376 
NATURE 
[ Feb. 15, 1883 
or in beds of shale interstratified with the sandstone, cutting 
off communication with the open air in this direction. Fissures 
traversing a dry sandstone in such a situation constitute an air- 
chamber which may clearly be of great capacity. On cutting 
ene of asystem of connected fissures, the first eftect is frequently 
to liberate a quantity of pent-up air or choke-damp, as at 
Solberge ; subsequently the opening becomes the sole channel 
by which equilibrium is preserved between the enclosed air and 
the atmosphere. It would however seem as probable that the 
opening should occur below the water-level as above it. In 
such a case the first effect of an expansion of the pent-up gases 
would be to force out water, and raise the level of the water in 
the well. The agitation of the water noticed in the well at 
Chester is probably due to the openings being partly above and 
partly just below the surface of the water. That they not infre- 
quently are wholly below appears probable from observations on 
springs and wells, for it has been noticed that in certain chalk- 
springs there is an increase in the amount of water flowing when 
there is a rapid fall in the barometer, though no rain may have 
fallen, and that under the same circumstances water recommences 
to flow from land-drains and percolation gauges. The gaugings 
of deep wells in the chalk have confirmed these observations and 
show that there is a rise in the water-level under a decrease of 
atmospheric pressure. These movements have been attributed 
to the expansion of the dissolved gases.1_ It is probable that the 
gases when given off by the water, rise into and occupy cavities 
from which there is no escape upwards. 
It is noticeable however that five certainly, and two probably, 
of the blowing wells described above derive their properties from 
fissures in the New Red Sandstone ; no case is known in either 
chalk or limestone, though these are soluble rocks peculiarly 
liable to contain caverns or widened joints. It is not improbable 
that the fissures are too numerous in these rocks, so that where- 
ever large hollows occur, there are also communications upwards 
with the open air. In sandstone on the other hand large 
hollows are of extremely rare occurrence, and in view of this 
difficulty it has been suggested that the Magnesian Limestone 
which underlies it about Northallerton, at a depth of about 400 
feet, and is known to be extremely cavernous, may have given 
way in places, and led to the formation of hollows in the sandstone. 
This explanation however is not applicable to the wells at 
Tamworth or Chester, where the sandstone is not underlain 
by limestone. It seems more probable that the strength of the 
air-currents should be taken in connection with the copiousness of 
the water-supply as indicative of the great extent of small ramify- 
ing fissures in some of the triassic sandstones. That the united 
capacity of such fissures must be very great to account for the 
phenomena is undeniable. The volume of air contained in the 
cavities at the Solberge well was estimated at about 10,000,000 
cubic feet, or as much as would fill a chamber measuring 217 
feet each way, length, width, and height.2 In making this 
estimate no allowance was made for aqueous vapour, or for air 
held in solution in the water, both of which would come off in 
increased quantities with a decreasing pressure. The former 
was known by the state of the meter to have been present in 
large quantities. But making every allowance for these 
causes of error, it is impossible to escape the conclusion that the 
fissures, small as they are individually, must in the aggregate 
form a reservoir of immense capacity. 
In concluding these remarks we may refer to the practical 
application of the knowledge of these properties in fissures. It 
has been noticed that the drains of large works begin to smell on 
the approach of rain, and there can be little doubt that this is 
partly due to the setting up of an outward current corresponding 
toafallinthe barometer. In fact every network of covered drains, 
and every covered cess-pool, where special provision for ventila- 
tion is not made, must constitute a natural blowing well. _ It is 
not our intention to discuss here the engineering details of drainage. 
It is sufficient to point out that by a faulty system of ventilation, or 
by the derangement of a system originally good, sewer-gas might 
be forced into a house with every fall of the barometer. 
Lastly we would allude to the effect of the barometer on the 
escape of fire-damp from coal-seams. Coal is a rock subject to 
jointing ; seams are not only broken through and displaced by 
faults, but for some distance from the main fracture are traversed 
by joints and smaller shifts resulting from the general strain. A 
brief visit to a fiery portion of a mine is sufficient to show the 
part played by these small clefts. On every side is heard the 
* Baldwin Latham, Report of the British Association for 1881, p. 614. 
