A range-wide morphological analysis of the Mexican cottontail (Sylvilagus cunicularius) has never been conducted using modern-day multivariate analyses. We took 26 craniodental measurements from the skulls of 155 adults. A discriminant function analysis identified four distinct morphological types. The four morphological types consist of two primary groups. The reddish-colored group occurs on the Pacific coastal plain from Sinaloa to Colima. The individuals from Sinaloa to northwestern Jalisco share various skull features, and those from southwestern Jalisco and Colima have other skull features in common. The constricted coastal plain in northwestern Jalisco appears to separate these two morphological types. The other primary group has a grayish dorsal pelage and occurs in the highlands (the Transvolcanic Belt and Sierra Madre del Sur) and Pacific coastal lowlands of Guerrero and Oaxaca. Within this group, the individuals from the highlands differ from those on the Pacific coastal plain of Guerrero and Oaxaca. Both morphological types have been collected along the western slopes of the Sierra Madre del Sur, one from sea level to 305 m and the other from 822 to 2,225 m. The red and gray cottontails might be distinct species, but we take a conservative approach, awaiting additional lines of evidence. Four distinct subspecies are recognized: Sylvilagus cunicularius cunicularius, Sylvilagus cunicularius insolitus, Sylvilagus cunicularius pacificus (here recognized), and Sylvilagus cunicularius new subspecies.

Using univariate and multivariate statistics, a detailed morphological analysis was conducted on the dentition and skull of long-tailed shrews from east-central United States previously referred to Sorex fontinalis Hollister, 1911, and S. cinereus Kerr, 1792. Based on the examination of 243 shrews, using 25 measurements, these 2 cryptic taxa have parapatric distributions near the border of the Appalachian Mountains and Ridge and Valley physiographic regions in eastern West Virginia, western Maryland, and southwestern Pennsylvania then eastward through central Pennsylvania to the Delaware River, there southward along the Delaware river and bay. Using univariate statistics, about 84% of individuals can be distinguished from each other. Using discriminant function analysis about 98% of individuals are separable with an average jackknife classification error rate of 11%. Both taxa were found within 3 to 54 km of each other at multiple localities without any evidence of morphological intergradation. However, at 1 locality, 5 of 7 specimens collected on the line of parapatry classified as unknowns and appear to have some morphological traits of both species. All morphological and distributional evidence presented is in agreement with those molecular studies that regard S. fontinalis as a distinct species, separate from S. cinereus.

Using univariate and multivariate statistics, I conducted a detailed morphological analysis on the dentition and skull of long-tailed shrews of the Sorex cinereus species group from east-central US. Based on the examination of 109 shrews, by use of 24 measurements, I identified 3 distinct morphological groups. These groups are referrable to: (1) Sorex cinereus cinereus (Masked Shrew) from southeastern New York, northern New Jersey, and northeastern Pennsylvania; (2) S . c . fontinalis (= S . fontinalis of some authors) (Maryland Shrew) from southeastern Pennsylvania, northeastern Maryland, and the Delmarva Peninsula of Maryland, Delaware, and Virginia; and (3) Sorex c . nigriculus (Tuckahoe Masked Shrew) endemic to southern New Jersey. The lower Delaware River and Bay may be a barrier that prevents intergradation between S . c . nigriculus and S . c . fontinalis or they could be 2 distinct species. Intergradation between S . c . nigriculus and S . c . cinereus in central New Jersey likely is occurring, although targeted sampling is needed to verify this. Intergradation between S . c . cinereus and S . c . fontinalis in east-central Pennsylvania is uncertain.

In the United States, Merriam's pinyon mouse, Peromyscus gratus (Merriam, 1898), occurs west of the Rio Grande in mountainous areas of southwestern New Mexico (Catron, Socorro, Grant, Sierra, and Hidalgo counties) and adjacent east central Arizona (Apache and Greenlee counties). In this region, P. gratus is locally sympatric with Peromyscus truei, except in Hidalgo County, where only P. gratus occurs and is rare or now extirpated. I provide methods for distinguishing P. gratus from P. truei, Peromyscus boylii, and Peromyscus nasutus. A morphological comparison of P. gratus from the United States and Chihuahua, Mexico, shows the U.S. population averages significantly smaller in overall cranial size except for a significantly longer palate and much shorter tail. Peromyscus gratus has a closely complementary distribution in the United States with P. nasutus, suggesting its habitat requirements may result in competitive exclusion, one from the other.

The skull morphology of four species of cottontails from the mountains and plateaus of southwestern United States and northern Mexico, currently recognized as Sylvilagus nuttallii, S. cognatus, S. robustus, and S. holzneri (=S. floridanus holzneri), was analyzed using multivariate statistics. Based on 26 or 30 measurements taken on each of 350 adult skulls, and formulation of 16 pooled samples, with each representing a different geographic population, the results show there are two species-level distinctions in skull type. Cottontails from southern Utah, northern and east-central Arizona, and northern New Mexico show morphological overlap among geographically adjacent samples and all have a highly arched skull. These are referable to S. nuttallii. Cottontails from central Arizona, central New Mexico, and western Texas southward along the Sierra Madre Occidental to Durango, Mexico, and Sierra Madre Oriental to central Coahuila, Mexico show morphological overlap among geographically adjacent samples and all have a relatively flat skull. These are referable to S. holzneri. Both S. nuttallii and S. holzneri are polytypic in the study area. Sylvilagus n. pinetis is restricted to the White Mountains of Arizona. Those north and west of the Colorado River in Arizona and Utah are referable to S. n. nuttallii and those from southeastern Utah, northeastern Arizona and northern New Mexico are referable to S. n. grangeri. The population of cottontails previously referred to S. cognatus from central New Mexico is indistinguishable from topotypical S. holzneri from southeastern Arizona as well as populations from southwestern New Mexico, Chihuahua, and Sonora, Mexico. Cottontails previously referred to S. robustus, from the Guadalupe Mountains of New Mexico and Texas, southward in the mountains of western Texas, and the Sierra Madre Oriental to central Coahuila, Mexico, average larger in overall size and, on average, have proportionately larger auditory bullae and are referable to S. holzneri robustus. The relationship of S. holzneri to other mountain/plateau-inhabiting taxa, outside the study area in central and southern Mexico remains to be resolved.

