152 
— a SS ee See 
water of crystallisation, which is present in roo parts by weight 
of the solution. C shows the weight of the dissolved substance 
in the anhydrous condition. The numbers in this column may 
be calculated from those in the second column by multiplying by 
the combining weight of the anhydrous and dividing by that of 
the hydrated substance. Column D gives the weight of the body 
in the dry state, which is dissolved in 109 parts of water, and is 
calculated by multiplying the numbers in column C by 1oo and 
dividing by 109—C :— 
Cx 100 
2 = 100-C 
Column E contains the number of atoms of the anhydrous salt 
in 109 parts by weight of water. The expression a/om is here 
synonymous with equivalent, The atom of hydrogen is taken 
at roa 7— 
D x 100 
a A (anhydrous) 
F gives the volume of the solution ; 100 parts by weight of the 
water of the solution being taken as 100 volumes :— 
D x 109 
spec, grav. 
P= 
G indicates the specific gravities of the solutions. H contains 
the volumeter degrees, according to the scale of Guy Lussac, 
which correspond to the specific gravities :— 
100 
H = Ge 
In column I are found the names of the observers, the tempe- 
rature, and the references to the sources from which the numbers 
were obtained. 
In this first table we find the various numbers corresponding 
to solutions of different states of concentration, In some cases 
the numbers are given for solutions at intervals of 1 per cent. of 
the salt, in others of 5 per cent., and in others of 10. The table 
commences with caustic alkalies, including ammonia, potash, 
and soda. Then follow the potassic and sodic carbonates, the 
chlorides of ammonium, potassium, sodium, lithium, aluminium, 
magnesium, calcium, strontium, barium, cadmium, and zinc, and 
stannous and stannic chlorides. The next section contains the 
bromides of potassium, sodium, lithium, magnesium, calcium, 
strontium, barium, cadmium, and zine ; whilst under the iodine 
compounds we find potassic, sodic, lithic, magnesic, calcic, 
strontic, baric, cadmic, and zincic iodides. | Next comes sodic 
hyposulphate, and the sulphates of ammonium, potassium, 
sodium, manganese, and iron, the double sulphate of iron and 
ammonium, magnesic sulphate, potassia~-magnesic sulphate, and 
the sulphates of zinc and copper. This series is followed by 
sections containing potassic chromate and bichromate, hydric 
disodic, and trisodic phosphates; hydric disodic, and trisodic 
arseniates ; nitrates of potassium, sodium, magnesium, strontium, 
barium, and lead ; chlorates of potassium and sodium ; bromates 
of potassium and sodium, iodates of potassium and sodium ; 
potassic ferrocyanide and ferricyanide ; plumbic acetate ; potassic 
and sodic tartrate; and Rochelle salt. The remainder of the 
table is devoted to the acids, and includes the following :— 
Hydrochloric, sulphuric, sulphurous, phosphoric, arsenic, nitric, 
acetic, tartaric, and citric. 
After the table follows a chapter discussing the relations exist- 
ing between the specific gravities of equally concentrated solu- 
tions ; and three others: On the change of volume produced by 
solution of salts ; on the change of volume produced on the dilu- 
tion of aqueous solutions ; and on the change of volume produced 
by mixing different solutions. ; 
The pamphlet concludes with a table extending over 19 pages, 
and containing the specific gravities of solutions, in most cases 
from 1 per cent. to nearly the point of saturation, though in some 
few instances they are given at every 5 per cent. ‘This table 
gives, in addition to those of the substances above enumerated, 
the specific gravities of solutions of sugarand alcohol. Dr. Ger- 
lach deserves the thanks of chemists and chemical manufacturers 
for undertaking the tedious labour of collecting and arranging in 
tables the large series of numbers which are found in this 
pamphlet. 
NATURE 
| Fune 23, 1870 
SCIENTIFIC SERIALS 
THE American Fournal of Science for May, 1870, contains a 
good article ‘Ona simple method of Avoiding Observations — 
of Temperature and Pressure in Gas Analyses,” by Wolcott 
Gibbs, M.D., Professor in Harvard University. 
