Systematics and Evolutionary Relationships of Spiny Pocket Mice, Genus Liomys

Authors

Hugh H. Genoways, University of Nebraska - LincolnFollow

Date of this Version

11-7-1973

Citation

Special Publication no. 5 (December 7, 1973) / Museum, Texas Tech University.

Comments

Copyright 1973, Museum, Texas Tech University. Used by permission.

Abstract

Introduction

Spiny pocket mice of the genus Liomys are members of the rodent family Heteromyidae and together with the genus Heteromys form the subfamily Heteromyinae. Their geographic range extends from northern Sonora, in western Mexico, and southern Texas southward to the vicinity of the Panama Canal Zone. Within this area, members of the genus occur mainly in dry to arid situations being replaced in areas of rain forest and cloud forest by members of the genus Heteromys. The vernacular name for Liomys is based on the fact that many of their hairs have been modified in the form of stiff, aristiform spines. However, my field companions have other names for members of the genus such as "green mice" and others that would not be appropriate to publish, because of the tendency of these rodents to rot quickly after being trapped and because their thin skin tears easily during preparation.

Although spiny pocket mice are relatively common inhabitants of much of Mexico and Central America, it was not until 1868 that representatives of the genus were described and not until 1902 that the generic name was proposed. Gray (1868) described the first two species that are presently included in the genus Liomys, although he placed them in the genus Heteromys Desmarest, 1817, where they remained until the genus Liomys was described. The two species (Heteromys irroratus and Heteromys albolimbatus) were based on only three specimens from Oaxaca-the holotype of irroratus from an unspecified locality in the state and the two specimens of albolimbatus from La Parada. A third species (Heteromys adspersus), now assignable to Liomys, was described by Peters (1874) from Panama. These three species together with other species of Heteromys known at the time were reviewed by Alston (1879-82) who recognized only two species, Heteromys desmarestianus and Heteromys longicaudatus, in Mexico and Central America. The species irroratus, albolimbatus, and adspersus were placed in the synonymy of H. longicaudatus. A fourth species (Heteromys alleni) was described by Coues (in J. A. Allen, 1881: 187-89) based on a specimen from Rio Verde, San Luis Potosí, although Allen in the same paper expressed some doubt as to the validity of the species in view of Alston's findings. Later, however, Allen (1891: 268-272) concluded on the basis of additional specimens from Brownsville, Texas, and Moroleón, Guanajuato, that Heteromys alleni was a valid species. The next descriptions of members of the genus appeared in 1893 when Thomas (1893a: 329-32, 1893b: 233-34) named three species--Heteromys bulleri, H. salvini, H. pictus--and a subspecies of H. salvini. In the first of these papers Thomas recognized all species named to that time and suggested that species of the genus could be defined by a combination of the characteristics based on number of plantar tubercles and hairy or naked soles of hind feet. Four years later, J. A. Allen (1897) described another species, Heteromys hispidus, from Compostela, Nayarit, and listed known members of the genus, dividing them into three groups based on Thomas' characteristics. It is of interest to note that Allen inadvertently omitted from his listing Heteromys pictus Thomas, 1893, with which hispidus is now considered synonymous.

The genus Liomys was formally described in 1902 by C. H. Merriam with Heteromys alleni Coues as the type species. Liomys was distinguished from Heteromys by Merriam mainly on the basis of the absence of secondary lobes or permanent enamel islands on the molars of Liomys that are present in Heteromys. In the same paper, Merriam described 11 new species and four new subspecies of Liomys and also Heteromys annectens, which is now considered to be a member of the genus Liomys. The following year Elliot (1903a: 146-47, 1903b: 233) described two new species, one from Morelos (Heteromys exiguus) and the other from Veracruz (Heteromys paralius). In these two papers as in his other major works of the next several years, Elliot (1904: 368-82; 1905: 316-23; 1907: 344-47) considered Liomys to be a subgenus of Heteromys. Goldman (1904: 82) characterized a species, Liomys parviceps, from Michoacán and in so doing accorded Liomys generic rank. In the next four years, J. A. Allen (1906: 211, 251, 1908: 652) described two new species and a new subspecies (Heteromys jaliscensis, Heteromys vulcani, Heteromys pictus escuinapae), and placed them in the genus Heteromys without commenting on the status of Liomys.

