Astron. Astrophys. 121,35-41 (1983) ASTRONOMY AND ASTROPHYSICS Properties of Am, <5Del, and 8 Set stars in the VBLUW system M. J. J. Wiertz and A. M. van Genderen Leiden Observatory, Postbus 9513, 2300 RA Leiden, The Netherlands Received December 1, accepted December 23, 1982 Summary. VBLUW photometric observations of 115 Am, 5 Del, and 3 Set stars are presented and discussed. The spread with respect to the main sequence in the two-colour diagrams (corrected for reddening if possible), is discussed and can be mainly attributed to gravity effects. Exceptions to this rule can be recognized from a comparison of their positions in these diagrams and in the reddening independent diagram [L — U~\/[B — L~] : some 3 Set stars must have suffered of relative high reddenings and others are presumably metal poor. Temperatures and gravities compare reasonably well with those of Davis Philip et al. (1976) based on uvbyfi photometry. However, the temperatures derived by Babu and Shylaja (1981, 1982) and based on spectral energy distribution, reveal that the photometric temperatures (and thus also gravities) are underestimated. This is caused by line blanketing on the colour indices b — y and V— B, of which the size as a function of the temperature can be roughly estimated. The relation between the temperature sensitive colour indices V-B (of the VBLUW system), b-y (ubvyfi system) and B2- V1 (Geneva system) for A/F type stars is investigated. It appears that the V— B is more sensitive to line blanketing than the other two indices. A PLC relation for the colour V— B is derived using Breger's (1979) relation for the colour b — y. Key words: photometry - metallic line stars - 5 Scuti stars - 3 Delphini stars 1. Introduction So far no extensive observations of Am, S Del, and <5 Set stars exist in the Walraven VBLUW photometric system. Only sporadic studies of a few specimen have been made: Ponsen (1963): gPup; Ponsen and Oosterhoff (1966): <5Sct and 5Del; Oosterhoff and Walraven (1966): all the three variables; van Genderen (1973): again 3 Del and Lub (1979) included these three variables in his discussion on the properties of RRLyrae stars in the VBLUW system. Since many observations and discussions exist of Am, 3 Del, and 3 Set in the Stromgren and Geneva photometric systems (see for references for example Hauck and Curchod, 1980; Gomez et al., 1981), it was worthwile to obtain also observations of this very complex field of A and F type stars in the VBLUW system. Send offprints requests to: A. M. van Genderen 2. The observations and reductions The observations were made with the simultaneous VBLUW photometer of Walraven attached to the 90-cm lightcollector of the former Leiden Southern Station (at the S A AO annex) in South Africa mainly in the years 1977. A description of the photometric system is given by Walraven and Walraven (1960), Rijf et al. (1969), and Lub and Pel (1977). In total 115 stars, usually brighter than the 7 th mag, were selected. They consist of 64 Am type stars, 9 3 Del and 42 3 Set type stars. Each program star was measured relative to standard stars in three to six nights. Table 1 tabulates the average standard deviation in the average photometric parameters for the three star types separately (for the VBLUW system in log intensity scale and for the UBV system with subscript J in mag). Mean errors are thus smaller. The larger standard deviation for the 3 Set star is likely caused by their variability. The classification of stars into these