Walter E. Howard

A 19-YEAR STUDY OF MICROTINE POPULATION FLUCTUATIONS USING TIME-SERIES ANALYSIS'

This report involves time-series statistical analysis (including concurrent physical and community variables) of the population dynamics of 4737 voles (Microtus californicus) trapped over 19 yr while emigrating from two study enclosures on a Northern California grassland. Population fluctuations of voles, as documented in the literature as well as in this study, generally cannot be described by periodic or regularly cyclic equations, but rather are either random or occasionally pseudoperiodic where autoregressive correlation explains irregular cycles on the basis of the popu- lations previous demographic history. For these two California vole populations, an autoregressive component accounted for approximately a third of the total variability in the population, while random extrinsic environmental variation explained almost all of the remaining variation. Weather played a key determinant role, influencing microtine populations both directly, and indirectly through an effect on vegetation. The distinction between periodic, pseudoperiodic, and random fluctuations in vole populations, and indeed in wildlife populations in general, cannot be dismissed as merely a question of semantics, because each entails a specific ecologic interpretation. The demographic characteristics of microtine rodents, although fluctuating more dramatically, were similar to those of other rodents.

SHEEP KILLING BEHAVIOR OF CAPTIVE COYOTES

One to 4 pen-reared coyotes (Canis latrans) and 1 to 6 sheep were released in 38 tests within a 1,600-m2 pen and their interactions were observed from a blind. One or more sheep were killed in 20 tests. Latency to attack averaged 47 minutes in 21 tests. Eight of 11 coyotes individually killed 16- to 32-kg lambs. The coyotes attacking sheep most frequently were 2-year-old males and females paired with these males. Yearling males attacked less frequently and unpaired females did not attack. Six of seven kills by male-female coyote pairs were made primarily by the male. Each coyote that killed sheep clamped its teeth in or near the larynx region of the sheep and held on until the sheep succumbed. This technique left characteristic tooth marks and hemorrhaging. The sheep appeared to die primarily of suffocation. Kill- ing time averaged 13 minutes in 20 tests. Food-deprived coyotes fed on the kills for an average of 25 minutes and consumed an average of 2.0 kg/kill. The body parts most frequently eaten from kills were digestive organs and muscle from the hind leg, neck, shoulder, and head. Three of four coyotes fed before tests killed sheep but did not feed on the kills. Defensive behavior by the sheep appeared to deter coyote attacks in 12 tests.

Microtine Population Fluctuations: An Ecosystem Approach Based on Time-Series Analysis

(1) This time-series analysis describes the bimonthly population dynamics of 5308 voles (Microtus californicus) trapped in pit traps set continuously for over 21 years, while emigrating from two study enclosures on a Northern California grassland. (2) A simple model is introduced to quantify seasonality in microtines and to relate seasonal population fluctuations to concurrent physical and community variables. Seasonal variation represented 33 to 46% of the observed total variability. (3) Weather played a determining role in nonseasonal microtine population fluctuations, both directly and indirectly, through an effect on vegetation. (4) Correlations between indicators of solar activity and nonseasonal microtine fluctuations may represent either a direct effect of those cosmic variables or a synchronous environmental effect of the Eltonian type.

Sheep losses to predators on a California range, 1973-1983.

Predation at the University of California Hopland Field Station was evaluated for an ll-year period beginning in 1973. Of those lambs placed on range, an average of 2.7% were killed each year by predators. An average of 1.5% of the ewes were killed. When the number of missing animals which were killed was estimated, the average annual predation rate for lambs and ewes killed was 10.4 and 3.8%, respectively. For all known ewe and lamb deaths, respectively, 45% and 26% were caused by predators, 14% and 28% died from causes other than predation, and 41% and 46% died from unknown causes. Of those sheep killed by predators, 89% were killed by coyotes (Canis r and superintendent and animal resource supervisor. University of California Hopland Field Station, 4070 University Road, Hopland, California 95449. I he authors thank D.M. Center. R.M. Dally. and C.E. Vaughn for reviewing the manuscript. Manuscript accepted November 20. 1984 livestock losses occurring during a few years. While short-term studies provide useful information, conclusions based on such research must be made with caution because factors such as predator density, environmental conditions, and availability of alternate prey vary annually and may influence predation. Generally, conclusions regarding livestock losses to predators are more reliable when long-term data are used; however, such data are rare. The purpose of this study was to evaluate livestock and economic losses to predators for a I l-year period beginning in 1973 for a. sheep operation in the northern coastal area of

COYOTE ESTROUS URINE VOLATILES

Samples of female coyote urine were taken once or twice each week during the winter and spring for two years. Headspace analysis was employed with Tenax GC trapping and GC-MS. Tenax trapping was started in less than 1 hr after sampling, and mild conditions were used to minimize losses of highly volatile and labile compounds. Thirty-four compounds were identified. They include sulfur compounds, aldehydes and ketones, hydrocarbons, and one alcohol. The principal constituent is methyl 3-methylbut-3-enyl sulfide, which usually comprised 50% or more of the total volatiles observed. The concentration of many constituents varied widely. This appeared to be quasiperiodic for five of the constituents, with a period of a few weeks, and with pronounced maxima at the peak of estrus. Apparently these compounds are 3-methyltetrahydrothiophene, methyl 3-methylbutyl sulfide, octanal, dodecanal, and bis(3-methylbut-3-enyl) disulfide. One or more of these compounds may have pheromonal activity in coyote relationships.

