1. "Casualties" includes both killed and injured, and unless otherwise stated indicates rates per annum.

2. Data is based on 18 years of exposure to landmines (1977 - 1994) [Roberts & Williams 1995:66, 67, 68, 74, & 429].

3. Data is based on an estimate of not less than 150 - 200 casualties per week [USAid 1994:21; USDoS 1994:15; World Resources Institute 1994:268; Roberts & Williams 1995:109].

4. Data is based on 2.5 years of exposure to landmines (1992 - 1994) [Roberts & Williams 1995:199].

5. Data is based on 28 years of exposure to landmines (1967 - 1994) [Roberts & Williams 1995:121, 125, 154, & 155, & 422].

6. Data is based on 14 years of exposure to landmines (1979 - 1992) [Davies, Dunlop, & McGrath 1994:58 & 162]. However, these estimate are conservative - the reality is probably much worse: '[a]s in Rattanak Mondul, those dying in remote areas sometimes never even make it onto official Cambodian statistics, let alone those held by foreigners' [Davies, Dunlop, & McGrath 1994:21], and '... for every victim who makes it to hospital, perhaps another two perish at the site of the injury, most of these never make it on to official injury statistics' [Davies, Dunlop, & McGrath 1994:58]. Therefore the above estimate for Rattanak Mondul is probably conservative. A more complete estimate would be about double this figure, at 1 per 1 107 or 362 per 100 000. This is to some extent validated by Davies, Dunlop, and McGrath [1994:22] who quote ICRC data indicating over 400 mine injuries per 100 000 people per year in Rattanak Mondul in 1992.

7. Data represents casualty rates for 1993 [USDoS 1994:25; World Resources Institute 1994:268].

8. Data represents casualty rates for 1994 - 1995 and is likely to under-estimate the severity of the problem [Anderson et al. 1993:189; MAG 1995:16; Roberts & Williams 1995:258-259].

9a. Data represents casualty rates for 1980 [Westing et al. 1983:125; Allen 1982:494; Maddocks 1994:55].

9b. Based on Westing et al.'s statement that '[d]uring the last five years [that is, 1978 - 1983], 30 to 40 people have been killed each year in that country, and 50 to 80 injured [1983:125], this rate can be extrapolated to give an estimated casualty rate of 1 casualty per 45 084 persons, or 2.218 casualties per 100 000 persons in 1995 [Westing et al. 1983:125; World Resources Institute 1994:269].

10. Data is based on a total of 15 000 victims over 25 years (1970 - 1995) [Roberts & Williams 1995:236], which is supported by Sheehan and Crolls who anticipated '550 men, women, and children will be killed or maimed by mines each year for the foreseeable future' [1993:11] . [Sheehan & Croll 1993:1 & 11; Vines, Anderson, & Schurtman 1994:5 & 42; World Resources Institute 1994:268; Roberts & Williams 1995:233, 235, & 236].

10a. Data is based on 14 years of exposure to landmines (1980 - 1993) [Aschiro et al. 1995:722-723].

11a. Data represents casualty rates for 1981 [Molski & Pajak 1985:26-27; Smogorzewski 1982:577].

11b. Data is based on Molski & Pajak's statement that '[e]ven at this late date, about a dozen people continue to be killed annually and twice that number injured' [1985:27], this rate can be extrapolated to give an estimated casualty rate of 1 casualty per 1 076 112 persons, or 0.093 casualties per 100 000 persons in 1995 [Molski & Pajak 1985:27; World Resources Institute 1994:269].

12. Data is based on Department of Health, Quang Tri province statistics [Monan 1994:19; Monan 1997:19]. Provincial population figure are not not available. Between 1985 and 1994, the casualty rate per annum ranged from a minimum of 1 killed and 15 injured in 1994 to a maximum of 1 killed and 149 injured in 1986 [Monan 1994:19; Monan 1997:19]. INTERPRETATION: In terms of impacts on human health per annum, landmine pollution is extreme in central Bosnia and severe in Mozambique, Cambodia, Afghanistan, and Angola. However, even after half a century, mines can still have a significant impact on human health. The Libyan data shows that mapping and fencing does not solve the problem. Even complete clearance, as in Poland's case, does not entirely eradicate the risk to human health.