The Epidemiology of Violent Mortality in Nairobi: A Study on  
Homicides, Accidents, and Suicides  
Wangai Kiama, M Med (Path), Frcpath (ECSA)  
Department of Pathology, Egerton University, P.O BOX 536-20115, Egerton-Njoro, Kenya  
Received: 21 November 2025; Accepted: 28 November 2025; Published: 03 December 2025  
ABSTRACT  
Background: Violent deaths, including accidents, homicides, and suicides, are major contributors to mortality  
in Nairobi, Kenya. The high prevalence of these fatalities presents a significant public health challenge,  
necessitating a better understanding of their distribution and contributing factors to inform effective intervention  
strategies.  
Objectives: This study aimed to determine the prevalence, causes and socio-demographic distribution of violent  
deaths in Nairobi, identify the main causes of these deaths (accidents, homicides, and suicides), and assess their  
relative impact on the city’s overall mortality.  
Methodology: A descriptive prospective study was conducted at the City Mortuary in Nairobi from June 1,  
2009, to May 31, 2010. A total of 2,566 autopsies were performed during the study period, of which 2,278  
(88.8%) were attributed to violent causes.  
Results: The study showed that violent deaths constituted 88.8% of all autopsies conducted. Accidents were the  
leading cause of violent deaths, accounting for 46.7% (1,064 cases), followed by homicides at 43.5% (990 cases)  
and suicides at 9.8% (224 cases). The overall prevalence of violent deaths in Nairobi was 61.6 per 100,000  
population. Prevalence rates for specific categories were 28.8 per 100,000 for accidental deaths, 26.8 per 100,000  
for homicides, and 5.4 per 100,000 for suicides.Statistical analysis using chi-square tests showed that violent  
deaths were significantly more common than natural deaths (χ² = 1857.89, p < 0.01). ANOVA results further  
indicated significant differences across the three categories of violent deaths (F = 57.32, p < 0.01).Violent deaths  
affected both males and females aged 10–60 years, with notable variation in age and gender distribution across  
accidents, homicides, and suicides. The mean age of victims was 32.8 years (range: 10–59). Males accounted  
for 2,095 (92%) of the 2,278 violent deaths, while females accounted for 183 (8%), resulting in an overall male-  
to-female ratio of 11.4:1.The 30–39 age group was the most affected, representing 914 deaths (40.1%), followed  
closely by the 20–29 age group with 844 deaths (37.1%). Together, these two age groups made up more than  
77% of all violent deaths. The male-to-female ratio was highest in the 20–29 age group (17.7:1), followed by  
the 30–39 age group (10.4:1).  
Conclusion: Violent deaths are a significant public health issue in Nairobi, with accidents and homicides being  
the leading causes. The findings emphasize the need for targeted public health interventions, including improved  
road safety measures, strengthened law enforcement, and expanded mental health services to prevent suicides.  
Recommendations: To reduce violent deaths, efforts should focus on enhancing road safety through better  
infrastructure and law enforcement, improving crime prevention via effective policing and addressing socio-  
economic disparities, expanding mental health services, and investing in programs to alleviate poverty,  
unemployment, and social inequality.  
Keywords: Violent deaths, homicides, suicides, accidents, age and gender, Autopsy, Nairobi.  
Page 262  
INTRODUCTION AND LITERATURE REVIEW  
Violent deaths, encompassing homicides, suicides, and accidents, represent a major public health challenge  
worldwide, especially in urban areas. These regions often experience a convergence of factors such as high  
population density, economic inequality, and social tensions, which can exacerbate the prevalence of such  
deaths. Nairobi, Kenya's capital city is no exception. With an estimated population of 3.7 million during the  
study period (June 1, 2009 – May 31, 2010), Nairobi faces increasing concerns over violent crime, road traffic  
accidents, and mental health issues, which contribute to a high rate of violent deaths.  
Nairobi’s rapid urbanization and socio-economic challenges—such as poverty, unemployment, and inequality—  
play a central role in driving violent fatalities. Gang violence, domestic homicides, and road traffic accidents  
(RTAs) are prominent contributors. Additionally, suicides are on the rise, pointing to a growing mental health  
crisis. The City Mortuary, the primary facility for medico-legal autopsies in the city, provides a unique and  
valuable data source for understanding the prevalence and causes of violent deaths. On average, the mortuary  
handles 200 bodies per month due to unnatural deaths, underscoring the urgency of addressing this public health  
issue.  
