This historic book may have numerous typos and missing text. Purchasers can download a free scanned copy of the original book (without typos) from the publisher. Not indexed. Not illustrated. 1891 Excerpt: ...residuals still occur at both ends of the scale. The same observations appear in Table XLV., p. 194. In this case the mean result for each star is used instead of the individual observations. Stars of the fifth and sixth magnitudes are the most numerous, and for them the agreement is good. Even here small systematic ...
Read More
This historic book may have numerous typos and missing text. Purchasers can download a free scanned copy of the original book (without typos) from the publisher. Not indexed. Not illustrated. 1891 Excerpt: ...residuals still occur at both ends of the scale. The same observations appear in Table XLV., p. 194. In this case the mean result for each star is used instead of the individual observations. Stars of the fifth and sixth magnitudes are the most numerous, and for them the agreement is good. Even here small systematic errors are perceptible, which become very marked for the bright and faint stars. The signs have opposite meanings in the two tables. In Table XXIV. a positive sign denotes that the star as measured appeared to be too faint as compared with other stars in the same plate. In Table XLV. a positive residual denotes that the star appears to be too bright according to its photographic magnitude as compared with the results given in the Harvard Photometry. To see how far the second source of error affects the residuals of Table XLV., they are grouped in Table XLVI. according to the magnitudes in the Harvard Photometry instead of those of the Draper Catalogue. The four portions of Table XLVI. correspond to the last seven lines of each of the four pages of Table XLV. The class of spectrum is given in the first column. For the bright stars the results are nearly the same for Tables XLV. and XLVI. In both cases the residuals are negative, and are larger numerically the brighter the star. Accordingly bright stars will appear brighter in the Harvard Photometry than in the Draper Catalogue, and if two such stars differ by one magnitude in the Harvard Photometry they will differ by less than that amount in the Draper Catalogue. This difference is probably real, and due to the slight difference in scale indicated in Table XXIV. Of course the large negative residuals throughout the portions of the tables relating to spectra of classes G and K are due to the red ...
Read Less