Olfactory Responses of Deer Mice to Douglas-fir Seed Volatiles

An attempt was made to identify the olfactory cues produced by Douglas-fir seeds which attract deer mice (Peromyscus maniculatus) to the seeds. The olfactometers used are described, and the merits of different statistical analyses of the data are discussed. The odors produced by whole Douglas-fir seed and by the endosperm were preferred among the fractions tested to date. Deer mice were repelled by Douglas-fir turpentine, cedar oil, and, to a lesser degree, one extract. Since the first attempts at artificial reforestation of logged or burned forests by direct seeding of coniferous species, birds and small mammals have been a major problem. In western Oregon in 1909 four thousand acres were sown (presumably hand-broadcast) to Douglas fir, Sitka spruce, and other conifers (Black, 1969). After near-complete failure (measured by a seedling count the following spring), an attempt was made to control the depredating animals. Then a second attempt at direct seeding failed. No large-scale seeding of Douglas fir was tried again until the late 1940's, when some new chemicals and methods were developed to control the seed eaters. That the consumption of conifer seed by animals poses a threat to the success of natural and artificial reseeding has been well documented (Gashweiler, 1967; Hooven, 1958; Spencer, 1954). Pregermination losses of seeds from all biotic factors, including vertebrates, invertebrates, molds, and fungi, often range from about 50 to 100 percent (Boyer, 196A; Laurence and Rediske, 1962). In the 75,000-acre Tillamook burn in western Oregon, 68 percent of the area was considered inadequately restocked (below 20 percent) by a massive reseeding operation (Black, 1969). Seed loss is clearly an economic liability. A number of vertebrate species are involved in the depredation of conifer seeds. Among the most common are Juncos (Junco oregonus), tree squirrels (Tamiasciurus sp.), ground squirrels (Spermophilous sp.), chipmunks (Tamias and Eutamias sp.) , shrews (Sorex and Blarina sp.), and deer mice (Peromyscus maniculatus). Deer mice, considered the most important vertebrate predator of conifer seeds because of their abundance, ubiquity, and voracity, were the subject of this investigation. They are reported to eat 200 to 350 seeds per night in captivity (Hooven, 1953, 1958; Packham, 1970). Success in preventing seed depredation has been moderately good from both toxic baits and seed treatments (repellents). Most of these compounds, however, are no longer registered for this purpose. This paper seeks efficacious seed protection in another direction. It has been shown that deer mice use olfactory cues to locate conifer seeds (Howard and Cole, 1967; Howard, Marsh and Cole, 1968). If the chemical or chemicals providing these olfactory cues can be identified, many new and innovative control techniques might then be attempted. Possible applications of this knowledge are discussed later. Howard and Cole (1967) and Howard, Marsh and Cole (1968) demonstrated that deer mice detect conifer seeds through olfactory cues -primarily if not exclusively. Single seeds buried one to three inches deep under peat moss in petri dishes were readily located and dug up by deer mice. Controls of empty petri dishes eliminated bias from tactile or humanodor cues. In identical tests under subdued light (0.25 foot-candles) and total darkness, detection did not significantly differ. This eliminated the possibility that visual cues were important in seed detection. Since deer mice use olfaction to detect conifer seeds, this study attempted to identify the olfactory cues involved. Some of the applications of knowledge gained about these olfactory cues are presented by Radwan (1970). He states that, once the active olfactory components are isolated and identified, it might be possible either to extract and remove *Current address is Montana Department of Livestock, 3617 Fassett Drive #2, Missoula, Montana 59801.