2 Reduction

The reduction of the 1959 material was made by means of the electronic computer of the central computing institute of the Leiden University. In this part of the reduction the authors were assisted by Mr. L. Maitimo. The 1960 observations were reduced by hand. Comparison with the 1959 material showed, for most of the 1960 material, systematic differences in the V colour, depending on the V-B colour index. This was caused by the use of a photomultiplier for the V channel with a colour sensitivity different from the one used in 1959. From comparisons between stars observed both in 1959 and in 1960 the following relation between V(1959) and V(1960) could be derived:

$$V(1959) = V(1960) - 0.254\{V(1960)-B(1960)\}.$$ (1)

The Walraven VBLUW colour system is constructed such that it is independent of the photomultipliers used, except for the V channel, where the red edge is determined by the photomultiplier used. Expression (1) was used to transfer the 1960 material to the 1959 photometric system. After applying this correction, it turned out that still systematic differences between 1959 and 1960 existed. In order to combine the 1959 and the 1960 results the latter were reduced to the 1959 system by applying corrections to the observations made in 1960. It turned out that a still better agreement with the 1959 observations could be obtained if these systematic differences were determined from and applied to the individual nights. In this way the values listed in Table 4 were obtained. To give an impression of the systematic corrections applied to the 1960 observations, the correction applied to the B-U colour index is -0.007 and to the U-W colour index amounts to +0.012. The other corrections were all considerably smaller; they amount to 0.002 on the average. These corrections to B-U and U-W are found again in a comparison of OB star observations made in 1965 and 1968 by the first author (van Houten, unpublished) with the 1959 values given here. Its origin is unknown.

It was assumed that the photomultiplier used in the V channel in 1959 was the same as the one used since 1965. However, comparisons of the values given in this paper with a compilation by Dr. E. Brinks (unpublished) of observations in the Lub & Pel (1977) system clearly showed a colour term in the V-B colour indices. The relation found is

$$(V-B)_{\rm st}=(V-B)_{59}-0.076(V-B)_{59}-0.003.$$ (2)

Here $(V - B)_{\rm st}$ refers to the Lub & Pel colour system and (V-B)₅₉ to the system used in this publication. A corresponding formula is valid for the V channel. It can be expected that a similar transformation can be found from a comparison of the present material with Walraven & Walraven (1977). The relation actually found is slightly different:

(V - B)₇₇ = (V - B)₅₉ - 0.055 (V - B)₅₉ + 0.001.

Here (V - B)₇₇ refers to the observations published in Walraven & Walraven (1977). The cause of the difference between (2) and (3) is probably that Walraven & Walraven (1977) was not corrected for the transformation found by Lub & Pel (1977):

$$(V - B)_{\rm st} = (V - B)_{77} - 0.026 (V - B)_{77} - 0.002.$$ (4)

If this is true, (4) can be derived from (2) and (3). Adding (3) and (4) yields:

$$(V - B)_{\rm st} = (V - B)_{59} - 0.079 (V - B)_{59} \nonumber$$

which is, within the limits of accuracy, equal to (2). It follows that for a transformation of the observations presented here to the standard system of Lub & Pel (1977) Eq. (2) should be used. Transformation of Walraven & Walraven (1977) to the same standard system needs Eq. (4).

In Table 4 the original 1959 reduction is listed in four decimals. This is more than the observational accuracy warrants, but the reduction was made in this way and it was judged an unnecessary loss of time to reduce them to a three-decimal value. All additions and corrections to this list were made with three-decimal values. Thus three-decimal values in Table 4 mean that an error in the original table has been corrected, that the observations in 1959 were combined with observations in 1960 or that the star was observed in 1960 only. These remarks are meant to explain why some entries in Table 4 have four decimals and other have only three. They are not meant to indicate which observations were made in 1959 and which in 1960. Readers who want to obtain this information can request them from the first author.