The long-tailed shrew, Sorex dispar Batchelder, 1911, and Gaspe shrew, S. gaspensis Anthony and Goodwin, 1924, from the Appalachian Mountains of North America have been characterized as genetically highly similar, and that one is morphologically a clinal variate of the other, i.e., there is a single species. I measured 24 characters of the skull on 196 shrews from throughout the range of the species. Geographic variation in skull shape and size was not gradual or continuous, but abrupt. These abrupt changes in morphology are associated with major water barriers, primarily the Connecticut River, middle Saint John River, and the Strait of Canso, which separates mainland Nova Scotia from Cape Breton Island. The morphological analyses presented here and previous genetic studies indicate that S. dispar and S. gaspensis are likely conspecific. Shrews with the largest skull occur from North Carolina north to Vermont and are referable to S. d. dispar with S. d. blitchi as a synonym. Shrews from New Hampshire northeast to southern New Brunswick and mainland Nova Scotia have a medium-sized skull and are referable to a new subspecies. Those from northern New Brunswick, Gaspe Peninsula of Quebec, and Cape Breton Island have a small skull and are referable to S. d. gaspensis. The skull morphology of S. d. gaspensis and the new subspecies are more similar to each other than to S. d. dispar. Results of this study differ from those of previous morphological studies because measurement error and within-group variation were reduced, which allowed for visibility of otherwise "hidden" between-group differences, or geographic variation.

Using 26 cranial measurements, we conducted a morphometric study on 113 adult skulls of the currently recognized rabbits Sylvilagus brasiliensis and S. sanctaemartae from Colombia, Ecuador, and Peru. Five morphological groups are identified of which 4 warrant species-level recognition. North and west of the Andes, S. gabbi occupies the lowlands of northern Colombia and northwestern Ecuador. Sylvilagus sanctaemartae is morphologically indistinguishable from S. gabbi and is a synonym of it. Two species live in the Andes of Colombia: S. fulvescens in the Western and Central Andes and S. apollinaris in the Eastern Andes. East of the Andes, Sylvilagus brasiliensis defilippi is found in the tropical Orinoco and Amazon basins of Colombia, Ecuador, and Peru, and S. brasiliensis andinus occurs in the Andes of Ecuador. These 2 subspecies intergrade in the Amotape-Huancabamba Zone of extreme southern Ecuador and adjacent west-central Peru. The eastern distributional limit of S. b. defilippi outside the study area in South America is unknown. An additional 50 skulls of S. gabbi and 14 of S. dicei were examined from Costa Rica and Panama to compare with the rabbits from South America. The data show that S. gabbi exhibits no noticeable geographic variation from Central America to northern Colombia and northwestern Ecuador. Sylvilagus dicei from the Cordillera de Talamanca of Costa Rica and Panama is morphologically most similar to S. fulvescens from the Western and Central Andes of Colombia and they are likely closely related. Methods are presented to identify all taxa and the distribution of each taxon is provided. This taxonomic arrangement is based on few specimens. Accordingly, species boundaries and phylogenetic relationships should be tested by focusing future collecting and morphological and molecular research in high-priority areas such as: 1) along the Colombian-Ecuadorian border from the Pacific Coast, across the Andes into the Amazon Basin; 2) Amotape-Huancabamba Zone of Ecuador and Peru; and 3) Eastern Andes of Colombia.

The effects of military tracked vehicle maneuvers on the vegetation of the Pinon Canyon Maneuver Site (PCMS), southeastern Colorado, were assessed from 1985 to 1987. Tracking decreased plant basal and litter cover and increased bare ground. The immediate effect of tracking was to reduce perennial warm‐season grasses [primarily blue grama, Bouteloua gracilis (H.B.K.) Lag. ex. Griffith] followed by the invasion of annual cool‐season grasses [sixweeks grass, Vulpia octoflora (Walt.) Rybd., and little barley, Hordeum pusillum Nutt.] and annual warm‐season forbs [sunflower, Helianthus annuus L., russian thistle, Salsola iberica Sennen & Pau, and kochia, Kochia scoparia (L.) Schrab.]. In untracked areas, herbaceous plant composition did not change; however, litter increased and bare ground and basal cover decreased. Changes in cover in untracked areas was attributed to above average precipitation and the cessation of domestic livestock grazing. Overall (tracked plus untracked areas), total cover increased on PCMS, but the proportion of annual cover also increased. Woody plant density decreased an average of 9% from 1985 to 1987. Long‐term management of the soil and vegetative resources of PCMS are discussed.