Globally, violent deaths represent a major burden. The World Health Organization (WHO) reports that  
homicides, suicides, and RTAs account for millions of deaths annually. In the African context, cities like  
Johannesburg and Lagos face alarmingly high homicide rates, and Kenya grapples with rising road traffic  
fatalities. Limited resources for law enforcement, inadequate mental health care, and poor infrastructure  
exacerbate the situation in many African nations.  
This study highlights the pressing need for comprehensive public health interventions targeting road safety,  
crime prevention, and mental health services. By focusing on the root causes of violent deaths in Nairobi,  
policymakers can implement targeted interventions to reduce these fatalities and improve public safety.  
Addressing these challenges is crucial for the well-being of urban populations, not only in Nairobi but in cities  
worldwide.  
MATERIALS AND METHODS  
This prospective descriptive autopsy study was conducted at Nairobi City Mortuary, the largest national referral  
center for forensic autopsies in Kenya. Ethical approval was granted by the University of Nairobi-Kenyatta  
National Hospital Ethics Review Committee (REF: KNH/UON-ERC/A/196), and informed consent was  
obtained from the next of kin of the deceased. The study aimed to assess the prevalence and patterns of violent  
deaths in Nairobi over a one-year period, from June 1, 2009, to May 31, 2010.  
Crude prevalence rates were calculated using a population estimate of 3.7 million for Nairobi, based on the 2009  
census data. Data were meticulously recorded on standardized data sheets to ensure accuracy and consistency,  
and were subsequently entered into the Statistical Package for Social Sciences (SPSS) for analysis. Descriptive  
statistics were used to generate an overview of the patterns and prevalence of violent deaths, providing insights  
into the distribution of homicides, suicides, and accidental fatalities.  
To explore relationships between variables, chi-square tests were used to assess associations between categorical  
factors such as gender, age, and cause of death. Additionally, analysis of variance (ANOVA) was employed to  
compare mean differences in continuous variables (e.g., age or time of death) across the different causes of  
violent death. These analyses helped identify significant trends and factors influencing violent mortality in  
Nairobi during the study period.  
RESULTS  
Distribution of Causes of Death in Nairobi  
During the study period, a total of 5,065 deaths were registered in Nairobi, with 2,566 bodies being presented  
for autopsy. Among these, the majority of deaths (88.8%) were attributed to violent causes. Specifically, 2,278  
Page 263  
deaths (88.8%) were classified as violent, including homicides, suicides, and fatal accidents. In contrast, only  
102 deaths (3.97%) were attributed to natural causes, suggesting that the majority of fatalities were due to  
external causes rather than diseases or aging. Furthermore, 62 deaths (2.4%) were categorized as  
"unascertained," meaning the cause of death could not be determined due to decomposition or insufficient  
information. Additionally, 124 deaths (4.83%) were excluded from the analysis, likely due to incomplete data  
or administrative reasons.  
A Chi-square test was performed to determine whether there were significant differences in the distribution of  
violent deaths versus natural causes and unascertained deaths. The chi-square test showed a significant difference  
(χ² = 1857.89, p < 0.01), indicating that violent deaths were significantly more prevalent than deaths from natural  
causes or those with unascertained causes. This highlights the need for targeted public health interventions to  
address issues such as crime prevention, mental health support, and road safety. (Table 1)  
Table 1: Distribution of the deaths by cause  
Cause of death  
Violence  
Number  
2278  
102  
Percentage  
88.8  
Natural causes  
Unascertained  
Excluded  
3.97  
62  
2.4  
124  
4.83  
Total  
2566  
100  
Distribution of Violent Deaths  
Out of the total number of deaths in Nairobi, 2,278 deaths (45%) were attributed to violent causes. These were  
further categorized into the following: Accidents: The leading cause of violent deaths, with 1,064 cases (46.7%  
of violent deaths), including road traffic accidents, drownings, burns, and other unintended injuries. Homicides:  
A close second, accounting for 990 cases (43.5% of violent deaths), highlighting significant levels of  
interpersonal violence and criminal activity in the city. Suicides: A smaller proportion, with 224 deaths (9.8%  
of violent deaths), indicating a notable mental health concern. Accidents and homicides together accounted for  
over 90% of all violent deaths in Nairobi. Addressing these issues requires multifaceted strategies such as  
improving road safety, enhancing law enforcement, and increasing mental health support.  