two types depends only on their spectroscopic character, while the classification into the <5 Set class is mainly based on their instability. A star can thus be of the 3 Del- as well as of the 3 Set type, like in the case of the prototypes 3 Del and 3 Set. Because of this overlap of classification criteria and confusing about naming, there is sometimes much ambiguity in literature about the true type of a star (see for an extensive discussion on these problems Breger, 1979, 1980). The V of the VBLUW system can be transformed into the V of the UBV system (with subscript J) with the aid of a formula given by Pel (1976). The V—B colour index of the VBLUW system can be transformed into the equivalent B— V colour index of the UBV system (with subscript J) by Table 7 in Walraven et al. (1964), but corrected for a slight change in the V passband (Lub and Pel, 1977). Tables 2 and 3 tabulate the photometric parameters (VBLUW and UBV systems) for the 115 stars. The number of times that a star has been measured is indicated in the column "N". Table 1. The average standard deviation in the photometric parameters for the three types of stars in the VBLUW system and the UBV system (with subscript J) Type V V—B B-U U-W B-L Vj (B-V)j (log intensity) (mag) Am 3 Del 3 Set 0.007 0.006 0.010 0.006 0.007 0.006 0.006 0.005 0.006 0.009 0.004 0.010 0.003 0.012 0.004 0.018 0.015 0.02b 0.015 0.015 0.012 © European Southern Observatory • Provided by the NASA Astrophysics Data System 36 Table 2. The photometric parameters of the Am stars in the VBLUW system (in log intensity scale) and in the UBV system (in mag scale and with subscript J) HR HD Name V V-B B-U u-w B-L N VJ (B-V)J 323 6619 0.094 0.049 0, 459 0. 136 0.205 3 6.63 0. 123 547 11522 0.410 0. 100 0.426 0. 160 0. 197 3 5.83 0.255 4599 104671 105702 a1 Cru 1.014 0. 109 0.404 0. 171 0. 176 3 4.32 0.280 4629 11 Vi r 0.461 0. 136 0.408 0. 158 0.231 3 5.70 0.320 0.255 4650 106251 12 Vir 0.411 0.100 0.410 0.143 0.202 3 5.83 4703 107566 5» rius 0.689 0.073 0.466 0.149 0.194 3 5.14 0.184 4794 109536 110575 110951 0.698 0.092 0.406 0.131 0.200 0. 196 3 5.11 0.232 4836 0. 169 0. 100 0.403 0. 140 0.157 3 6.44 0.255 0.330 4847 32 Vir 0.622 0. 124 0. 420 0.209 3 5.30 4872 111588 0.462 0.246 0.045 0.477 0.377 0. 146 0. 189 5 5.71 0.112 4900 112097 41 Uir 0.112 0.153 0.190 3 6.24 0.285 5008 115331 0.413 0.076 0.441 0.132 0.209 4 5.82 5.87 0.192 0.104 5040 116235 64 Vir 0.399 0.042 0.429 0.099 0.192 4 5093 117651 0.209 -0.007 0.384 0.083 0. 143 4 6.35 -0.025 5094 117661 73 Vir 0.343 0.078 0.444 0.128 0.208 4 6.00 5.91 0.197 5171 119938 0.377 0.107 0,399 0.139 0.201 4 0.275 5349 125158 0.658 0. 105 0.437 0.152 0.229 5 5.21 0.265 5359 125337 A Vi r 0,942 0.415 0.053 0.447 0. 126 0. 199 4 4.51 0. 133 5401 126504 0. 122 0.426 0.155 0.221 4 5.B2 0.310 5531 130841 c*1 Lib 1 .651 0.058 0.437 0. 127 0. 195 4 2.74 0. 148 5577 132219 59 Haa 0.483 0.102 0.441 0. 158 0.207 4 5.65 0.260 5591 132851 60 HU3 0.414 0.066 0.464 0.144 0.202 4 5.83 0. 166 5663 135235 0.360 0.078 0.419 0. 134 0.201 4 5.96 0 . 197 56S2 135730 0.232 0 . 545 0 . 068 0.437 0.139 0 . 208 0.208 4 6.28 5.50 0.172 0. 163 5762 138413 0.065 0.458 0.151 4 5848 140417 »1 Lib 0.579 0.091 0.410 0.146 0.205 4 5.41 0.230 5872 5875 141296 0.297 0.535 0.115 0.373 0.153 0.195 4 6.11 0.290 0.115 141378 141795 0.046 0.428 0.110 0. 197 4 5.53 5892 £ Ser 1 .260 0.058 0.432 0. 120 0.199 4 3.71 0.148 5900 142049 144197 0.431 0.855 0.141 0.407 0.166 0. 