This is the authors' abstract. We don't add key points for this paper.
Median, standard error, and standard deviation of a season of can be estimated by probit analysis from the observed ratio of newborn to females at three or more times during the season, and once at its end. The interval between sampling times need not be constant nor the sample sizes equal. This method circumvents the need to keep a population under constant observation, yet yields estimates of the frequency distribution of birth dates as precise as those from a captive population. Applied to data collected in 20 field hours from impala (Aepyceros melampus) in Zambia, the method estimated the median date of birth as 16 November with a standard error of less than a day, and the standard deviation of birth dates as 7 days. J. WILDL. MANAGE. 38(3):552-556 In 1953, Zuckerman (p. 933) complained of the dearth of precise information on the season of of wild mammals. Little has changed since then. Sadlier (1969) in reviewing the ecology of mammalian reproduction was unable to point to a single instance of mean, standard error, and standard deviation being measured for the season of of a naturally occurring mammalian population. The problem is largely technical. The season of can easily be determined for a captive population by recording the number of that occur each day, and calculating directly from these the mean, standard deviation, and standard error of the birth dates. But this method is a fragile flower defying transplantation to the field. Natural populations can seldom be kept under constant and plenary observation. Field observations on the timing of are usually incomplete and can be summarized no more precisely than as a subjective evaluation of or as the earliest date at which newborn were seen. Both peak period and onset of births lack the precision needed to answer large questions about reproductive ecology, physiology, and evolution. They are seldom precise enough, for example, to determine the regression of birth dates on latitude or day length, and they are unequal to exploring the relationship between variance of birth dates and environmental favorability, or the role of normalizing selection in stabilizing the mean and spread of these dates. The alternative of using only birth records from zoological gardens is unsatisfactory for two reasons. Firstly, the number of zoos with sufficient birth records for sufficient species is depressingly small, and secondly, there is some evidence (Caughley 1971) that under the relaxed selection pressures prevalent in zoos the season of can expand and drift away from its natural timing. No entirely satisfactory alternative to field measurement suggests itself, but the technical difficulty of measurement tends to frustrate such efforts. Three methods of circumventing the technical problem of field measurement have been advanced. If a growth curve is available, the age of fetuses can be extrapolated forward to provide a sample of birth dates as was done by Laws and Parker (1968) and Hanks (1969) for elephants. Conversely, the ages of newborn can be extrapolated backwards to give the same information, a technique used by Poole and Pilton (1964) to define the season of for kangaroos. These two methods 552 J. Wildl. Manage. 38(3):1974 This content downloaded from 157.55.39.238 on Sat, 02 Jul 2016 05:33:17 UTC All use subject to http://about.jstor.org/terms ESTIMATING BIRTH DATES Caughley and Caughley 553 Table 1. Frequencies of newborn and of females other than newborn during the season of births. D E A B C Breeding % Day Newborn Females females to date
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Caughley et al. (1974) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: