may mention tinfoil, sheets of copper and brass 0'2 mm. thick, a sheet of aluminium o'4 mm. thick, a steel lamina o'05 mm. thick, a silver leaf o'r mm. thick, a paper booklet, containing twenty-one gold leaves, a glass sheet, o'i mm. thick, a sheet of mica of o'15 mm., a plate of Iceland spar of 0.4 mm., a block of paraffin of I cm., a beech board 1 cm., a plate of ebonite of I mm., etc. Fluor spar is but slightly transparent with a thickness of 5 mms., similarly sulphur 2 mms. thick, and glass 1 mm. thick. These results I only give as a first indication, for when they were obtained, the co-existence of four different species of radiations, which may have very different properties, was not taken into account (note 5).

It will be highly interesting to investigate whether other sources, and in particular the sun, do not emit analogous radiations to those we are dealing with in the present communication, and also whether the latter produces any calorific action (note 6).

Now, ought these radiations in reality to be considered as akin to the large wave-length radiations discovered by Professor Rubens?

с

Their common origin in the emission of an Auer burner is favourable to such a view, as is also the opacity of rock-salt and of water. But on the other hand, for Auer rays, the transparency of metals and other substances opaque to Rubens' rays constitute an apparently radical difference between the two sorts of radiations (note 7).

On New Sources of Radiations capable of traversing Metals, Wood, etc., and on New Actions produced by these Radiations (May 25, 1903).

While investigating whether radiations analogous to those whose existence I recorded in the emission from an Auer burner (see p. 13) are not to be met also in other sources of light and heat, I established the following facts: the flame of an annular gas-burner emits such radiations; the chimney, however, should be removed, on account of the absorption of the rays by glass. A Bunsen burner scarcely

produces any.

A piece of sheet-iron or silver, heated to dull redness by a Bunsen burner, placed behind them, gives off rays at about the same rate as an Auer burner.

A plate of polished silver was arranged so that its plane made an angle of 45° with the horizontal plane. This plate having been heated to cherry-red by a Bunsen burner, its upper face emitted rays analogous to those of an Auer burner. A horizontal pencil of these rays, after traversing two sheets of aluminium of 0'3 mm. total thickness, sheets of black paper, etc., was concentrated by a quartz lens ; with the aid of the small spark, the existence of four focal regions was ascertained. I further found that the action on the spark was much more pronounced when the spark was arranged vertically-that is, in the plane of emission-than when it was normal to this plane. The new radiations emitted by the polished plate are therefore polarized, as are the light and heat emitted at the same time. The silver plate having been covered with lampblack, the intensity of emission increased, but the polarization disappeared.

The foregoing leads one to think that the emission of radiations susceptible of traversing metals, etc., is an extremely general phenomenon. First observed in the emission of a focus tube, it was also met in that of ordinary sources of light and heat. For shortness, I will henceforward designate these rays by the name of "N" rays.1

I would draw attention to the fact that these "N" rays comprise a very large variety of radiations; for while those which issue from an Auer burner have refractive indices greater than 2, there are others, amongst those emitted by a Crookes' tube, whose index is inferior to 1.52, for if a pencil of these rays is made to impinge on an equilateral quartz prism, parallel to the edges and normal to one of the faces, an emerging pencil is obtained which is very much spread out.

Up to this time the only means of detecting the presence of "N" rays was by their

1 From the name of the town of Nancy, these researches having been made at the Nancy University. In conformity with a usage which has become established, I now employ the letter "N" instead of "n," which I had at first adopted.