304 REPORT—1904. 
In the case of school children we must measure a sample of 1,000- 
2,000 for each age interval of 12 months. The total number of school 
(ese? = 18,467. Dividing this number 
into 12 age groups each of 1 year, each group will contain 1,539. This 
number is only just a sufficient sample. Hence, it follows that it will be 
necessary to measure the whole of the children in the primary schools. 
There are also a number of children in addition to this in the secondary 
schools. 
Approximately it will be necessary to measure 
800,000 adults. 
8,000,000 schooi children. 
children in each district is 
Each district should be measured once in ten years. In order to keep 
the staff constantly employed, the measuring of the whole population may 
be spread over the whole ten-year period. It will be necessary therefore 
to measure per annum 
80,000 adults 
and 800,000 school children 
Taking 250 working days per annum, it will be necessary to measure 
bie, A0vdiabrines 
320 adults per day, 
and 3,200 school children per day. 
If we have a staff of 20 surveyors it will be necessary that each 
surveyor measures 
16 adults and 160 school children per day. 
This, I think, allowing for unavoidable delays in getting at the per- 
sons to be measured, is a reasonable amount of work to expect from one 
surveyor per day, assuming that only four dimensions are measured. 
It has been suggested that instead of whole-time surveyors, teachers 
should be employed part of their time in measuring the children in their 
own schools. 
Permanent surveyors, being constantly employed at the same work, 
and frequently tested by superintendent surveyors, would measure very 
much more accurately and probably at less cost to the State than the 
teachers. 
The objections which apply to teachers would apply in a less degree 
to factory inspectors. If these men in the course of their ordinary duties 
come frequently to London, Edinburgh, and Dublin, the cost of in- 
structing them would thus be materially reduced. But owing to the fact 
that only part of their time could be devoted to measurement, more sets 
of instruments would be required, and their skill would be inferior to that 
of permanent surveyors. 
An approximate estimate of the cost of whole-time and part-time 
surveyors shows clearly, I think, that the employment of a specially 
trained permanent staff of surveyors to carry out the whole anthropo- 
metric survey, instead of employing teachers, factory inspectors, or other 
existing officials, will be much more economical and much more efficient. 
In the Appendix is given a record of some of the anthropometric work 
which has been carried out or initiated in the United Kingdom in the 