Therefore, by 1911 when Goldman's comprehensive work, "Revision of the spiny pocket mice (genera Heteromys and Liomys)" appeared, 25 species and five subspecies were recognized in the genus. Goldman recognized nine species and 17 subspecies, besides the nominal races, based upon his study of 1,003 specimens. In addition, he described one new species, Liomys guerrerensis, from Omilteme, Guerrero, and one new subspecies, Liomys irroratus pretiosus, from MetIaltoyuca, Puebla. Although Goldman did not accord them subgeneric rank, he did divide the genus into three species groups--irroratus group, pictus group, and crispus group. This was the last major work to appear dealing with Recent species of the genus until 1948, although in the intervening years descriptions of one species, Liomys anthonyi (Goodwin, 1932a: 2), and three subspecies--Liomys salvini aterrimus (Goodwin, 1938: 4), Liomys irroratus pullus (Hooper, 1947: 47), Liomys irroratus acutus (Hall and Villa, 1948:253)--appeared, and Liomys vulcani was reduced to subspecific status under Liomys salvini (Goodwin, 1946: 374). In 1948, Hooper and Handley (1948) presented a synopsis of the known races of Liomys irroratus and analyzed the trends in geographic variation within the species. Since 1948, the only papers dealing with the taxonomy of the genus were a description of a new species, Liomys pinetorum (Goodwin, 1956b: 2-3), from Chiapas and a new subspecies, Liomys irroratus yautepecus (Goodwin, 1956a: 7-8), from Oaxaca; the latter was subsequently arranged as a synonym of L. i. irroratus (Goodwin, 1969: 148). The only nominal fossil member of the genus is Liomys centralis described by Hibbard (1941a: 349) from the Rexroad Fauna of western Kansas, although other fossil and subfossil material has been reported from San Josecito Cave, Nuevo León, by Cushing (1945: 185) and Jakway (1958: 320), and from caves in southern Tamaulipas by Koopman and Martin (1959: 2, 6) and Dalquest and Roth (1970: 226). Therefore, at the beginning of my study 11 Recent species, 22 Recent subspecies, and one fossil species were recognized in the genus.

The purpose of this study was first to examine the taxonomic relationships of the species of the genus Liomys and, after forming a new systematic arrangement of the taxa, to examine the evolutionary relationship of the species within the genus and to evaluate their relationships to members of the genus Heteromys. The present work is easily divisible into three sections--nongeographic variation, geographic variation, and an assessment of evolutionary relationships--although no one section is independent from the other two. The section on nongeographic variation examines variation with age, secondary sexual variation, individual variation, and variation resulting from molt. Study of geographic variation and specific relationships using both univariate and multivariate statistics has shown that only four of the 11 species recognized prior to this study are actually distinct; these species are Liomys irroratus (seven subspecies), Liomys pictus (four subspecies), Liomys salvini (three subspecies), and Liomys adspersus (monotypic). In addition to these four species, one other monotypic species, Liomys spectabilis, recently described by me (Genoways, 1971) is recognized, resulting in a total of five species within the genus. Because a fossil record of the genus is nearly nonexistent, I have attempted to analyze the relationships of these five species to each other and to members of the genus Heteromys by study of such characteristics as external and cranial morphology, structure of unworn premolars and molars, bacular morphology, morphology of glans penes, comparative karyology, sperm morphology, ectoparasite faunas, structure of pterygoid bone, pelage characteristics, morphology of hind feet, and reproductive patterns. The section on specific relationships also should be used as an appendix to the section on geographic variation because all of these same characteristics were used in determining the species to be recognized. I decided to place all of the specific characteristics in a single section for ease of comparison rather than scattering them throughout the sections on geographic variation within the accounts of individual species.