III. That the thickness of the system of rings is increasing from the inner margin of the dusky ring to the outer border of the ring C, as proved by the form of the shadow of the planet thrown upon the rings.

IV. That the cloud-forms seen near the outer border of the ring Cattain different heights, and change their relative position, either by the rotation of the rings upon an axis, or by some local cause; as indicated by the rapid changes in the indentation of the shadow of the planet.

V. That the inner portion of the dusky ring disappears in the light of the planet at that part which is projected upon its

disk.

VI. That the planet is less luminous near its limb than in the more central parts, the light diminishing gradually in approaching the border.

VII. That the dusky ring is not transparent throughout, contrary to all the observations made hitherto; and that it grows more dense as it recedes from the planet; so that, at about the middle of its width, the limb of the planet ceases entirely to be seen through it.

VIII. And, finally, that the matter composing the dusky ring is agglomerated here and there into small masses, which almost totally prevent the light of the planet from reaching the eye of the observer.

Cambridge, Dec. 1, 1875.

ART. LIX.-Curve of Eccentricity of the Earth's orbit; R. W. MCFARLAND.

THE diagram on the following page gives the curve of eccen tricity as found in Croll's work, and also in Stockwell's. The former was computed with a view of explaining the climate of past time; the latter, in examining the moon's mean motion. It will be seen that the curves have the same general form, differing in the value of the ordinates, more than in the position of the maximum and minimum points. I have recomputed the values by Le Verrier's old formulæ, and find Croll's figures correct in most cases, and not in error to the amount of 001 except in one instance. My work extends over 1,100,000 years, and is for points nearer together than Croll's. Stockwell's is doubtless nearer correct than the other, but there is substantial agreement. The two curves for the last 40,000 years, and for the next 30,000, are almost identical; so nearly is this the case, that only one line is drawn in the figure for those periods.

The following table gives the difference between perihelion and aphelion, the sun's mean distance being 92,000,000 miles.

At 40,000 the difference disappears.

Ohio Agricultural and Mechanical College, Columbus, April, 1876.

ART. LX. On a Bolide of January 31st, that passed over Kentucky; by J. LAWRENCE SMITH.

On the afternoon of the 31st of January, at five and one-half o'clock, while crossing one of the streets of Louisville, my attention was suddenly arrested by a magnificent meteor crossing the heavens. I first saw it at 60° above the horizon, and it disappeared to my view behind the houses at an elevation of about 20°. It was pear-shaped, and very bright, and remained in view for two or three seconds. Its apparent size was about onesixth that of the disk of the moon. It did not separate while under my observation, nor did I hear any noise. On asking, through the medium of the public prints, for the observations of others, I received some fifteen communications, from an area one hundred and twenty miles in diameter.

To a number of the observers an explosion was visible, producing several flashes of light, of different colors. This occurred about ten degrees above the horizon. On exploding, all the fragments disappeared instantly except the largest, which also disappeared before reaching the horizon. Some observers saw sparks flying off from the ball, and a short stream of light behind it. One or two think they heard a whizzing noise and at the time of bursting heard the explosion. All agree in stating that the direction was from northwest to southeast. Nothing has yet been heard of any fragments having been collected. My opinion is that it fell about the range of the Cumberland Mountains in Kentucky, or in the northeast of Tennessee.

This is the third bolide in three consecutive years that it has been my good fortune to witness, in their passage over Louisville, but the fragments of none of them have been obtained. They all were passing from the northward to southward.

ART. LXI.-Notes on the Sensitiveness of Silver Bromide to the Green Rays as modified by the Presence of other Substances; by M. CAREY LEA.

SEVERAL investigations made at different times during the past years on the sensitiveness of silver bromide to rays of different refrangibilities, led me to the conclusions: 1st, that its sensitiveness to the different rays could be distinctly modified (increased or diminished) by the presence of various bodies, colored and colorless; 2d, that no relation could be traced between the color of the modifying body and the refrangibilities of the rays to which the sensitiveness was modified. During the past winter, I have carefully re-examined the question, which is an important one both in its theoretical and practical relations, and have found my conclusions in all respects confirmed. And I have during the past winter been occupied with a single portion of the subject, namely, the action of the green rays, as a special study.

In this investigation I have pursued the same general method as before; that is, I have used colored glass whose transmitted rays have been carefully studied with the spectroscope. There is no doubt in my mind that this method of examination is capable of giving results as valuable as those obtained by the use of the spectrum. It may be said in fact that the subject requires for its full elucidation, the use of both methods. To those who may imagine that the results of the exposure to the spectrum are the more reliable, it may be interesting to have specified the weak points of that method.

The relative strength of the impression produced by different portions of the solar spectrum on a sensitive surface will always depend upon the intensity of the light employed and the length of exposure. It has already been proved that silver iodide and bromide are sensitive to every part of the spectrum. It follows that any and every part of the spectrum may be photographed upon plain iodide or bromide if only the exposure is sufficiently prolonged. The longer the exposure, the stronger will be the impression produced by the less refrangible rays as compared with the more. Also, it is known that by a system of masking, Prof. John Draper has succeeded in photographing the whole spectrum at once; that is to say, he was able to hold back the action of the more refrangible rays until that of the less was sufficiently strong. The masking was of course done with red or yellow media. Now, when pigments of these colors are extended over sensitive films, what is this but a sort of masking, which retards the action of the more refrangible

rays while freely permitting that of the less? Even supposing these pigments to be chemically inert they would check the action of the blue end of the spectrum and render it possible by a longer exposure to obtain both together. Even a longer exposure would not be absolutely essential, a more powerful or a longer development may take its place, being rendered possible by the diminished impression of the blue end.

If, then, it is alleged that by coloring a film of silver bromide with red pigments, the sensitiveness to yellow rays is increased, we are at once moved to reply that such a result is no proof of a chemical or photochemical action exerted by the red pigment; that precisely the same result might be expected if the red substance were chemically inert, or if it were extended over a glass surface and simply interposed in the path of the rays, between the prism and the film, without even coming into contact with the latter. Many results that have been published are liable to this fatal objection. It has been proposed to modify the form of experiment by applying the color to the front of the plate and exposing on the back, through the glass. But even this does not remove the difficulty. The collodion film containing the silver bromide is exceedingly thin and when moistened, very porous and absorbent. Any soluble color applied, dissolved in water or alcohol, instantly penetrates it through and through and even moistens the glass under it. And all the colors mentioned by other experimenters as having been subjected to this experiment, are soluble. It is on these results that theories have been based, and they are all susceptible of full and complete explanation in the manner just mentioned.

In matters of photographic experiment such as these now under discussion, there are always three distinct factors: The sensitiveness of the matter, the force of the impression (depending upon the intensity of the light and the length of exposure), and the development. This makes the investigation difficult and deceptive. If we take two identical sensitive films, and submit them to the same exposure, we may get quite different results by varying the development, or with an identical film and development, by varying the exposure. The amount of error and deception liable to be introduced in this way is known to none but those made familiar with it by experience, and consequently for accurate results, these sources of error must be eliminated. There is but one way of doing this: a film must be taken, one portion of it must be washed over with the substance whose action is to be studied, and then the two portions, the plain and that which has been treated, must be simultaneously exposed and simultaneously developed. Both portions must receive the same rays, of equal intensity and for au