A Drift-Fence Pit Trap that Preserves Captured Rodents

The efficiency of lead fences of hardware cloth to direct rodents into funnel traps is discussed; a new type of pit trap that excludes birds and automatically preserves captured rodents is described. Rodents entering the trap fall through a trap door into a gallon jar containing alcohol. The pit traps primarily catch dispersing rodents; resident populations do not appear to be destroyed. Each pit trap yielded 2.5 to 45 times as many rodents as did Sherman live traps set for only one night.

Amount of Food Eaten by Small Carnivores

Bassariscus astutus .—An adult female ring-tailed cat was captured in a bakery in Madera, California. Mrs. Stella Garner kept it as a pet for two months before giving it to me for study at the San Joaquin Experimental Range. She fed the cat raisins, prunes and table scraps. She said it did not like bread but that it ate some cooked and fresh meats. The ringtail weighed 744 grams when I took it to the Experimental Range, but it soon gained and fluctuated between 900 and 1,000 grams on a diet of rodents and birds. All food items fed the ring-tailed cat for nearly six months were weighed and the amount consumed was ascertained. To determine daily consumption, a period of 44 days was selected. In that period the cat consumed 3,988 grams of rodents and birds for an average daily ration of 90.6 grams, which is about equal …

Canid sex attractant studies.

Behavioral studies using anestrous female beagles, and olfactory tests with dogs in the absence of females, indicate that methylp-hydroxybenzoate cannot be considered a key sexual attractant for male beagles, even though this compound has been found in estrous vaginal secretions.

Aversion to strychnine sulfate by Norway rats, roof rats, and pocket gophers

Abstract Sprague-Dawley and wild Norway rats ( Rattus norvegicus ), roof rats ( R. rattus ), and pocket gophers ( Thomomys bottae ) were employed in drinking trials of distilled water and 0.01, 0.05, and 0.5% strychnine sulfate solutions for periods of 10 minutes (rats), 20 minutes (pocket gophers), and 24 and 48 hours for all species. Both the wild Norway and the Sprague-Dawley rats apparently found strychnine highly unpalatable, especially the 0.5 and 0.05% concentrations. They were able readily to discriminate even the weakest strength and acquired a marked aversion to all the strychnine solutions, whereas the roof rats developed an aversion only for the 0.05 and 0.5% concentrations and pocket gophers never learned to avoid completely even the 0.5% strength, though they were made ill. These tests suggest that strychnine sulfate may be usable as a rodenticide for R. rattus , if it is prepared at the proper concentration. Twelve of 15 pocket gophers and 1 of 5 roof rats drank lethal amounts of the solutions, whereas none of the 27 laboratory and wild Rattus norvegicus did.

MOLT IN THE POCKET GOPHER, THOMOMYS BOTTAE

hair, if apparent in the two seasonal pelages of molting gophers, were also entered on the card. A total of 789 of these recordings was obtained from 198 individuals. Fourteen gophers living on the plot were dyed as an additional procedure, but this was at a time when trapping was less frequent. Nevertheless, during the ensuing fourteen months, ten of the dyed animals were recaptured a total of 21 times. Kill-trapped gophers.-Flat study skins were prepared from 355 gophers trapped on and around the study plot at the Experimental Range. Only a few skins were obtained from June to September-the season when rangeland gophers show little evidence of surface activity and are difficult to trap. The area of pigmentation on the flesh-side of the skins, indicative of follicular activity, was measured with a planimeter. Similar measurements were also made of the area of each skin that was in summer or winter pelage. The position of molt lines and other data were also obtained from nearly 100 animals collected from irrigated alfalfa fields at Davis. Dyed and bleached gophers.-A technique of dying hair black and of bleaching black hair to light brown was developed by the senior author and D. G. Constantine in 1949 (unpublished). The technique had been employed on studies of molt in cattle and swine, and it was found applicable to other mammalian species as well. To dye the animals, a black human hair dye with an oil base was used. Since it was not water soluble, the dye persisted until the hair was shed. It was diluted with an equal amount of 30-volume (3 %) hydrogen peroxide. Granulated soap was added until a thick liquid was obtained. The soap made application easier,