150 0.218 4 5.77 0.357 0.235 5980 5992 6129 i Nor 0.093 0.436 0.213 4 4.72 144426 0.234 0.032 0.459 0. 129 0.181 4 6.28 0.079 148367 Oph 0.892 0.073 0.414 0.121 0.203 4 4.63 0. 184 6193 150366 0.321 0.081 0.429 0.137 0.208 4 6.06 0.205 6250 151956 47 Her 0.555 0.038 0.443 0.116 0.200 4 5.48 0. 094 6366 154783 0.371 0.107 0.443 0. 177 0.222 3 5.93 0.270 6358 155375 0.115 0.029 0.443 0.110 0.192 3 6.58 0.073 6611 161321 0.267 0.075 0.473 0. 137 0.215 3 6.19 0.189 6957 170920 171819 61 Ser 0.367 0.059 0.504 0. 149 0.182 3 5.95 0. 148 6986 0.417 0.096 0.440 0. 169 0.200 4 5.82 0.242 0.200 6988 171856 0.375 0.079 0.472 0. 155 0.204 3 5. 92 7011 172546 26 S3r 0.262 0.096 0.429 0. 146 0.209 3 6 .20 0.242 7274 179009 X Pav 0.244 0.076 0.478 0. 177 0.200 3 6.25 0.192 7369 182490 183552 2 S3e 0.234 0.435 0.022 0.121 0.418 0.458 0.098 0.174 r> 6.29 5.77 0.054 7411 0.192 0.222 3 0.307 7431 184552 51 SSr 0.489 0.066 0.463 0. 146 0.215 3 5.64 0.166 7498 7510 186219 70 Psv 0.592 0.098 0.434 0. 183 0.210 3 5.38 0.247 186543 V Tel 0.607 0.072 0.448 0. 174 0.210 3 5.34 0.181 7579 188097 0.446 0.090 0.463 0,203 0.217 3 5.74 0.227 7624 189118 91 S2r 0.627 0.066 0.435 0.122 0.201 0.206 0.205 3 5.30 0.166 7990 198743 M Act r 0.843 0. 126 0.411 0. 159 3 4.75 0.320 0.158 B018 199443 0.383 0.063 0.445 0. 130 3 5.91 8045 200052 0.322 0,017 0.438 0.120 0. 193 3 6.07 0.041 8278 206088 i Cap 1 .268 0. 126 0.450 0. 174 0.225 4 3.68 0.345 8293 206546 0.256 0.097 0.439 0.150 0.215 3 6.22 0.244 8295 206561 206677 44 Cap 0.390 0.102 0.428 0.405 0.160 0. 194 4 5.88 0.260 8302 45 Cap 0.366 0 .086 0.130 0. 195 4 5.95 0.237 8362 8410 208149 0.283 0.625 0.093 0.457 0. 176 0. 154 0.207 0.217 3 6.15 5.30 0.235 0.225 209625 32 Aa r 0.089 0.453 4 8583 213464 58 Ao r 0.181 0.117 0.389 0. 158 0. 188 5 6.40 0.298 0.031 8616 214484 0.448 0.013 0.445 0.126 0.163 4 5.75 8662 215545 0.113 0.116 0.414 0.162 0.208 4 6.57 0.294 8722 216823 T1 Gru 0.458 0.087 0.115 0.460 0.435 0.166 0.217 4 5.71 0.220 8944 221675 14 Psc 0.391 0.160 0.220 4 5.88 0.291 3. The two-colour diagrams In order to construct the two-colour diagrams with reddening free colour indices, most stars were therefore corrected with the reddenings given by Davis Philip et al. (1976) in the uvbyfl system: E{b — y). For a number of stars no reddening was available. Therefore we derived them ourselves by the same method as described by Davis Philip et al. and Davis Philip and Egret (1980). For this purpose the necessary uvbyfi parameters were taken from the catalogue of Hauck and Mermilliod (1980). If stars were also not listed in this catalogue, their reddenings were adopted to be zero. Nearly all reddenings are small anyway, because of the proximity of the program stars. The reddenings seldom surpass E{b-y) = 0.03 or E(B-V)j = 0.05 or E{V-B) = 0.02. If the computations of the reddenings resulted into negative values, or when the reddenings listed by Davis Philip et al. were negative we adopted them to be zero. The total number of stars with zero reddening is about 30. The reddening E(V— B) for the VBLUW system could then be computed with the aid of the relation: E(V-B) = E(b-y)/1.653 (Lub and Pel, 1977). Figure 1 shows the three two-colour diagrams for the Am stars (the panels on the left) and the 8 Del and 5 Set stars (the panels on the right). The main sequence and reddening trajectories are indicated. In the V—B/B— U diagrams we sketched the Teff = constant lines according to the computations of Lub and Pel based on the Kurucz (1979) models for solar composition and umj = 2kms_1. In order to prevent confusion only the logg = 3.5 line is shown. The main sequence is situated in between the lines for logg = 4 and 4.5. © European Southern Observatory • Provided by the NASA Astrophysics Data System M. J. J. Wiertz and A. M. van Genderen: VBL UW photometry of Am, <5Del, and «5 Set stars Table 3. The same as Table 2 but now for the 5 Del stars (first group) and the S Set stars (second group) 37 HR HD Name V V-B B-U U-W B-L N V (B-V) 421 8829 47 Cet 0.512 0.121 0.358 0. 