length of axis of twenty inches; the lower whistle is of the same diameter, but of a length of axis of fourteen inches. The note of the shorter bell is a fifth of that of the longer. This arrangement gives a melodious sound, unlike that of ordinary locomotive whistles, and on that account has extraordi. nary merit; its character being strongly distinct from that of steamboat whistles. In reference to the audibility of signals in different kinds of weather, it was found that a sound moving against the wind, and inaudible to the ear on the deck of a schooner, was heard by ascending to the masthead. In general, it was stated that when the fishermen in the morning, on the Banks of Newfoundland, hear the sound of the surf to the leeward, or from a point toward which the wind is blowing, they take this as an inevitable indication that in the course of from one to five hours the wind will change to the opposite direction from that in which it is Llowing at the time. General Duane states that the fogsignals at Cape Elizabeth, and at Portland Head, which are respectively nine and four miles southeast of Portland, can be heard in the latter city much better during a heavy northeast snow-storm than at any other time, although the sound comes to the city in nearly direct opposition to the course of the wind. The most perplexing difficulty, however, arises from the fact that the signal often seems to be surrounded by a belt in which the sound is entirely inaudible. Thus, in moving directly from the station, the sound is audible for the distance of a mile, is then lost for about the same distance, after which it is again distinctly heard for a long time. This action is common to all sound signals, and has been at times observed at all the stations; even at one where the signal is situated on the bare rock, twenty miles from the mainland, with no surrounding objects to affect the sound. Rep. Lighthouse Board, 1874, Appendix.

OY CELESTIAL PHOTOMETRY.

Professor Thury communicates to the Physical Society of Geneva a very full description of a new photometer adapted to astronomical purposes, and also some general considerations upon photometry. It is not at present necessary, as it was fifty years ago, to insist upon the importance of photometric observations in astronomy. We know that the problems

of the distribution of the stars in space, and the gradual mod-
ifications that celestial bodies undergo in their own nature,
are intimately connected with the intensity of the light we
receive from them or that they send out; but it would be
impossible to find a collection of photometric observations
sufficient to serve as a basis for safe deductions. Either the
accuracy of the observations is too small, or there are not
enough of them. The observations that we have are due,
for the most part, to experienced observers, and the differ-
ences between their methods of research fully explains the
diversity of their results. In order to make their observa-
tions comparable with each other, and to eliminate causes of
error peculiar to each method, it is necessary to institute
comparative observations by making use of each of the
methods employed hitherto in photometry. The results of
such an investigation, which has already been commenced in
Germany, will probably be, first, a general accordance of the
figures obtained by different methods, sufficient to give con-
fidence in their exactness. Second, the knowledge of the
means proper to bring about such an accordance; that is to
say, a knowledge of the universal corrections and of the in-
provements necessary to be introduced into the apparatus,
and the methods of employing it. Finally, we shall know
which of the photometric methods permits the greatest de-
gree of exactness, and which offer special advantages. In
general the photometers hitherto employed may be divided
into two categories: First, those where the object affected by
the light is the eye itself. Second, the physical and chem-
ical photometers, where some inert body is modified by the
light. Of these latter, that of Leslie and the photographic
photometers are instruments especially adapted to meas-
ure separately the intensity of different kinds of radiations.
Visual photometers are divided into two classes.
we diminish the brightness of the light until it disappears
from the sight, or, rather, until it becomes too feeble to en-
able us to distinguish certain definite details of objects, and
we then calculate the quantity of the diminution by know-
ing the methods employed to produce it. These are the
photometers of extinction; such are those of Arago, Xavier,
and Maistre. The second class of visual photometers is that
of comparison, where the two lights present themselves at

1

the same time to the eye, and we diminish the brighter one
until it becomes of equal intensity to the feebler. The quan-
tity of this diminution measured as a fraction of the primi-
tive intensity expresses the comparative brightness of the
two lights. In place of estimating directly the equality of
the brightness of the two images, we can oppose them one
to the other, giving birth to phenomena such as will render
their perfect equality more sensible--for example, by trans-
forming and transmitting the inequality of intensity into the
production of color. The photometers of comparison which
possess the greatest perfection can perhaps be called pho-
tometers by opposition; such are those of Wild, Bunsen,
Dove, and one of Arago's photometers. The essential part
common to all photometers by comparison and by extinc-
tion is that which is designed to diminish the intensity of
the light, under the condition that the quantity of this dimi-
nution shall be exactly measurable. To this end, recourse
has been bad to the eight following methods: First, the ab-
sorption of the light by an apparently transparent medium
of variable thickness. Second, reflection from a polished
surface at a variable angle. Third, reflections from one,
two, three, or more surfaces respectively, at an invariable
angle. Fourth, the reduction of the intensity by a deviation
by means of reflection of a portion of the light. Fifth, re-
duction of the intensity by two polarizing systems or planes
of polarization by inclining them to cach other at an invari-
able angle. Sixth, reduction by bifurcation of the ray in a
double refracting prism. Seventh, reduction by the increas-
ing separation of the rays of a conical pencil. Eighth, ex-
clusion of a portion of the beam of light which enters the
pupil of the eye. This exclusion may be accomplished by
means of a diaphragm placed either near the eye or in front
of the objective of the telescope, or within the terrestrial
telescope at the place occupied by the small diaphragm of
the quadruple ocular. Of all the combinations that we have
enumerated, Thury has chosen one of the most simple, viz., a
photometer having a variable diaphragm and reflecting mir-
l'ois. His instrument is adapted to a four-and-a-half-inch
refractor of excellent defining power, and has since 1868
been applied especially to the nebulæ and the components
of double stars. The photometric system adopted is, there-