2 Proc. York Geol. and Polyt. Soe., of. cit. 
monotonous hissing or bubbling of the escaping gas, often 
accompanied by the deeper note of a ‘‘ blower,” or one of those 
larger channels often observed in connection with faults: The 
gas is continously given off as a result of a slow decomposition 
taking place in the coal, and the amount that comes off indicates 
a great extent of connected fissuring. For though cavities 
charged with gas under pressure and liable to exhaustion 
are found, yet large ‘‘ blowers ’” commonly continue active for 
years, and must therefore drain a large area of theseam. While 
the movement of the gas in the blower differs from that of the 
air in sandstone fissures, in being always in one direction, namely 
outwards, it is at the same time evident that the same cause 
which induces an outward current in the well would cause an 
increase in the outward current from the coal. The increase 
would be proportional to the capacity of the fissures ; a fall in 
the barometer from thirty to twenty-nine inches for example 
would cause th of the body of gas stored in the fissures to be 
added to the ordinary outflow. The liability to explosion with a 
falling glass has long been a subject of observation. When it is 
considered that a wide margin is usually allowed in the ventilation 
to ensure the sufficient dilution and removal of fire-damp, and that 
a number of other contingencies may bring about an explosion, 
it becomes evident tlat a powerful cause must be operating to 
make the influence of the barometric changes perceptible. 
A, STRAHAN 
SOCIETIES AND ACADEMIES 
LONDON 
Royal Society, January 25.—‘‘ Internal Reflections in the 
Eye,” by H. Frank Newall, B.A. 
The author in this paper describes the appearance and investi- 
gates the cause of a faint light seen under certain circumstances 
now to be related :—Stand opposite a uniformly dark wall in a 
darkened room. Direct the eye to any point in front, and 
keeping the eye fixed, and being ready to perceive any appear- 
ance out of the line of direct vision without moving the eyes 
towards it, hold up a candle at arm’s length, and move it to and 
fro over about two inches ona level with the point fixed, and a little 
to the right or left of it. The faint light may be seen moving 
with a motion opposite to that of the candle on the other side of 
the point of direct vision. 
Near inspection of the light shows it to be an inverted image 
of the candle, about equal in size, very faint. 
Reasons related in the paper lead the author to offer the 
following explanation : the physical cause of the faint light or 
“ghost,” is light which, proceeding outwards from the image ot 
the candle, formed on the retina by the lens, is reflected back on 
to the retina by the anterior surface of the lens. This second 
image is ‘‘referred”’ outwards, and seems as if produced by a 
faint source of light outside the eye. 
The effects of alterations of the state of accommodation on the 
appearance of the ghost are described; the question as to 
whether the retina is to be regarded as a screen or as a regular 
reflector is discussed; and the results of calculations based on 
numbers given by Helmholtz for his schematic eye are noted as 
forming a difficulty in the explanation. : 
If the candle be replaced by sunlight, further observations are 
to be made : (1) signs of the faulty centering of eye-surfaces, as 
shown by the fact that the sun and its ghost do not arrive at the 
centre of the field of vision together ; (2) signs of oblique re- 
flection at a concave mirror, as shown by the fact that the ghost 
is circular in only one state of accommodation, whilst in other 
states it is extended either in a horizontal direction for near 
focus, or in a vertical direction for distant focus. 
To about four out of fifteen persons the author has failed to 
show the ghost ; but no relation is as yet observed between the 
visibility of the ghost and the kind of sight of the observer, as 
defined by the ordinary terms, long- and short-sightedness. 
A second ‘ ghost,” probably due to reflections entirely within 
the lens, is referred to in the paper: but this, on account of its 
indistinctness, has not been investigated, except to establish the 
fact that its motion is the same in direction as that of the candle 
in the circumstances above related. 
February 1.—On the Electrical Resistance of Carbon Contacts, 
by Shelford Bidwell, M.A., LL.B. : 
The experiments described in the paper were undertaken with 
the object of investigating the changes of resistance occurring in 
carbon contacts under various conditions. 