137 0. 190 3 5.57 0.307 2094 40292 43760 0.628 0.043 0.112 0.140 0.380 0.151 0.191 3 5.29 6.74 0.285 0.354 2255 6 Mon 0.459 0. 182 0.211 3 3228 68703 0. 160 0.114 0.426 0.121 0.218 3 6.46 0.290 3649 74198 157919 0.893 0.003 0.159 0.105 0.422 0.105 0.162 3 4 .64 0.004 0.400 6492 6561 55 Oph 1 .029 0.389 0.193 0.219 4 4 .28 159876 S Ser 1 .329 0.434 0. 159 0.210 5 3.53 0.265 8322 207098 s C3P 1.590 0.110 0.395 0. 153 0. 197 4 2.88 0.280 8787 218227 e Gru 1 .028 0.167 0.422 0.195 0.240 4 4.28 0.420 242 4919 fi Phe 0.462 0.141 0.416 0. 183 0.216 A 5.70 0.357 6870 8511 ES Tuc -0.257 0.091 0.376 0.150 0. 170 3 7.50 0.230 401 AV Cet 0.250 0.090 0.409 0.140 0.155 0. 199 4 6.23 0.227 0.302 431 9065 WZ Scl 0.094 0.119 0.388 0.185 5 6.62 9133 XX Scl -0.826 0.081 0.437 0.132 0.199 4 8.93 0.205 515 10845 VY Psc 0.116 0.105 0,444 0.154 0.201 0.195 3 6.57 0.265 812 17093 24550 uu Ari 0.666 0.089 0.405 0.132 0.155 2 5.19 0.225 V479 Tau -0,219 0. 164 0.430 0.225 1 7.39 0.413 1225 24832 DL Eri 0.270 0. 101 0.426 0.159 0.207 2 6.18 0.257 1298 26574 or" Eri 1 .122 0.126 0.413 0. 168 0.207 3 4.05 5 .55 0.320 1351 27397 27459 V483 Tsu 0.522 0.654 0.119 0.398 0.134 0.202 0.207 2 0.300 1356 Y696 Tau 0.092 0.423 0.118 2 5.22 0.232 1412 1611 23319 32045 e* T3U 1.392 0.085 0.452 0.130 0.200 3 3.38 0.215 s Eri 0.832 0.107 0.452 0.194 0.197 3 4.78 0.270 1653 32846 X Cae 0.218 0. 124 0.387 0. 167 0. 180 3 6.31 0.313 2100 40372 V1004 Ori 0.375 0.087 0.446 0.170 0,204 4 5.92 0.220 2107 40535 V474 Mem 0.295 0.118 0.409 0.172 0.203 3 6.12 0.300 2707 2989 55057 V571 Mon 0.572 0.15.3 0.114 0.428 0. 179 0. 135 0.207 3 5.43 0.290 62437 67523 AZ CMi 0.078 0.460 0. 202 4 6.48 0.197 3185 >> Pup 1 .625 0.160 0.425 0. 197 0.238 5 2.79 0.403 3265 69997 HQ H«a 0.224 0. 123 0.427 0.125 0.219 2 6 .29 0.310 3524 75747 RS Cha 0.320 0.089 0.420 0. 175 0. 199 3 6.06 0.225 3588 77140 FZ Vel 0.685 0.093 0.115 0.450 0.150 0.158 0.213 3 5.15 0.235 100363 SU Crt -0,701 0.377 0.187 1 8.61 0.260 106384 FG Vir 0.127 0.104 0.400 0.141 0. 195 3 6.54 0.265 4684 107131 FM Com 0.176 0.070 0.422 0.119 0. 157 0,186 1 6.42 6.69 0. 156 5005 115308 DK Mir 0.066 0. 127 0.396 0. 189 6 0.320 116994 V743 Cen -0.718 0.118 0.413 0.165 0. 193 5 8.65 0.300 5788 138917 i Ser 1 .222 0.101 0.426 0.165 0.205 4 3.80 0.257 6290 152830 V644 Her 0.198 0.131 0.382 0.154 0.201 6 6.36 0.330 153747 -0.221 0.054 0.446 0.130 0. 178 6 7.42 0. 136 6581 160613 0 Ser 1 .049 0.031 0.459 0.126 0.183 6 4.25 0.077 170625 172748 V668 Crs -0.277 0.085 0.145 0.472 0.416 0. 145 O.glS 6 7.55 0.215 0.368 0.472 7020 i Set 0.833 0.177 0.217 4 4.77 174553 V369 Set 1.007 0.520 0.190 0.447 0.157 0.229 0.212 6 9.36 7331 181333 Y1200 AcU 0.095 0.454 0.163 5 5.56 0.240 7340 7524 181577 P SSr 1.167 0.081 0.440 0.154 0.207 5 3.94 0.298 186786 NZ Pav 0.327 0.795 0.125 0.404 0. 174 0.193 5 6.04 0.315 7859 195961 f> Psv 0.173 0.434 0. 198 0.234 5 4.86 0.433 7928 197461 £ Del 0.967 0.106 0.402 0. 150 0. 192 4 4.44 0.270 8006 199124 201707 EM Aar 0.110 0.100 0.402 0, 147 0. 