892 
full of suggestive thoughts for physicists, chem- 
ists and biologists, who are too apt to overlook 
the many metaphysical conceptions used by 
them in their most rigid scientific investigations. 
On the other hand, the scientist or the ‘plain 
man,’ accustomed to use trans-subjective things 
as his models of reality, is tempted to say 
that the ‘reality,’ with which Professor Ladd 
deals, is only a metaphysical abstraction, quite 
of a kind with the ‘ stream of consciousness’ con- 
ception of the Self, which is adopted as his 
model. If a ‘stream of consciousness’ had 
no channel in the bed-rock of real things to flow 
in, such a man might ask, How could any 
knowledge of the reality of the Self arise? 
From a common sense point of view, such a 
criticism would appear to be valid, since our 
idea of, as well as our term for, reality is ob- 
tained from the thing (Latin, res). The thing- 
ness of the thing is reality ; this does not, how- 
ever, invalidate the theory that the ‘ground of 
things’ may be, metaphysically, in the same 
class with the Self. The ‘plain man’ will, 
however, contend that it is by reason of its 
derivation from the thing, as its ground, that 
the conception of reality derives its meaning, 
and he will naturally infer that the putting of 
reality and the self into the same class will re- 
duce self to a ground of a particular trans-sub- 
jective thing, viz., of its physical organism. Only 
when we take the point of view of the author, 
by adopting self in contrast to thing as our 
model of reality, do we reach the conclusion 
that reality is the selfness of the thing and of 
all things. This volume is of chief value to the 
scientific student for the light it throws, from 
this view point, upon some of his most diffi- 
cult problems. Henry S. WILLIAMS. 
New Haven, Conn., November, 1899. 
DETERMINATION OF THE DENSITY OF WATER 
AT 4° ©, BY THE INTERNATIONAL 
BUREAU OF WEIGHTS AND 
MEASURES, 1899. 
THE interest attaching to the recent Report* 
on this subject is two-fold in that this constant 
*Détermination de la masse du décimeétre cube 
d’eau. Rapport préliminaire présenté au Comité In- 
ternational des Poids et Mesures dans la séance du 18 
avril, 1899, par M. le Dr. Ch.-Ed. Guillaume. 
SCIENCE. 
[N. S. Vou. X. No. 259. 
is the connecting link between the metric units 
of capacity and mass as well as in most scien- 
tific volumetric measurements, and in that the 
present result bears the hall-mark of the insti- 
tution that has given us our accurate standards 
of length, mass, and temperature. That the 
investigation was conducted by M. Guillaume, 
whose rare ability in quantitative research has 
become widely recognized through his memoirs 
as adjunct of the Bureau and through his ad- 
mirable ‘Thermométrie de Précision,’ ‘ Unités 
et talons,’ ete., is abundant guarantee that no 
refinement known to modern metrology has 
been omitted in this work. 
After a discussion of the method, results, and 
sources of error, the report concludes : 
‘Ror the present it is probable that in adopt- 
ing for the specific mass of water the value 
0.99 995 or 0.99 996 the error committed will 
not exceed 2 centigrammes per kilogramme. 
We hope to be able by an exhaustive discussion 
of the measurements to reduce a little more 
these limits of uncertainty.’’ (Translation.) 
The method was the familiar one of weigh- 
ing a solid of measured dimensions successively 
in air and in pure water from which the dis- 
solved air had been withdrawn. Four hollow 
cylinders were used, two of bronze and two of 
brass. Their diameters ranged from 14.4 to 
6.6 centimeters, and the height of each was 
about the same as the diameter. The corre- 
sponding weights of water displaced ranged 
from about 2 to 0.2 kilogrammes. The mean 
temperatures of the water when the weighings 
were made were about 8°, 8.°5, 9°, and 15°, 
these being selected, except the last, as giving 
about the maximum weight of displaced water. 
The linear dimensions of the cylinders were 
measured at a large number of systematic- 
ally distributed points by the usual comparator. 
Sliding contact bars bearing reference marks 
were brought into contact with the cylinder at 
opposite ends of a diameter or of a height, and 
the distance between the marks measured by the 
microscopes and standard scale. This distance, 
less that found when the stops were in direct 
contact, gave the desired dimension. The den- 
sity of the water was reduced to 4° by means 
of the tables of the expansion of water from 
the measurements of M. Chappuis (See Procés- 
