ON OFFICIAL STATISTICS. 429 



An account of the persons arriving l)y sea is, it is believed, 

 kept with a fair degree of accui'acy in all tlie colonies ; but the 

 proper officers seem to be unable, in any of the colonies, to secure 

 an equally accurate return of the departures ; the reason being 

 that a large proi)ortion of those who leave for the neighbouring 

 colonies without taking their passages beforeliand (which is a very 

 common practice) escape their notice. The estimates of population 

 are affected by this more than by any other cause, (except perhaps 

 overland emigrants), and, if allowance be not made for it, they 

 have always been found to largely overstate the truth. It will 

 be remembered that, at the census of 1881, the estimates of popu- 

 lation previously made in Victoria, New South Wales and Queens- 

 land were too high by 112,000, and I have ascertained that at a 

 late census taken in Queensland during 1886, it was found that 

 the population had in live years been over-estimated by 11,500, 

 whilst at a census taken in New Zealand during the same year, it 

 was found that the population had in a like period been over- 

 estimated by 7,400. To counteract this tendency to over-estimate, 

 an allowance is now made in two of the colonies, but this amounts 

 to little more than a guess, and it would be more satisfactory if 

 some means could be arrived at whereby an account of the depar- 

 tures could be kept with the same accui'acy as that of the arrivals. 

 If the ports at which the passengers embark and disembark, 

 were stated correctly, the returns of the departures might be 

 adjusted with a sufficient decree of accuracy by means of i-eturns 

 of the arrivals in the other colonies, but since so much of the 

 intercolonial passenger traffic is carried on by the large mail 

 steamers which call at several Austi'alian ports, and carry passen- 

 gers besides to and from Europe, and the East, the passenger 

 returns are so mixed up that I have found them entirely valueless 

 for making cori'ections of this nature. There is much room for 

 improvement in the immigration and emigration statistics of all 

 the colonies, and I have more than once suggested to the Govern- 

 ment of my own colony that an effiart should be made to bring 

 about a conference between the different officers charged with 

 keeping the records of arrivals and departures with the view to 

 their amelioration. I have also urged the desirability of an 

 endeavour being made to procure returns — or at any rate reliable 

 estimates — of persons crossing the borders from colony to colony ; 

 but this I have been told is altogether hopeless, and the state of 

 matters in this regard it is to be regretted, appears likely to 

 remain unsatisfactory. 



The registration of deaths is fairly complete in all the colonies, 

 Vjut the registration of births is veiy far from being so in most of 

 them, as each successive census has proved. This not only 

 disarranges the estimates of population, but as the deaths of infants 

 under one year of age are compared with the births for the pur- 

 pose of determining the infantile mortality, it makes this appear 