In absolute determinations of nitrogen and other gases, accu- 
rate observations of temperature and pressure are, in the ~ 
ordinary methods of analysis, necessary, and when made require 
subsequent calculations which, when the analyses are numerous, 
become rather tedious. By the following simple method these 
observations may be altogether dispensed with, and the true 
weight or the reduced volume of the observed gas, obtained 
at once by a single arithmetical operation, 
“A graduated tube, holding about 150 cubic centimetres, is 
filled with mercury, and inverted into a mercury trough. Two 
thirds or three fourths of the mercury are then displaced by air, 
care being taken to allow the walls of the tube to be slightly 
moist, so as to saturate the air. This tube may be called the 
companion tube; the volume of air which it contains must be 
carefully determined in the usual manner by five or six separate 
observations, taking into account, of course, all the circum- 
stances of temperature and pressure. The mean of the re- 
duced volumes is then to be found, and forms a constant 
quantity. The gas to be measured is transferred from the re= 
ceiver in which it is collected, into a (moist) eudiometer tu 
which is then suspended by the side of the companion tube, 
and in the same trough or cistern. Both tubes being supported 
by cords passing over pulleys, it is easy to bring the level o} 
the mercury in the two tubes to an exact coincidence. Thi 
pressure on the gas is then the same in each tube. The tem 
perature is also the same, as the tubes hang side by side in th 
room set apart for gas analyses, and are equally affected by an: 
thermometric change. It is then only necessary to read off 
the volumes of the gas in the two tubes to have all the data 
necessary for calculating the weight of the gas to be measured, 
: . . As the observed volume of the air in the companion 
tube is to the observed volume of the gas in the measuring 
tube, so is the reduced volume of the air in the first—previously 
determined as above—to the reduced volume of the gas to b 
measured. This method of course applies to the reduction of 
any gaseous mixture whatever to the normal pressure and 
temperature. . . . . In practice, a companion tube 
filled with mercury will last with a little care for a very long 
time. Even when filled with water I have found that excel- 
lent results may be obtained, and that the tube will last fer 
some weeks. Williamson and Russell, in their processes for 
gas analysis, have employed a companion tube for bringing “4 
how 
© 
gas to be measured to a constant pressure, but the applica- 
tion made above is, I believe, wholly new.” 3 
: 
SOCIETIES AND ACADEMIES 
LonDOoN " 4 
Royal Society, May 19.—‘‘A Ninth Memoir on Quantics.” 
By Prof. Cayley. j 
It was shown not long ago by Prof. Gordan that the number # 
of the irreducible covariants of a binary quantic of any order is 
finite (see his memoir ‘‘ Beweis das jede Covariante und Invariante 
einer binaren Form eine ganze Function mit numerischen Co- 
efficienten einer endlichen Anzahl solcher Formen ist,” Crelle, 
t. 69 (1869), memoir dated 8th June 1868), and in particular that 
for a binary quantic the number of irreducible covariants (in- 
cluding the quantic and the invariants) is = 23, and that for a 
binary sextic the number is = 26. From the theory given in 
my ‘* Second Memoir on Quantics,” PAi/. Zrans. 1856, I derived — 
the conclusion, which as it now appears was. erroneous, that for 
a binary quintic the number of irreducible covariants was infinite, 
The theory requires, in fact, a modification, by reason that cer- 
tain linear relations, which I had assumed to be independent, 
are really not independent, but, on the contrary, linearly 
connected together: the interconnection in question does not 
occur in regard to the quadric, cubic, or quartic; and for 
these cases respectively the theory is true as it stands; for 
the quintic the interconnection first presents itself in regard 
to the degree 8 in the coefficients, and order 14 in the 
variables ; viz., the theory gives correctly the number of — 
covariants of any degree not exceeding 7, and also those of the 
degree 8, and order less than 14; but for the order 14 the 
theory as it stands gives a non-existent irreducible covariant — 
a lt at SS A le 
