537 
of Edinhurfjli, Session 1883-84. 
phenomena are not thereby affected ; the matter is one purely of 
arithmetical convenience. Had the subdivisions been according to 
the powers of ten, these conversions and their attendant labour 
would have been saved. But it is not now and then only that 
these irksome conversions occur ; they pervade every calculation in 
geodesy, navigation, astronomy. The estimate is not too high, that 
they double the labour of computation. 
In astronomical works there is abundant evidence of the need for 
a change. While the reckoning of longitude in signs, used sixty 
years ago, is discarded in favour of the counting in degrees all 
round, thirds are quite disused, the second is divided into tenths 
and hundredths. The arguments for the planetary disturbances 
are given, not in degrees, but in thousandth parts of the entire re- 
volution. 
There is no work having greater authority in these matters than 
that most admirable one, the Nautical Almanac y and every page 
thereof proclaims the need for decimals. The right ascensions, decli- 
nations, latitudes, and longitudes are given to decimals of the second. 
Now, if the division of the second into 100 parts be better than 
into 60, why should we not adopt, as John Newton did in the 
Trixfonometria Britannicciy the centesimal division of the degree? 
There is, and there can be, ho argument in favour of division by 
60 down to seconds, that will not hold as well for thirds and for 
fourths ; and the same instinct for convenience which leads to the 
decimal division of the second would, if it had its own way, lead 
to that of the degree, of the quadrant, and of the day. 
But ease of calculation is not the only consideration. The sun’s 
daily right-ascensions, to hundredths of a second, are accompanied 
by a column of variations in one hour ; this, which is really needed 
for the sake of the inept computer, saves the division by 24. But 
this column occupies the place of the actual differences, needed by 
the strict computer for taking into account the variation of the 
variations. With decimal graduation one column would suffice for 
all, and the compiler of the almanac v/ould be spared the labour. 
The same may be said of the sun’s declinations and of the moon’s 
hourly places, which are accompanied by variations in 10 minutes. 
In the last-mentioned there is a remarkably strong instance of the 
awkwardness of sixties. Thus the variation in declination is to be 