197 5 6.58 0.255 8102 EW Ao r 0.149 0.108 0.438 0.167 0.211 5 6.48 0.273 It is obvious that Am stars can be somewhat hotter than the other two types of stars, since they extend further to the blue, even up to 10,000 K. Most of the scatter in the diagrams is certainly due to the gravity effects, especially the B — U index is most sensitive for gravity differences, while the B—L is the least sensitive. From other studies we do know that many stars of these three classes are slightly evolved. Nevertheless it is necessary that a few other causes which can take part in the scatter are shortly discussed, although they are considered to be much smaller. 3.1. Binarity According to Abt (1965) and Abt and Bidelman (1969) all Am stars may possess a fainter and thus redder main sequence star as a companion. Consequently their colours may be slightly reddened, depending on the brightness of the companion. It is to be expected that this influence is small, not more than a few percent, since the redder the companions are, the lower their luminosities are. 3.2. Blanketing effects The enhanced metal line strengths, especially in the case of the Am type stars, may redden the V—B colour index. According to Babu and Shylaja (1982) the effect on (B— V)} is of the order of 0.2 mag near Teff ~ 10,000 K, but diminishes towards zero near Teff ~7000K (see Sect. 5). However, the position of the two hottest Am stars in Fig. 1 (HD 117651 and HD 214484) is on the right place considering the spectral types assigned to them by Houk and Cowley (1975) and by the HD catalogue as AOV and AO, respectively. Perhaps the Am characteristics are very weak in these stars. According to Hauck and Curchod (1980) the metallicity, being defined in the Geneva system by zlm2( = m2(obs) —m2(Hyades)), is greater for cooler Am stars than for hotter ones (their Fig. 3). Although the B, L, U, and W passbands are very sensitive for metallicity, the effect on the colour indices B—U and U—W will be partly cancelled. Further, a certain amount of scatter in Fig. 1 will be introduced by the fact that each metallic line star has its own spectroscopic characteristics. 3.3. Reddening by interstellar dust Inaccuracies in the reddening values applied in Fig. 1 and for a small number of stars even unknown reddenings, will in general introduce very small extra scatter of probably not more than a few hundredths of a mag. The average reddening which has been © European Southern Observatory • Provided by the NASA Astrophysics Data System 38 0 0.1 0.2 0 0.1 0.2 0.5 U-W I I I _ Am _ / 'V-* X i i i o 6 Del — "\ 'jo Scl • o • \ loooY . j* f J 9000«Y 7000 | ' 8000A | | 1 1 ' 1 A Jl A' 33 10000V /«» ••// 9000 \ / ^^T» / '7000 V'/ ' 8000 I I I — • — i i i I I I * •5*f*\ + + • **+ i i i 0 0.1 0.2 0 0.1 0.2 V-B Fig. 1. The two-colour diagrams for the Am Stars (on the left) and 5 Del and 6 Set stars (on the right). Sketched are the main sequence (full curve), a number of Teff = const lines and the log g = 3.5 line (dotted curves) in the V— B/B—U diagrams and the reddening trajectories (arrows). Special symbols will be explained in the text applied is E(B— 1^ = 0.02 mag only. A few 5 Set stars for which no proper reddening could be derived by the method mentioned before and which likely have suffered a much higher reddening, are HD 174553 (V 369 Set), HD 195961 (gPav), and HD 24550 (V 479 Tau). They are indicated by crosses in the right panels of Fig. 1. According to Hall and Mallama (1980) the reddening of the first