Fesruary 18, 1904] 
NATURE a7 
(4) Lastly, the subsidence of land might give a suit- 
able foundation on which a reef could build up to the 
surface either as an atoll or subsequently to form one. 
It has no doubt constantly produced large areas where 
(1) and (2) could act, but, where it has directly brought 
about the formation of atolls, it has probably been a 
purely local phenomenon. In no case can the mere 
existence of any atoll or reef be considered as evidence 
of any subsidence. Certain elevated reefs appear 
| 
| 
| 
possibly to have owed their origin to this mode of form- | 
ation, but there is no definite evidence in support of 
it for any existing reef, though certain reefs in the 
East Indies and to the south-west of the Indian Ocean 
may have so originated. 
The first three modes of formation resemble one | 
another in that to produce the characteristic forms of 
reefs they depend on factors which can be clearly seen 
in progress at the present day. 
may be found the same influences at work as produced 
its structure and appearance. The topography of the 
numerous submerged tropical banks of the Indo- 
Pacific region, that reach within 50 fathoms of the 
surface, strongly supports (1) and (2), but the examin- 
ation of others at greater depths, such as the Saya 
de Malha and Nazareth, is exceedingly desirable. 
More light, too, is imperatively demanded on the con- 
ditions and life of atoll slopes and shoals from 50 to 
500 fathoms, where the characteristic deep-sea con- 
ditions would seem to prevail. The study of the fauna 
also down to these depths over any large region or 
ocean may confidently be expected to throw a flood of 
light on the distribution of marine forms, and thus 
enable us to predict with some additional degree of | 
certainty the former distribution of land and sea. 
J. StaNLey GarDINER. 
PHOTO-TELEPHONY. 
THE transmission of speech by light was first | 
realised nearly a quarter of a century ago by the 
invention of Prof. Graham Bell’s photophone, a full 
description of which will be found in Nature, vol. 
XXlil. p. 15. The transmitting instrument contained a 
small silvered disc or diaphragm of thin microscope | 
cover-glass, which was clamped around the circum- | 
The | 
ference like the diaphragm of a_ telephone. 
receiver was a large parabolic mirror, at the focus of 
which was fixed a selenium cell in circuit with a tele- 
phone and battery. A beam of light from the sun or an 
electric lamp was reflected by the silvered diaphragm 
to the parabolic mirror, which concentrated the rays 
upon the selenium cell. The speaker’s voice was 
directed upon the back of the diaphragm, causing it 
to vibrate in correspondence with the sound waves; 
the rapid changes in the curvature of its surface which 
accompanied the to and fro movements of the central 
parts of the diaphragm varied the concentration of the 
light upon the selenium, and since the conductivity of 
the selenium varied with the illumination, sounds were 
produced in the telephone similar to those by which the 
transmitting diaphragm was agitated. 
Though the performance of the photophone was for 
short distances surprisingly perfect, it failed for several 
reasons to give satisfactory results when the trans- 
mitter and the receiver were separated by more than 
two or three hundred yards, and for a long time all 
attempts to render the invention practically useful were 
abandoned. The resuscitation of the photophone in a 
modified form has resulted from the recent discovery 
of the ‘‘ speaking arc.”’ 
A few years ago Dr. H. T. Simon, of Géttingen, 
having noticed the curious sounds given out by an are 
lamp when one of its leads happened to be near a wire 
NO. 1790, VOL. 69] 
Indeed, on each atoll | 
| 
supplying intermittent currents to an induction coil, 
was led to try the effect of superposing a microphone 
current upon the current feeding the lamp. With this 
object he interposed the secondary coil of a suitable 
transformer into one of the lamp leads, while the 
primary coil was connected in circuit with a battery 
of two or three cells and an ordinary microphone, the 
latter being placed in a distant room. Words spoken 
into the microphone were distinctly repeated by the arc, 
which was found to constitute an excellent telephone 
receiver. The simple arrangement originally em- 
ployed by Dr. Simon has since been modified in some 
details, notably by Mr. W. Duddell, who in 1g00 
caused an arc lamp to address a large audience in the 
hall of the Institution of Civil Engineers.‘ That such 
an effect should be produced by so small a cause as 
the feeble induced microphonic currents is surprising ; 
Mr. Duddell has, however, shown experimentally that 
a periodic variation of the order of 1 part in 10,000 of 
the mean current supplying the are will alter the vapour 
column sufficiently to produce sound-waves. 
No change of luminosity while the arc is talking can 
be recognised by the eye, but if the light is caused to 
pass through a transverse slit upon a moving kine- 
matograph film the developed negative shows a 
succession of narrow bands of varying brightness, 
indicating that considerable changes actually do occur 
in the intensity of the light. 
Prof. Graham Bell suggested in 1899 that the speak- 
ing arc might be used as transmitter in phonographic 
work. His suggestion has been followed up with 
conspicuous success by Dr. Simon himself and by Mr. 
Ernst Ruhmer, of Berlin, the latter having been able 
to transmit intelligible speech across distances up to 
15 kilometres. As a transmitter he uses a search- 
light projector having a parabolic mirror of silvered 
glass at the focus of which is the speaking arc; the 
carbons are placed horizontally along the axis of the 
mirror with the positive carbon outwards. A small 
telescope serving as a finder may conveniently be 
attached to the projector. The light is received, as in 
the older apparatus, by another parabolic reflector 
having a cylindrical selenium cell fixed axially at its 
focus. It is chiefly to the peculiar quality of his 
selenium cells that Mr. Ruhmer attributes the excel- 
lence of his results. The cells are of a pattern pro- 
posed by the present writer in 1880 (NaTuRE, vol. xxiii. 
p. 59). Two thin wires serving as electrodes are 
wound spirally, very close together but not touching, 
around a cylinder of unglazed porcelain, upon which 
a fine double screw-thread (to receive the wires) has 
been formed before baking. The surface of the 
cylinder is covered with a thin coating of selenium, 
which is afterwards crystallised; thus the two wire 
electrodes are joined throughout their length by the 
sensitive substance. Great importance is attached to 
the mode in which the crystallisation is effected. 
Vitreous selenium may be crystallised either, as is 
usually done, by gradually heating it up to a tempera- 
ture somewhat above 100° C. or by melting it at.a 
high temperature—about 250°—and letting it gradually 
cool down. In the latter case the crystalline surface 
appears to be of a coarsely granular structure, and it 
has long been known to the writer that selenium thus 
prepared is much less affected, as regards conduc- 
| tivity, by changes from darkness to light than the fine- 
grained variety obtained by the other process. This 
is confirmed by Mr. Ruhmer, but he has added the 
further interesting observation that the coarse-grained 
kkind—‘ soft ’’ selenium he calls it—is immensely more 
sensitive than the other, or ‘‘ hard ’’ selenium, to small 
changes of illumination, and, moreover, that it re- 
1 A detailed description of his apparatus is given in Journ. Inst. 
Electrical Engineers, vol. xxx. p. 240. 