A One-Way ANOVA was conducted to examine differences in the number of violent deaths across the different  
causes. The results indicated a significant difference in the number of deaths across the categories (F = 57.32, p  
< 0.01), suggesting that the distribution of violent deaths significantly varies by cause, with accidents being the  
most prevalent, followed by homicides and suicides. (Table 2)  
Table 2: Distribution of violent deaths by category  
Cause of death  
Accident  
Homicide  
Suicide  
Number  
1064  
990  
Percentage  
46.7  
43.5  
224  
9.8  
Total  
2278  
100  
Page 264  
Prevalence of Violent Deaths  
Nairobi's population during the study period was estimated at 3.7 million. The overall prevalence rate of violent  
deaths was calculated at 61.6 per 100,000 population, based on the 2,278 recorded violent fatalities. This  
prevalence rate indicates that violent deaths constitute a significant portion of overall mortality in the city.  
Accidents: The most common cause of violent death, with 1,064 cases, yielding a prevalence rate of 28.8 per  
100,000 populations. Homicides: Accounted for 990 deaths, with a prevalence rate of 26.8 per 100,000  
population, reflecting a significant level of interpersonal violence. Suicides: While less common, 224 suicides  
were recorded, with a prevalence rate of 5.4 per 100,000 population, indicating a need for increased mental  
health and suicide prevention initiatives.  
A Chi-square test was conducted to examine whether there were any significant differences in the prevalence of  
violent deaths based on different causes (accidents, homicides, suicides). The results of the chi-square test  
revealed a statistically significant association (χ² = 32.58, p < 0.01), indicating that the distribution of violent  
deaths is not random and varies significantly between accidents, homicides, and suicides. (Table 3)  
Table 3: Prevalence of violent deaths  
Category of Violent Death  
Overall  
Number  
2278  
1064  
990  
Prevalence rate(100,000)  
61.6  
28.8  
26.8  
5.4  
Accident  
Homicide  
Suicide  
224  
Age and Gender Distribution of Violent Deaths  
Violent deaths affected both male and female individuals aged 10-60 years. The overall age and gender  
distribution of victims varied across different causes of death (accidents, homicides, and suicides). Key findings  
include: The mean age of victims was 32.8 years (range: 10–59 years).Males accounted for 2,095 deaths (92%)  
out of the total 2,278 violent deaths, while females accounted for only 183 deaths (8%). This gives an overall  
male-to-female ratio of 11.4:1. The 30-39 age group was most affected, with 914 deaths (40.1%), followed by  
the 20-29 age group with 844 deaths (37.1%). These two age groups combined account for more than 77% of  
all violent deaths. The male-to-female ratio was most striking in the 20-29 age group (17.7:1), followed by the  
30-39 age group (10.4:1). (Table 4)  
Table 4: Age and gender distribution of victims of violent death in Nairobi, Kenya  
Age Range Frequency/ Gender  
Male  
34  
Female  
10  
TOTAL (%)  
44 (1.9)  
M:F Ratio  
3.4:1  
10-19  
20-29  
30-39  
40-49  
799  
834  
376  
45  
844 (37.1)  
914 (40.1)  
413 (18.1)  
17.7:1  
10.4:1  
10.2:1  
80  
37  
Page 265  
50-59  
53  
11  
64 (2.8)  
4.8:1  
TOTAL  
2095  
183  
2278 (100)  
11.4:1  
Statistical Analysis  
Chi-Square Test for Distribution of Death Causes:  
Chi-Square Value (χ²): 1857.89,p-value: < 0.01  
The significant result indicates that violent deaths are disproportionately more common than deaths from natural  
causes or those with unascertained causes.  
ANOVA for Distribution of Violent Deaths by Cause:  
F-value: 57.32,p-value: < 0.01  
This result indicates significant differences in the number of violent deaths across the different causes (accidents,  
homicides, suicides).  
