Project topics and materials
Get Complete Project Material File(s) Now! »
Population Size
The size of a population depends upon three factors or forces operating upon it–reproduction, mortality, and movement. Reproduction has an additive effect on population numbers, mortality a reducing effect, and movement either one or the other. The details of reproduction will be discussed in the following major section.
Several methods were used to arrive at estimates of the population size of 0. irroratus but none were satisfactory. Three methods, the Lincoln Index, the Schumacher-Eschmeyer Method, and the Schnabel Index, were used as a noncumulative basis of estimates for each time the traps were examined and then a mean estimate calculated for the entire session. The differences, however, between the morning and afternoon estimates were too considerable for these methods to be of valid use. The Lincoln Index was used on a cumulative basis, combining the trapping results from each month before calculation, and appeared to be satisfactory when trapping \vas conducted on a regular basis with a minimum interval between sessions (Fig. 13). The population estimates during the first year of trapping were reasonably w·ell in line with the total number captured. It was only after the first year, when intervals between trapping sessions became greater, that the standard error of the estimate became greater and that estimates did not appear to be within reason.
Hayne (1949) stated that a lengthy interval between the preliminary marking period and the subsequent period when the population is sampled inevitably results in an overestimation of the size of the population.
This was obviously the case in using the Lincoln Index on a cumulative basis for 0. irroratus. If trapping is conducted in approximately the same way during each trapping period in a long-term study a population estimate is basically a duplication of the actual number captured, the latter merely being a lower number. The most important information is whether the population fluctuated from one period to another and why; complex means of estimating populations in cases such as the present study are unnecessary. Therefore, with the exception of the following paragraph on density, discussions in this report will refer to the actual number of individuals captured on the study grid~ as shown in Fig. 14.
The density estimates here used are based upon the actual number of animals captured (Fig. 14) and the area which 0. irroratus occupied, excluding single capture stations (Fig. 12). This area was equivalent to 1,575 ha (15 750m2). The density estimates, which are undoubtedly low because they are based only upon the trappable population, ranged from a low of 17 per hectare to a high of 72. The mean density of £• irroratus from all the trapping periods was 36 animals per hectare, or one animal per 278 m2 • The density estimates per hectare for each trapping period were as follows:
1970, March-17, April-27, May-34, June-29, August-28, September-18, October-23, November-24, December-41; 1971, January-39, February-41, May-72, July-58, November-58; 1972, May-33. Roberts (1935) stated that Otomyinae are usually plentiful wherever they occur. Ansell (1960) said the same of 0. angoniensis in Zambia. Hanney (1965) stated that in specific areas in Malav1i Q• angoniensis is the most important species in biomass, but this speaks more for its size than it does for numbers. Dieterlen (1968) stated that a true indication of the population size of £· tropicalis cannot be reached because they are difficult to trap. This would be true only in respect of snap-traps, as will be discussed below under Response to traps (see Individual Behavior).
The population size of 0. irroratus on the grid fluctuated considerably during the study period (Fig. 14). The smallest number of animals captured was 27 in March 1970 and the greatest number was 113 ln May 1971. In 1970 the population reached its peak in May. It appears that the peak during 1971 was also in May, but no trapping was conducted in the months immediately before and after in order to define the exact peak time. Excluding the first trapping period in March the population in 1970 appeared to reach its minimum in September (29). The low probably would have been at approximately the same time in 1971, and it is surmised that the trappable population then would have been between 70 and 80 individuals. The peak in May is the result of this being the end of the period when the young are born and the population has been steadily increasing up to that time. The low in September follo\vS the winter nonbreeding period and precedes the appearance of the first young of the season. September follmvs the winter and is also the period when the food supply is at its poorest \vhile the mortality rate is likely to be at its highest (see Mortality).
The number of animals captured during Nay 1971 ( 113) \vas more than twice that captured during Hay 1970 (53) and Hay 1972 (52). It is difficult to ascertain the direct causes of the increase in the population to its peak in May 1971 as these are usually multiple and complex. Favorable climatic conditions are often considered the basic cause of population increases. The 12-month period beginning in July 1970 and ending in June 1971 can be used as a basis for easy reference in examining temperature and rainfall. The overall mean temperature in nearby Pretoria during that period was less than 0,02 per cent above normal (Table 2). The rainfall at the Reserve was 13,7 per cent above normal (Tablel). Considering only these two variables it appears as if the above average rainfall \vas the most important factor in the population increase. Temperature cannot be excluded as an important factor as it is possible that despite above normal rainfall there is also a need for normal or above normal temperatures at the same time.
