Afrosoricida Stanhope 1998

Wilson, Don E. & Reeder, DeeAnn, 2005, Order Afrosoricida, Mammal Species of the World: a Taxonomic and Geographic Reference (3 rd Edition), Volume 1, Baltimore: The Johns Hopkins University Press, pp. 71-81 : 71

Afrosoricida Stanhope 1998

Families: 2 families with 19 genera and 51 species in 2 suborders:

Suborder TENRECOMORPHA Butler 1972

Family Tenrecidae Gray 1821 (10 genera with 30 species and 2 subspecies)

Suborder CHRYSOCHLORIDEA Broom 1915

Family Chrysochloridae Gray 1825 (9 genera with 21 species and 27 subspecies)

Discussion: Traditionally included in the Lipotyphla (= Insectivora sensu stricto). Various molecular studies ( Madsen et al., 2001; Murphy et al., 2001 a, b; Springer et al., 1999) and syntheses of morphological and molecular data ( Asher et al., 2003; Liu et al., 2001) support a clade containing tenrecs and golden moles, which Stanhope et al. (1998) named Afrosoricida . This name is inappropriate since this clade does not include soricids, and could lead to confusion with the soricid subgenus Afrosorex Hutterer, 1986 . Noting that Tenrecomorpha Butler, 1972 may be a prior, and more explicit name for this clade following Simpson’s (1945) guidelines for naming superfamial taxa, Bronner et al. (2003) nevertheless accepted Afrosoricida because this name is entrenched in the recent literature.

While Afrosoricida is widely used as a name for a tenrec-golden mole clade (e.g., Cao et al., 2000; de Jong et al., 2003; Douady et al., 2002 a,b; Hedges, 2001; Helgen, 2003 a; Scally et al., 2001; van Dijk et al., 2001; Waddell et al., 2001), some authors have argued that Tenrecoidea McDowell, 1958 is the prior valid name for this taxon (e.g., Archibald, 2003; Asher, 2000; Malia et al., 2002; Mouchaty, 1999). Tenrecoidea, however, was first used by Simpson (1931) as a corrected superfamily name for Cententoidea, and comprised various "zalambdodont" taxa ( Solenodontidae , Potamogalidae , Tenrecidae and extinct Palaeoryctidae and Apternodontidae ) but not chrysochlorids (assigned to a separate superfamily Chrysochloridea ). McDowell’s (1958) restriction of Tenrecoidea to include only golden moles and tenrecs (thus identical in composition to Afrosoricida ) implied a fundamentally different grouping concept to that of Simpson (1931); usage of this name arguably violates Simpson’s (1945:33) guidelines (29-30) for reasonable emendation, and also his recommendation that superfamily names (ending in –oidea) should be avoided. Priority, which Simpson (1945:33) advocated as a deciding criterion only ". when other things are about equal." is thus insufficient to justify acceptance of McDowell’s (1958) revised "Tenrecoidea". Even if accepted, this name should be ascribed to Simpson, 1931 ( Simpson 1945:32 – guideline 23).

Following Simpson (1931), the term "tenrecoid" has also been widely misused as a general name for a vaguely defined grouping of (extinct and extant) taxa characterized by zalambdodonty, even though it has become clear that some of these were not technically zalambdodont, and that zalambdodonty may have arisen independently several times (e.g. Broom, 1916). This further militates against its stricter nomenclatorial use, even at taxonomic ranks below order.

Tenrecomorpha is also a problematic name for the tenrec-golden mole clade (c.f. Bronner et al., 2003). Butler (1972) used this name for a suborder of the Lipotyphla (Insectivora sensu stricto) to accommodate only the Tenrecidae , and assigned chrysochlorids to the separate suborder Chrysochloridea Broom, 1915 . Butler (1988) tentatively accepted a common origin of these families, but argued for their early separation. Eisenberg (1981:63) showed Tenrecomorpha as a separate clade from Chrysochloridea , but elsewhere (p. 113) included chrysochlorids in Tenrecomorpha without giving any character support. While Mouchaty et al. (2000) and Waddell and Shelley (2003) recently used Tenrecomorpha in this context, this name has most consistently been used to include only tenrecs (e.g. MacPhee and Novecek, 1993) and consequently is better applied at the subordinal level to separate tenrecids from chrysochlorids, in accordance with both morphological and molecular data suggesting early phylogenetic divergence of these taxa. Following Bronner et al. (2003), we therefore reluctantly accept Afrosoricida as the most specific name for a tenrecid-chrysochlorid clade.

The supraordinal affiliation of the Afrosoricida remains controversial. Molecular data strongly support an affinity within the Afrotheria, whereas morphological data suggest a closer relationship to lipotyphlans. Lipotyphlan monophyly, however, is only weakly supported by cladistic analyses of morphological data ( Asher, 1999) and phylogenetic analyses of combined anatomical and molecular data strongly support the inclusion of afrosoricids within Afrotheria ( Asher et al., 2003) .