1

fore, that of the visual photometer by extinction. The enfeebling of the light is obtained by reflection from one or more mirrors situated between the objective and the ocular, and by a diaphragm having a variable opening placed in front of the objective. This diaphragm is composed of sixteen thin rectangular plates, sliding simultaneously and uniformly each in the direction of its length, and the direction of the radius that passes through the centre of the objective. The polygon of a variable diameter and symmetrical form is the real aperture of the objective. When the aperture of the telescope is diminished too much, the dimensions of the false disks of the stars increase, and the diffraction rings that surround the false disk become so modified that the conditions of visibility are no longer the same for two stars viewed with very different apparatus. It is necessary, therefore, to correct this source of error by diminishing the brightness of the brightest stars, not by contracting the aperture, but by introducing the use of mirrors. A comparative table is given, showing the relative effects of mirrors and diaphragms. Two classes of scales have been adopted in expressing the orders of brightness of the stars. The photomctric scale of Sir John Herschel was based upon the simple fact that the intensity of light diminishes as the square of the distance. The brightness of the stars belonging to the first, second, third, etc., magnitudes on his scale was therefore respectively one quarter, one ninth, one sixteenth, etc. The system more generally followed is such that the brightness of a star of any order is always a certain constant fraction of the brightness of a star of the next succeeding order, so that the arithmetic series of magnitudes corresponds to a geometrical series of intensities. The constant ratio in this system would naturally be so chosen as to change as little as possible the magnitudes that have been somewhat arbitrarily assigned to the stars by many generations of astronomers. The actual photometric series of Sir John Herschel accords remarkably well with the ordinary scales of magnitudes, if we simply multiply his magnitudes by 1.41, and take for the unit of intensity a star equal to that of Alpha Centauri. But the photometric scale of this astronomer of: fers grave inconveniences, which have hitherto prevented its being adopted. The smallest star visible in the twenty-foot

reflector of Sir William Herschel, and which is at least of the twentieth order of magnitude according to the scale used by this astronomer, belongs in fact to the three hundred and twentieth order of magnitude on the photometric scale. The geometric scale offers none of these inconveniences, although, on the other hand, it leaves something arbitrary in the choice of the constant factor of the progression. In both scales the standard of magnitude must be adopted as the fixed point of departure; this is an arbitrary point, whose selection demands much careful consideration. The choice of this unit of brightness may depend upon the following considerations: First, it may be a star of invariable brightness (if such exist). Second, it might be an artificial light, if we take means at hand for producing a light of constant value. Third, it may be determined by the effect upon either the eye itself, or upon the inert bodies that are employed in the photographic process. As regards the eye, it should be remembered that the image found upon the retina depends upon the more or less perfect adjustment of the eye of the observer, and, second, that the aperture of the pupil is variable within very considerable limits. These two latter sources of uncertainty may be remedied by simple means, when it will be found that it is highly convenient to adopt as a standard the faintest stars visible to the normal eye. This unit having been determined by many observers for many stars, the average of all will be a unit representing the average sensitiveness of the human eye, and independent of fluctuations in the brightness of the stars, and which therefore is sensibly constant.--Bibliothèque Universelle, 1874, 209.

FLOW OF AIR THROUGH ORIFICES. An extensive series of observations has been made upon the flow of air through orifices, and its discharge under great pressures, by Professor Fliegner and Dr. Zeuner, of the Polytechnic School at Zurich. The velocity of discharge can be obtained theoretically from the kinetic theory of the constitution of gases, according to which theory the molecules are, at relatively great distances from each other, moving in straight lines, except when they impinge on each other, or on the walls of the contained vessel, in which cases they rebound as if perfectly elastic. Applying