Chi-Square Test for Prevalence of Violent Deaths by Cause:  
Chi-Square Value (χ²): 32.58,p-value: < 0.01  
The significant result suggests that the prevalence of violent deaths significantly varies between the causes of  
accidents, homicides, and suicides. -  
These statistical results reinforce that violent deaths are a significant public health issue in Nairobi, with  
accidents and homicides being the leading contributors. This calls for targeted public health strategies, such as  
improving road safety, reducing criminal activity, and enhancing mental health services.  
DISCUSSION  
Prevalence and Demographic Distribution of Violent Deaths in Nairobi  
The results of this study highlight a significant prevalence of violent deaths in Nairobi, with a crude prevalence  
rate of 61.6 per 100,000 population. This rate underscores the critical role violent deaths play in the city's overall  
mortality burden. The breakdown of causes reveals that accidents and homicides account for over 90% of all  
violent deaths, emphasizing the need for focused interventions to mitigate these fatalities.  
Age and Gender Distribution of Violent Deaths  
A demographic analysis reveals distinct patterns in the distribution of violent deaths based on age, gender, and  
geographic location. Age Distribution: The age group most affected by violent deaths was 30-39 years,  
accounting for 40.1% of cases, followed closely by 20-29 years, contributing 37.1% of violent fatalities. These  
young to middle-aged adults represent over 77% of all violent deaths, which highlights that young adults are the  
most vulnerable. The 10-19 years and 50-59 years age groups were far less affected, contributing only 1.9% and  
2.8%, respectively. Gender Distribution: Males were overwhelmingly affected by violent deaths, comprising  
92% of victims (2,095 out of 2,278 deaths). The male-to-female ratio of 11.4:1 was most pronounced in the 20-  
29 years age group, with a 17.7:1 ratio, and slightly less in the 30-39 years (10.4:1) and 40-49 years (10.2:1) age  
groups. Females accounted for only 8% of the total violent deaths, emphasizing the gender disparity in violent  
mortality. Geographic Distribution: The study did not explicitly examine geographic variation in violent  
deaths; however, data from other reports suggest that urban slum areas, such as Kibera and Mathare, are  
disproportionately affected by violence, particularly homicides and gang-related violence. Areas with poor  
infrastructure and high levels of poverty also tend to have higher rates of road traffic accidents (RTAs).  
Page 266  
LIMITATIONS  
While this study provides valuable insights into violent deaths in Nairobi, several methodological limitations  
should be considered:  
a). Selection Bias: The reliance on data from a single mortuary introduces potential bias, particularly if the  
mortuary primarily serves certain geographic regions or socio-economic groups. This could skew the  
representation of violent deaths, especially if areas outside the mortuary’s catchment area, such as remote or  
informal settlements, are underrepresented.  
b). Underreporting: Not all violent deaths may have been presented for autopsy, and some deaths, especially  
those in underserved or remote areas, might not have been recorded due to logistical challenges or the absence  
of forensic investigations. Additionally, unascertained deaths may be more common than reported.  
c). Data Exclusion: A total of 124 cases were excluded from the analysis, which could introduce gaps,  
particularly if these excluded cases were disproportionately affected by certain factors, such as age, gender, or  
cause of death. The exclusion of these cases might limit the comprehensiveness of the findings.  
d). Absence of Temporal and Seasonal Analysis: The study did not examine temporal patterns, such as  
seasonal variations in violent deaths, which could offer important insights into trends and dynamics of violence  
in Nairobi. A deeper exploration of when violent deaths occur could help identify patterns linked to specific  
times of the year or particular social conditions.  
e) . Lack of Qualitative Data: The study relied on quantitative data and did not incorporate qualitative  
approaches such as case studies or interviews. Including qualitative data in future studies could provide valuable  
context, helping to uncover the root causes of violence and offering insights for more culturally appropriate  
interventions.  
f) Selection Bias and Generalizability:This study is based on data from a single mortuary, which may introduce  
selection bias in two key ways:  
i)Non-representative Sample: The mortuary may not capture all violent deaths in Nairobi, particularly those in  
peripheral or rural areas. Deaths in these regions may be less likely to be brought in for autopsy, leading to a  
potential underestimation or overestimation of certain causes of death.  
ii) Missing Cases: The exclusion of 124 cases may reflect issues such as incomplete data or administrative  
errors. These missing cases could distort the findings, particularly in terms of cause of death and demographic  
breakdown.  
These limitations should be considered when generalizing the findings to the broader population of Nairobi.  
