Imagini ale paginilor
PDF
ePub

(4) The largest number of the variable stars of Type III are found on the borders of the Galaxy.

(5) This zone is from 15° to 25° of Galactic latitude.

(6) These variable stars show a tendency to form into groups.

I have in addition investigated the variation and period of these stars and can find no connection with their position in reference to the Galaxy. Lastly to trace any connection between the variable stars and the stars of Type III, I plotted down all the known variables of III !! and III !!!. These show a distinct tendency to grouping, but not with any reference to the Galactic latitude of 20° as in the case of the variable stars. The third type stars increase in number as the Milky Way is approached, and in two of the cases, where there are groups of variables in Cygnus and Delphinus, the third type stars group as well; on the other hand, there is a strong group at Gal. long. 60°, lat. S. 28°.

TOW LAW, DARLINGTON, ENGLAND.

July 6, 1899.

OBSERVATIONS OF COMET SPECTRA.

By W. H. WRIGHT.

COMET 1898 (Perrine).

THE spectrum of this comet was observed visually on May 9, 1898. It was of the usual type: the three characteristic bands superposed on continuous spectrum, which was relatively strong. Photographic observations were not made.

COMET 1898 VII (Coddington).

Observations on June 11 by Professor Campbell and the writer showed the spectrum to be of the usual type, except that the banded spectrum was very faint as compared with the continuous. The yellow and blue bands were seen only with difficulty. To the writer the banded spectrum seemed to be stronger in the outlying parts than in the nucleus. The spectrum was again observed on June 16 with practically the same results, except that the bands in the nucleus seemed to be relatively stronger than on the previous occasion. The spectrum was too faint to be photographed.

COMET 1898 X (Brooks).

Observations were made on November 3, 1898, by Professor Campbell and the writer. The three chief bands were easily visible, the one in the green being apparently much brighter than usual. It was roughly estimated as being from four to six times as bright as the others. The continuous spectrum was very weak, being visible only in the nucleus with a wide slit Nothing was visible above or below the three bands.

COMET 1899 a (Swift).

The visible spectrum of this comet resembled that of 1898 x, inasmuch as it consisted almost entirely of bright lines.

The

three carbon bands were bright, but the dispersion used was too low to show further details.

The spectrum was photographed with a spectrograph attached to the twelve-inch equatorial on May 4, 8, and 11. The photograph of May 8 was the best, showing the blue band resolved, and the cyanogen lines at AA 3871 and 3883, with considerable detail between.

Further photographs were obtained on June 5 and 6 with a spectrograph attached to the thirty-six-inch telescope. By the use of a correcting lens, placed one meter in front of the slit, a color curve satisfactory for this class of work is obtained over the entire region covered by the spectrograms. The following are the results of measurements:

[blocks in formation]

The slight correction (— A. U.) for motion of the comet in the line of sight has been taken account of, but the results are uncorrected for the effect of slit-width as discussed by Professor H. Kayser (A. and A. P., 13, 367). The general shift toward the violet indicated by the spectrogram of June 6, is not, however, entirely to be accounted for by the Kayser effect, as will be seen by the following consideration: The hydrogen tube used on

this date for comparison contains an impurity giving the bands at λ 3871 and 3883. With reference to these the corresponding comet bands are perceptibly shifted toward the violet. The spectrograph used in these determinations is one constructed for the purpose, and which has not otherwise been tested. Although it is in many respects most efficient, I am inclined to consider that the shift referred to is instrumental, the result probably of manipulation of the slit while photographing the comparison spectrum.

Most of the lines here observed have counterparts in the spectra of carbon and cyanogen, and the majority of these are doubtless to be accounted for by the presence of the two substances. In addition to those indicated above there is a line of wave-length 4099.2, which approximates closely to that of the comet line 4101. Professor Kayser, however, thinks the two are not identical, the cyanogen line being in his opinion much too faint. The identity of the comet line being open to doubt, it is of interest to note that it is very close to H8 (4101.9). The proximity must not, however, be held to prove identity.

In appearance, the head of the comet was very diffuse, with a nucleus some 4′′ in diameter. In the spectrum of the nucleus A 4052 is fully as bright as λ 3870, but the latter line extends out into the fainter parts of the comet's head more than four times as far as the former. In fact the lines XX 3870 and 3880 experience only a gradual increase in brightness in the region of the nucleus, and extend the entire length of the slit. In the cases of all the other lines the change is quite abrupt. This must be taken to indicate a marked difference between the spectrum of the nucleus and the spectra of the outer parts of the head.

The spectrograms have been compared with some of comet 18936 (Rordame) and 1894 b (Gale), secured by Professor Campbell with a spectrograph attached to the thirty-six-inch telescope. The correspondence is not exact, as Professor Campbell had not the advantage of the photographic corrector referred to above. This accounts sufficiently for such differences as exist in the estimates of relative intensity of the lines. Taking into

consideration the difference in color curves there is no evidence of any variation in the type of spectrum.

The constants of the spectrograph used in these determinations are:

[blocks in formation]
« ÎnapoiContinuă »