All the adult females captured ln November 1971 were either gravid or lactating and in the next trapping period (May 1972) the population was expected to be tremendous. Instead, the population in May 1972 was greatly reduced. It is not certain what caused this decline, but in January 1972 the rainfall was 92,4 per cent above normal for the month with most of it occurring on two consecutive days. TI1e result was that the entire study grid was flooded. The area of the grid that suffered most was the lowest or « wet » part. Evidence indicates that the water level on the low part during the flood may have been 1 to 1,5 m. The force of the flood was considerable as two very large and heavy fallen tree limbs found lying in the grid aftenv-ards would have had to be carried from at least 200 to 300 m away. About 1 km upstream, where the stream enters the Reserve, a 250 m section of strong security fence was washed m.;ray. One of the most important factors concerning the flood would have been its suddenness. If the water rose gradually the animals may have had time to escape. The data kept at the Rietvlei ~laterworks, 3 km dm .. mstream from the grid, indicates, ho\vever, that the water level rose very suddenly, lessening any chance the animals had to escape. This catastrophic event therefore appears to have been responsible for the tremendous population reduction between November 1971 and May 1972 (Fig. 14). Of the 52 0. irroratus captured during May 1972, 18 (34,6 per cent) were marked prior to the flood. This suggests that some individuals did manage to escape and later return, many even returning to their previous home ranges (see discussion below under Home Range). Because there \·lere no juveniles captured in May 1972 (Fig. 16) it appears as if the flood may have disrupted the social structure and along with it the breeding potential. Also, the rainfall for each month following the flood, up to and including May, was below average. This could also have affected breeding, as discussed above. Adults were most abundant in the May 1972 catch (Fig. 16), indicating that the area was probably repopulated as a result of immigration. Schulz (1953) found that o. irroratus readily moved into an area depopulated by trapping. Flooding could also have been the reason for the relatively low number of animals, and the absence of juveniles, in March 1970, as a similar flood had occurred in October 1969, during v1hich 178 mm of rain fell. This amounted to 266 per cent above the normal (67 mm) fo.r the month of October (see Table 1). The majority of the rainfall during that month (112 mm, 63 per cent) occurred on one day. On that particular:. day the area which was to become the study grid was flooded. The effect of flooding upon populations of rodents has been reported by several authors (Grinnell, 1939; Blair, 1939; Stickel, 1948; McCarley, 1959; Ruffer, 1961). Grinnell (1939) and Blair (1939) concluded that flooding may be one of the most important causative factors of fluctuations in population numbers of terrestrial species in low-lying areas. Blair stated that it may even produce virtual extermination of certain species in some parts of their range. Stickel (1948) and Ruffer (1961) found that floods had little effect on the population size of Peromyscus leucopus. McCarley (1959) stated that short-term flooding produced no detrimental effects on the mice, but that flooding over a three week period produced a 70 per cent decrease in the population. The probable reason that there was less effect on the populations in these last three studies was that the rodents concerned were either arboreal or semiarboreal.
Composition
Populations have a definite structure and composition which are constant at any specific time, but which vary with age. A population of rodents at any given moment has a particular composition of males, females, adults, subadults, and juveniles. However, as little as one month later the structure and composition will have changed because of births, deaths, and ageing of individuals.
The age ratio of 0. irroratus in the study grid was calculated for each trapping session and was based upon the following categories: Juvenile–An individual too young to breed and still distinguishable from breeding adults on external characters. These individuals are equivalent to those in age class 0, usually weigh less than 50 g, and range in age from birth to five weeks.
Subadult–An individual which has not bred but which externally resembles an adult. These individuals are equivalent to those ~n age class I and part of class II, usually weigh between 50 and 90 g, and range in age from five weeks to three months.
Adult–An individual which has bred. These individuals are equivalent to those in the last part of age class II and older, usually weigh in excess of 90 g, and range in age from three or four months to about two years. The fluctuations in size of each age group from one trapping sess~on to another are represented in Fig. 16. No juveniles were present ~n March, August, and September 1970, July 1971, and May 1972. The absence of juveniles from July to September is easily explained.
