Jamieson, 1994).
Two palinurids, two astacids, and three thalassinoids were variably used as the outgroup. The palinurid taxa Panulirus and Thenus were used as the outgroup in the principle analysis because of the relatively undisputed basal position of the Palinura within the Reptantia, based on fossil (Glaessner, 1969), molecular (Vaughn & Traeger, 1976) and somatic morphological (Scholtz & Richter, 1995) evidence. As with the Palinura, the Astacidea previously have been designated a basal or stem position within the evolution of the reptantian decapods. Makarov (1962: 23) stated "The group from which all Anomura originated was that of the Astacura,......" and Glaessner (1969) indicated that the fossil record of the Astacidea is as old, if not older than, that of the Palinura (excluding the Glypheoidea).
The relationship of the infraorder Thalassinidea to the other decapod infraorders has long been the subject of debate. The thalassinoids are considered by some crustacean workers (Burkenroad, 1963; Bauer, 1986) to have important links with the other decapod infraorders and to have played an integral role in the evolution of the Decapoda. They have previously been linked with the infraorder Anomura on both larval (Gurney, 1942; MacDonald et al., 1957) and adult morphological (de Saint Laurent, 1973; Martin & Abele, 1986) evidence. In recent cladistic analyses of adult somatic characters the thalassinoids are depicted as the sister group of the Anomura (Poore, 1994; Scholtz & Richter, 1995).