The Mammals of Warren Woods, Berrien County, Michigan Occasional Paper of the Museum of Zoology, Number 86 — Story, Setting & Ideas

A Habitat-First Framework

Rather than organizing by species, Dice structures his study around habitat types. He maps the preserve's topography—ravines draining to the Galien River, flood-plains, beech-maple forest—and assigns each mammal to a specific environment. For example, the prairie vole (Microtus ochrogaster) appears in sedges and rushes of cleared upland, while the northern white-footed mouse (Peromyscus leucopus noveboracensis) is recorded across multiple habitats. This method allows Dice to infer ecological preferences without speculation. Readers should note how the habitat list functions as a key: each entry corresponds to a later species account, making the paper a reference work as much as a narrative.

Breeding Data as Evidence

Dice's tables of reproductive records are the study's empirical core. For each captured individual, he notes age, number of embryos, and embryo length in millimeters. The entry for Peromyscus leucopus noveboracensis on July 22: “adult, 5, 10 5 mm.” exemplifies his precision. These data points, collected over a single summer (July–August 1920), reveal breeding peaks and litter sizes. Dice does not interpret these patterns; he presents them as raw material for future analysis. Readers should examine the tables for species-specific rhythms—for instance, the house mouse (Mus musculus) shows a wide range of embryo counts, from 7 to 11, while the Norway rat (Rattus norvegicus) reaches 9.

A New State Record and Local Knowledge

Among the findings, Dice announces a new state record: the prairie vole (Microtus ochrogaster) in Michigan, previously misidentified as the Pennsylvania vole. This claim is supported by specimen data from Warren Woods and a nearby wheat field. Dice also incorporates information from local residents—George R. Fox, director of the Warren Foundation, and trapper William Schmidt—to list mammals not observed on the preserve, such as the Virginia opossum and star-nosed mole. These secondary sources extend the study's geographic scope while acknowledging its limits. The inclusion of anecdotal evidence alongside systematic trapping shows Dice's pragmatic approach to field biology.

Reading the Tables and Maps

The paper includes a sketch map of Warren Woods (Fig. 1) indicating mammal habitats, though the map itself is not reproduced in the text. Dice's verbal description compensates: he notes the distribution of flood-plain forest, beech-maple forest, and cleared areas. The tables of trapping records are dense but reward careful reading. For example, the entry for Zapus hudsonius hudsonius (Hudson Bay jumping mouse) on July 27 lists only “subadult, 0,” suggesting a single capture. Such sparse records contrast with the dozens of entries for common species. Readers should cross-reference habitat descriptions with species occurrences to understand Dice's classification logic.

Dice's paper is best approached as a primary source document rather than a finished ecological synthesis. The data tables and habitat lists invite comparison with modern surveys of the same preserve. Readers interested in historical ecology will find value in the precise measurements and local informant accounts, which together capture a moment before further development altered Warren Woods. The study's value lies not in conclusions drawn, but in the meticulous record it preserves.

Reading Dice’s careful notes on those Michigan mammals, I kept thinking how quietly he documented a place that would soon change. That same tender attention to a single creature’s place in the world lives in A New Subspecies of Wood Rat (Neotoma mexicana) from Colorado — Reading Companion. It felt like holding a small, precise photograph of something almost forgotten.