Future research could benefit from incorporating data from multiple mortuaries across different regions of the  
city, including underserved and peri-urban areas.  
g) Study Population Denominator and Catchment Area  
The study’s population denominator is based on 2,566 autopsy cases during the study period, but this may not  
fully represent the entire population of Nairobi, which was estimated at 3.7 million during the same period. The  
denominator used for calculating prevalence may be biased, as it does not account for discrepancies in how  
deaths are recorded or how autopsies are performed. The catchment area for the mortuary may also be limited  
to specific neighbourhoods or socio-economic groups, which could further affect the generalizability of the  
findings. To improve accuracy in future studies, the denominator should better reflect the entire population of  
Nairobi, while also accounting for the socio-economic and geographic disparities that influence the likelihood  
of a death being brought in for autopsy.  
Page 267  
CONCLUSION  
This study highlights the critical public health challenge posed by violent deaths in Nairobi, particularly those  
caused by accidents and homicides, which together contribute significantly to the city's overall mortality rate.  
With a prevalence rate of 61.6 per 100,000 population, the findings underscore the urgent need for  
comprehensive interventions. The statistical analyses reveal that addressing key factors such as road safety, law  
enforcement, and mental health services is essential for reducing the incidence of violent deaths.  
Furthermore, tackling the root socio-economic drivers of violence, including poverty and inequality, is vital for  
addressing both homicide and suicide rates. A multi-faceted approach is necessary to create long-term change  
and improve public safety in Nairobi. Immediate priorities should include enhancing road safety through  
improved infrastructure and stricter traffic enforcement, as well as strengthening crime prevention strategies to  
reduce homicides. Expanding access to mental health services is also crucial for reducing suicides, while  
addressing socio-economic inequalities, particularly youth unemployment, can help prevent violence by  
providing alternatives to crime.  
In the long term, urban planning and infrastructure improvements are key to building safer and more resilient  
communities. By integrating these strategies, Nairobi can create a safer, healthier environment for its residents,  
ultimately reducing violent deaths and improving overall public health outcomes.  
RECOMMENDATIONS:  
To reduce the burden of violent deaths in Nairobi, a multi-faceted approach is essential. Based on the findings  
of this study, several key recommendations have been identified, each addressing a different aspect of the issue  
and prioritized according to urgency and potential impact.  
a) Road Safety Improvements (Priority: High)  
Enhancing road safety should be a top priority, given that accidents account for nearly half of all violent deaths  
in the city. Immediate actions should focus on improving road infrastructure, such as upgrading poorly  
maintained roads and enhancing traffic flow in high-risk areas. Additionally, stricter enforcement of traffic laws  
is essential to address issues such as speeding, reckless driving, and non-compliance with road safety regulations.  
Public education campaigns promoting safe driving practices can also play a critical role in reducing road traffic  
accidents. By improving both infrastructure and behavior on the roads, it will be possible to significantly reduce  
the number of traffic-related fatalities in Nairobi.  
b) Crime Prevention and Law Enforcement (Priority: High)  
Addressing homicides, which account for a substantial proportion of violent deaths, requires a focused effort on  
strengthening law enforcement and improving public safety. This includes tackling gang violence, which is a  
major contributor to the high homicide rates, particularly in informal settlements. Community policing should  
be prioritized as a strategy to foster better relationships between law enforcement and local communities, making  
it easier to prevent and respond to criminal activity. Furthermore, investment in crime prevention programs that  
engage vulnerable youth and provide alternatives to gang involvement is crucial. Enhancing law enforcement  
capacity and providing adequate resources for crime prevention initiatives will help curb the high homicide rates  
in Nairobi.  
c) Mental Health Services Expansion (Priority: Medium)  
Although suicides constitute a smaller proportion of violent deaths, the prevalence of mental health issues  
remains a critical public health concern. Expanding mental health services and integrating them into primary  
healthcare settings is essential to improving access to care for individuals at risk of suicide. Efforts should also  
focus on reducing stigma surrounding mental health care, which often prevents individuals from seeking help.  
