56.29% being of males, 43.69% being of females, and 0.02% of unknown sex. The crude death rate accordingly dropped from 5.5 to 4.9, which is one of the lowest rates in the world. While due tribute must be paid to the efficiency of the Colony's medical and health services, it must be acknowledged that this exceptionally low rate to a large extent results from the age structure of the population, which contains an abnormally high proportion of young people. The infant mortality rate dropped 6.5 from 32.9 to 26.4 per thousand of live births, thus continuing the uninterrupted downward trend in this rate since 1950.
158. Table XXXV is a new Table giving the numbers of deaths registered by districts in 1962, 1963 and 1964. It shows that of the 18,113 deaths in 1964, 5,918 (32.7%) were on Hong Kong Island, 10,707 (59.1%) in Kowloon and New Kowloon, and 1,488 (8.2%) in the New Territories.
159. Table XXXVI gives the numbers of registered deaths by age groups during the five years 1960 to 1964. This shows that taking males and females together, the number of deaths in each age group up to 60-64 (except the 15-19 group in which there was an increase of 28) declined, and small increases prevailed in the age groups from 65-69 upward.
160. In relation to the statistics in Table XXXVI it is advisable to repeat the warning given in previous Reports. The vast majority of the ages of the deceased are given to the District Registrars by relatives according to the Chinese reckoning, under which, for instance, if a child is born the day before Chinese New Year, it is regarded as being two years old on the following day. These Chinese ages are converted by the District Registrars to ages according to the European reckoning, but obviously errors can easily be made in the process. There is, moreover, no doubt but that a great many people are extremely hazy about their own and their relatives' ages. A further complicating factor is the tendency of relatives to exaggerate the age of the deceased, it being not uncommon in public announcements to add one month to the age for every year of the deceased's true age. The District Registrars, however, use their best endeavours to ascertain the correct age, and the figures given for the various groups are probably pretty fair approximations.
Accident and Suicide Deaths
161. Table XXXVII shows the numbers of deaths in 1964 by accident and suicide classified by causes, sex, and age group. This revealed that 1,346 people died by accident or suicide as follows, the 1963 figures being given in brackets for comparison:
46