870 REPORT—1885. 
It is essential that one other warning be given. After all the care 
which can be exercised in eliminating sources of error, it is not at all 
certain that the quantities of goods, though expressed in the same 
denominations, are in truth alike in the substance of the unit. A ton of 
coals cannot vary in quantity, or greatly in quality, from year to year, 
but a yard of cotton piece goods may greatly alter in width, thickness, 
and tineness of texture, according to the make which happens to be sale- 
able at the time. Fabrics of mixed materials, such as cotton or jute 
conjoined with wool or silk, do greatly change in the proportions of the 
dearer substance, as well as in size or texture, according to the fashion of 
the day. There is good reason to believe that in all these respects. the 
yards of both cotton and woollen goods are now intrinsically less valuable 
than they were in past years, and in whatever degree this may be true of 
these and other goods, the comparison of goods by the price, weight, or 
measurement must be fallacious. As an instance of this, it happened that 
the value given for a consignment of shirtings to New Zealand was 
challenged as being preposterously low. The production of invoices proved 
the figures of both yards and money to be absolutely correct, and that the 
gcods were really described as shirtings. But further inquiry elicited the 
fact that these goods were destined not to make shirts for Europeans, or 
even for Maories, but to form shrouds for the carcases of sheep in the 
refrigerating chambers of the vessels bringing them home. For this 
purpose a much inferior article was serviceable, although in the trade 
accounts it would necessarily be grouped with those of far higher quality 
and price. 
But it is high time to leave these wearisome details and descriptions, and 
to point out a few of the purposes for which the tables may serve. They 
must be regarded as means, not ends; tools wherewith crude and unsightly 
statistics may be reduced into useful, and it may be comely shapes. 
Like interest tables and logarithmic numbers, though equally tedious to 
construct, and far more difficult to describe, they may save a vast amount of 
labour to those by whom they are used, and lead up to results of real utility. 
In the first place, the extreme variation of price within short periods 
of time, and the equal irregularity of their recurrence, forbid the assump- 
tion that gold is now or has very lately been altering in value. I do not 
say that it may not have become appreciated, but if so the evidence to 
sustain the theory must be sought elsewhere. When coals rose from 
884s. in 1863 to 20°49s. in 1873, and in 1883 sank to 9:20s. it cannot be 
affirmed that the change was in the metal rather than the mineral. Con- 
trast this with cotton yarn, which in the three years named stood at 
26°01d., 17°76d., and 12-°25d. Like the gold itself, to procure which we 
must crush down the quartz in which it lies imbedded or wash away the 
sand with which it is mixed before the grains are discovered, so the 
hard mass of statistics, or the encumbering multitude of figures, must be 
ground or boiled down if we would eliminate the truth from the sur- 
rounding sources of error so prone to lead the judgment astray. 
In the next place, by bringing out so clearly the great accessions to 
the bulk of our trade, and yet limiting these changes within narrower 
bounds than many economists are disposed to admit, we discover the 
severe fall which prices have sustained, and hence the enormous quantity 
of goods we have to give for an equal or lesser amount of money. Com- 
paring 1883 with 1873, for instance, we find the actual value of the goods 
we sold to have been 15,000,0007. less, or a fall of 6 per cent., whereas. 