By offering more accessible and inclusive mental health support, the rate of suicides can be reduced. Public  
Page 268  
awareness campaigns aimed at educating the public on mental health and available services will help mitigate  
the cultural stigma associated with seeking help.  
d) Social and Economic Programs (Priority: Medium)  
Addressing the root causes of violent deaths, particularly homicides, requires a more comprehensive approach  
focused on poverty alleviation and youth empowerment. Nairobi's high levels of economic disparity contribute  
to social instability and violence, especially in marginalized communities. Initiatives that promote job creation,  
education, and skills development can provide vulnerable populations, particularly youth, with alternatives to  
violence and criminal activities. By reducing socio-economic inequalities, the risk factors associated with  
violence can be mitigated. Additionally, community-driven social programs that focus on economic  
development and empowerment will foster long-term peace and stability.  
e). Urban Planning and Infrastructure Improvements (Priority: Low)  
Finally, urban planning and infrastructure improvements are essential for addressing the broader social  
determinants of violent deaths. While this recommendation is of a lower priority compared to the immediate  
need for improved road safety and law enforcement, it remains a critical long-term strategy. Improving  
infrastructure in marginalized neighbourhoods, such as better street lighting, public transportation, and the  
creation of community centres, can help reduce the conditions that allow violence to thrive. Adequate  
infrastructure promotes greater social cohesion, safety, and access to essential services, all of which can help  
lower rates of violence.  
REFERENCES:  
1. African Union (AU). (2017). African Union Road Safety Charter. African Union Commission.  
2. Amegashie, D. K. (2016). The Burden of Road Traffic Accidents in Sub-Saharan Africa: Causes and  
Impact. Journal of African Public Health, 5(3), 89-97.  
3. Dahlberg, L. L., Krug, E. G., Mercy, J. A., Zwi, A. B., & Lozano, R. (2016). Violence: A global public  
health problem. In E. G. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, & R. Lozano (Eds.), World Report  
on Violence and Health (pp. 1-56). World Health Organization.  
4. Mugisha, J., Ndugwa, R., & Rwambali, J. (2020). Mental Health and Suicide Rates in Nairobi: A Need  
for Immediate Action. Journal of Mental Health and Wellbeing, 12(2), 45-52.  
5. Matzopoulos, R., Prinsloo, M., & Ramalapa, M. (2015). The Epidemiology of Violence in South African  
Cities: A Comparative Study of Johannesburg and Cape Town. South African Medical Journal, 105(3),  
162-167.  
6. Nairobi City County. (2018). Nairobi County Crime and Safety Report. Nairobi City Government.  
7. Wambugu, J. (2015). Patterns and Causes of Homicides and Accidents in Nairobi: A Mortality Analysis.  
East African Medical Journal, 92(8), 350-358.  
8. World Health Organization (WHO). (2019). Suicide Prevention: A Global Imperative. WHO.  
9. World Health Organization (WHO). (2020). Violence and Injury Prevention: Global Status Report.  
World Health Organization.  
10. World Health Organization (WHO). (2021). Global Status Report on Road Safety 2021. World Health  
Organization.  
11. Elvik, R. (2009). The Handbook of Road Safety Measures. Elsevier.  
12. Hawton, K., & van Heeringen, K. (2009). Suicide. The Lancet, 373(9672), 1372-1381.  
13. Kiriinya, R., et al. (2014). Addressing mental health in Kenya: Challenges and opportunities. The Lancet  
Psychiatry, 1(1), 15-16.  
14. Lind, J. (2006). Urban Violence in Nairobi: A Social and Political Analysis. African Studies Quarterly,  
8(1), 45-63.  
15. Mackay, M., et al. (2010). Urban Transport Safety: The Case of Nairobi. International Journal of Urban  
Transport, 8(2), 114-130.  
16. Mok, P. L., et al. (2016). Mental health services in sub-Saharan Africa: Challenges and opportunities.  
African Journal of Psychiatry, 19(2), 1-9.  
Page 269  
17. Norris, T., et al. (2015). Violence and Crime in Urban Sub-Saharan Africa. The Journal of Urban Health,  
92(3), 452-463.  
18. Peden, M., et al. (2004). World Report on Road Traffic Injury Prevention. World Health Organization.  
19. Stewart, F. (2008). Horizontal Inequalities and Conflict: Understanding Group Violence in Multi-Ethnic  
Societies. Palgrave Macmillan.  
20. Wandera, S. O., et al. (2017). Urban Violence and Crime in Nairobi's Informal Settlements. Journal of  
African Studies, 21(3), 255-276.  
Page 270