600 
appear to break out capriciously both in point of size 
and position, yet when observed perseveringly for a 
series of years, a recurrence of phenomena, so far at 
least as the total area covered by spots is concerned, 
becomes evident. The period required to complete this 
cycle is variously estimated, a little over eleven years 
being that most generally accepted. On Thursday last, 
a remarkable paper by Prof. Piazzi Smyth, Director of 
the Royal Observatory, Edinburgh, was read at the Royal 
Society, in which the results given by thermometers, 
buried at different depths in the earth and observed for 
upwards of thirty years, were tabulated and discussed. 
The main deduction from these observations was that the 
temperature of the earth, divested of the effects of tran- 
sient atmospheric changes, seemed subject to a secular 
law. This secular variation in the earth’s temperature 
may of course be due to secular changes going on ‘deep 
in the structure of the globe itself ; but it may be ascribed 
also, and with a far higher degree of probability, to 
changes in the heat-supplying power of the sun. 
There is one extremely important fact connected 
with these changes, namely, that one of them is accom- 
plished in 114 years; that is, exactly in the sun spot 
periods as determined by Wolf, and identical, or nearly 
so, with the period obtained from the Kew observa- 
tions. 
But the spots are only one of the known evidences of 
changeful activity going on in the great central luminary. 
The form, disposition, and dimensions of the prominences, 
and the distribution of the chromosphere, are visibly un- 
dergoing constant alteration. May these phenomena not 
also have their period of recurrence? And may not they, 
equally with the spot outbreaks, stand in some relation to 
what formerly used to be considered purely terrestrial 
phenomena, namely, magnetism, electricity, humidity, tem- 
perature, and rainfall? To carry the hypothesis one step 
further ; if there is a physical relation between the solar 
changes and meteorological occurrences, and if the solar 
changes are subject to laws which cause them to recur in 
regular series, have we not in this secular arrangement a 
clue by means of which climatic variations may be studied 
with greatly increased effect? Is not, in short, the sys- 
tematic study of solar phenomena extremely important 
from a meteorological point of view? 
If this hypothesis, which is one daily gaining strength, 
be but probably sound, the careful observation of 
the physical phenomena of solar eclipses becomes an 
urgent necessity, as calculated not only to afford more 
just and more noble views of the constitution of the 
universe, but to confer on mankind the same power with 
respect to climatal vicissitudes, which we already possess 
with respect to astronomical phenomena, the power of 
confident prediction, which will never be ours until we 
have a firm grasp of the secular laws by which those 
vicissitudes are governed. 
If these views be generally accepted, as we know they 
will be by those acquainted with the subject, we need not 
fear that anything which the knowledge and devotion of 
our astronomical physicists, or the power of the Govern- 
ment, can supply, will be wanting to effect the due ob- 
servation of the Solar Eclipse of December in a manner 
creditable to our age and our country. 
A. 
NATORE 
[April 14, 1870 
ON THE BASIS OF CHEMICAL NOTATION 
(Ce are so much in the habit of employing 
what are called chemical symbols, that they are 
liable occasionally to forget the realities symbolised; 
while persons interested in the realities of chemistry, but 
not themselves chemists, are apt to call in question the 
propriety of employing any such symbols at all,—looking 
upon the entire system of chemical expression as an 
arbitrary one, having its chief warrant from authority, and 
not only throwing an unwarrantable gloss upon the facts, 
but frequently overshadowing them. ‘That the accepted 
system of chemical notation is, indeed, to some extent 
arbitrary, and that it does throw more or less gloss upon 
the facts, may be admitted at once as indisputable; but 
nevertheless its relation to the facts is so simple and 
direct, and its utility as a means of illustrating and 
classifying the facts is so remarkable, that its justification 
ought not to prove a seriously difficult labour. 
It being the especial business of chemists to consider 
every material object in relation to the kind of matter 
of which it-is composed, they have gradually become 
acquainted with about sixty different kinds of matter that 
are unalterable in their kind by way of subtraction. The 
entire matter of a piece of iron, for instance, may cease to 
exist as iron, and, by an accretion of other matter, appear 
in the form of rust. But, though alterable in this way by 
the addition of other matter than iron to it, it is experi- 
mentally unalterable by the subtraction of other matter 
than iron from it. Now the sixty or more different kinds of 
matter having this property of unalterability by subtraction, 
though never declared to be in their essence elementary, 
are always tacitly assumed to be so; and chemical 
changes are accordingly interpreted in a definite way 
which, on this particular assumption, would appear to be 
the only legitimately possible way, but which, irrespective 
of this particular assumption, can only. be regarded as one 
of several more or less probable ways. 
Making the assumption, however, with eyes shut or 
open, as the case may be, chemists are able to learn, by 
analysis, the respective weights of the different elementary 
substances constituting a given weight of any compound 
substance. The results of the analyses are, of course, ex- 
pressible in various modes ; the most obvious, and, so to 
speak, impersonal mode, being the centesimal one—the 
setting forth of so many parts by weights of the respective 
constituents in 100 parts by weight of the particular com- 
pound. But in the case of several different compounds 
having one or more common constituent, the relationship 
of composition subsisting between the different com- 
pounds is much better brought out by taking some 
common constituent as a constant, and the other con- 
stituents as variable in relation thereto, rather than by 
taking all the constituents alike as variable. 
Now, among the sixty or so elements, hydrogen is 
characterised by this peculiarity, that in nearly all the 
compounds of which it is a constituent, it exists in a 
smaller proportion by weight than any other constituent, 
while in absolutely all its compounds it exists in a smaller 
proportion by weight than some other constituent; so that 
in the compounds which it forms with but one other kind 
of elementary matter, its proportion by weight is always 
less than that of the other elementary matter with which 