This represents the winter period from approximately two months after breeding ceases (Hay), until spring yJhen breeding has begun (August), but before the first young have been born. The absence of juveniles in March 1970 and Hay 1972 ~n both instances represented a similar situation. The situation was preceded by a month of exceptional, above average rainfall (five and four months before, respectively–Table 1), v1hich resulted in flooding of the grid, This was followed in both cases by a period of below average rainfall (Table 1). It is possible that the flooding disrupted the lives and social structure of the animals sufficiently to cause a halt in breeding for some time afterwards. Therefore, no juveniles had entered the population, The percentage of adult, scrotal males during Harch and April 1970 was low for that time of the year (Fig. 20), and is a good indication that breeding in males was more seriously disrupted than thBt of the females. 1ne below average rainfall which followed may also have been the key factor in the interruption of breeding, or the interruption could have been caused by a combination of the t\vO factors. In 1970 there were juveniles present in April and May but the rainfall was below average prior to April and average in April. This would appear to rule out lack of rain as being the sole limiting factor. It is possible that the absence of juveniles in the first trapping session, March 1970, was the result of the differential trapability of animals of different ages. Wiley (1971) stated that adults of the eastern woodrat, Neotoma floridana, generally dominated the subadults and juveniles. It also stands to reason that subadults would in turn dominate the activity of juveniles. Watts (1970) found the presence of large male red-backed voles, Clethrionornys gapperi, in some way inhibited the first approach to traps by juveniles. Because March \vas the first time the animals in the grid were ever subjected to traps it is possible that adults and subadults totally dominated juveniles in response to the traps. Dieterlen (1968) stated that the loss of young £· tropicalis must be relatively high because so few were caught. It 1s possible that \vhat Dieterlen experienced v1as not a high loss of juveniles, but a similar differential trapability in terms of which juveniles were dominated in their response to traps. The percentage of marked animals recaptured in the follm.ving month (April 1970) was low (47, 7 per cent) indicating that initial response to traps ‘tvas poor. The animals captured would, therefore, have been the more dominant ones and possibly any juveniles were excluded from capture. During Harch there was one less trap at each station on the « wet » part of the grid, which could have limited the chance of capture, especially if this trap dominance of juveniles does occur in _2. irrorat~.
The highest percentage of juveniles 1n the population occurred in May, October, and December 1970, and May and November 1971 (Fig. 16). The abundance of juveniles appears to be at its highest near the beginning and at the end of the breeding season, which begins in August and ends in Hay. The percentage of subadults in the population ranged from a low~n October 1970 (8 per cent) to a high in March 1970 (63 per cent) (Fig. 16). The high percentage of subadults in March is probably the result of unusual or abnormal conditions (see previous discussion concerning flooding and juveniles). Disregarding March, subadults were most numerous in June 1970 (57 per cent) and July 1971 (53 percent). The high percentage during these t »tvo months could be a result of the following: lack of or reduction in the number of juveniles born into the population within the past five weeks; and maturing of an increased number of juveniles born more than five « tveeks before. The percentage of adults in the population ranged from a low ~n June 1970 (36 per cent) to a high in September 1970 (86 per cent) (Fig. 16). The low in June is primarily the result of an increased number of juveniles and subadults within the population. The high in September was the result of the maturation of the previously (MarchAugust) encountered subadults to adults and the absence of any new young animals being born into the population.
INTRODUCTION
DESCRIPTION OF STUDY AREA
MATERIALS AND METHODS
ACKNOWLEDGEMENTS
TAXONOMY AND MORPHOLOGY
-Taxonomic Status
-Morphology
-CONTENTS
-Distinction Between 0. irroratus and o. angoniensis
-External and Cranial Measurements
-Age and Seasonal Variation
DISTRIBUTION
-Geographic
-Ecological
POPULATION DYNAMICS
-Population Size
-Composition
-Mortality
-Dispersal, Home Range, and Territoriality
REPRODUCTION
-Reproductive Capacity
-Litter Size
-Gestation Period
-Sexual Maturity
-Breeding Season
-Reproductive Potential
POSTNATAL GROWTH AND DEVELOPMENT
-Description at Birth
-Physical Grmvth
-Behavioral Development
ACTIVITY
-Daily Cycle
-Variations and Relationships
BEHAVIOR
–Individual Behavior
-Locomotion
-Exploration
-Rest and sleep
-Grooming
-Feeding
-Elimination
-Marking
-Nest building, burrowing, and runway formation
-Response to traps
–Social Behavior
-Communication
-Adult-offspring behavior
-Adult interactions
-Sequences of behavior during encounters
-Sexual behavior
-Territorial behavior
ECONOMIC IMPORTANCE
-Influence on the Environment
-Relationships with Man
-Competition with man
-Competition with domestic animals
-As a source of food
-As a laboratory animal
-Public health implications
SUMMARY AND CONCLUSIONS
LITERATURE CITED
FIGURES
TABLES