one may amount to E(B— V)j = 0A2 adopting that its luminosity class is III. If we adopt that the star's intrinsic position is amidst the other d Set stars, we find E(B — V)j = 0.17 + 0.05 mag. This is in good agreement with the previous value. The reason that we suspect the other two stars of also relative high reddenings will be explained in Sect. 4. According to Davis Philip et al. (1976) the reddening for the second star is E(b — y) = 0.022, but this may well be underestimated by a factor of four. We shall return to this in Sect. 4. Thus we conclude that the bulk of the scatter in Fig. 1 is likely introduced by gravity differences. As a possible extreme specimen we like to mention the Am star HD 170920 (21 Ser) (triangles in Fig. 1). Its position in the V-B/B-L and V-B/B-U diagram points into the direction of a very low gravity: logg~3 but according to its position in the V — B/U — W diagram it may be somewhat higher. We shall indicate this star also in further diagrams by a triangle. Spectroscopy should throw more light on its evolutionary status. 4. The [L-U]/[B-L] diagram In Fig. 2 we show the reddening independent two-colour diagram [L— L/]/[B — L]. Both indices are defined as [L-U]=(L-[7)-0.21(7- B) and [B-L] = (B-L)-0A3(V-B). Sketched are a few Teff = const lines for logg in the range 3.5-4.5 for solar composition and a microturbulence of umj = 2kms~1 based on the computations of Lub and Pel (1977). The empirical main sequence is represented as a dashed curve. According to Pel et al. (1981) a slight zero shift had to be applied on the cool part of the theoretical relations, in order to match them with the empirical main sequence. This zero point shift decreases at hotter temperatures, but since we do not know yet exactly how much, we adopted them to be zero for 8000-10,000 K. © European Southern Observatory • Provided by the NASA Astrophysics Data System M. J. J. Wiertz and A. M. van Genderen: VBL UW photometry of Am, 8 Del, and <5Sct stars 39 0.15 0.20 0.15 0.20 [B-L] Fig. 2. The reddening independent two-colour diagram [L— U~\I[B—L~] for the Am stars (on the left) and 5 Del and 5 Set stars (on the right). Sketched are the main sequence (dashed curve) and a number of Teff = const lines (full curves) It can be seen that some overlap exists between the theoretical relations for 7000-8500 K and 8500-10,000 K. The relations for Teff<7000K (not shown) overlaps partly those for Teff>7000K. The benefit of this diagram is that reddening effects caused by interstellar dust are not present. The position of the three 5 Set stars (crosses) are now normal with respect to the other <5 Set stars, not only confirming the reddening estimated by Hall and Mallama (1980) for HD 174553 (V 369 Set), but also confirming our suspicion that HD 195961 fePav) and HD 24550 (V479Tau) must be reddened too by relative large amounts. We estimate for both stars reddenings in the order of E(B— V)j~0.12 + 0.04mag. This reddening is much higher than Davis Philip et al. determined for HD 195961 (ePav) (see Sect. 3). The abnormal Am star HD 170920 (61 Ser) from Fig. 1 is also peculiar here because of its high [L — [/] index (triangle in the left panel of Fig. 2) and possibly indicating a low gravity. One Am star, HD 104671 (#' Cru) (encircled in Fig. 2 in the left panel), shows a rather large deviation with respect to the other Am stars, caused by its low value for [£ — L]. The position in the V— B/B — L diagram of Fig. 1 is then also relatively high above the main sequence (encircled in Fig. 1 left panels). The spectral types assigned to it by means of the strength of the metal lines varies between A 5 and A 8 (see for references the catalogue of Hauck and Curchod, 1980). In the Geneva photometric system the star is striking because of its abnormal low metallicity index (Am2 = — 0.031) compared to its temperature and colour index (B2 — V1 = +0.084). In fact its position at the left of the solar composition grid of Fig. 2 also points into the direction of a metal under-abundance, since an other benifit of this diagram is that [B — L] is a sensitive metal index parameter in contrast with [L— !7] (Pel and Lub, 1978, their Fig. 1). However a quantitative number cannot be given because of the difficulty to obtain accurate physical data of peculiar stars from photometry alone (see Sect. 5). In this context we can also mention the 3 Set stars encircled in the right panel of Fig. 2 and which may be also metal deficient stars: HD6870 (BSTuc), HD9065 (WZScl), HD 32846 (XCae), HD 100363 (SUCrt), and HD 115308 (DKVir). Indeed, Breger (1979) remarks in the "Notes" belonging to his Table 1 that HD6870 and HD 100363 are known to be metal deficient. Thus presumably they are Population II 5 Set stars with high space velocities (Breger, 1979). The metal deficiency of the other three stars is not yet confirmed by other means (they are indicated in Fig. 2 by question marks). 5. Determination of Teft and log<7 We investigated how well Teff and loggi derived from the VBLUW and the uvbyp photometric systems agree with each other. These parameters for the latter system are given for most of the stars by Davis Philip et al. (1976). Since we are dealing with stars with abnormal spectral characteristics, it is also important what Teff and logg are when they are derived by means of spectral energy distribution studies. These studies (Babu and Shylaja, 1981, 1982) are more independent of things like spectral anomalies and small reddenings (but not independent of effects caused by duplicity). For the purpose of Tell and logg determination, Fig. 2 would be most suitable, were it not that overlapping problems of certain temperature regions play a too important role. Therefore we have chosen the V— B/B — U diagram (Fig. 1). The grid of theoretical colours is however only suitable for V—B> 0.045, again because of overlapping problems and the converging of Teff- and logg = const lines at bluer colours. The result is that for the Am stars our values for Teff and \ogg are on the average 100 K and 0.4 lower, respectively, than those of Davis Philip et al. These differences practically disappear for the <5 Del and 5 Set stars. Thus there is a satisfactory agreement between the two methods. However individual differences between both methods may amount to 300 K and 0.8 in logg. As discussed before, blanketing by metallic lines in the (B — V)j index (and thus also in V—B and b — y) cannot be ignored, especially for the hot stars. Indeed, when spectral types are considered, the photometric temperatures appear to be too low by many hundreds of degrees. We found that 13 Am stars from the catalogue of Davis Philip et al. (of which four in common with us) are also studied by Babu and Shylaja. It appears that the temperatures of the latters are indeed hotter than the photometric ones. Figure 3 shows the relation between this difference (Teft(ph)— Teff(BS)) and the true temperature (based on the energy distribution) of Babu and Shylaja Tetf(BS). The right hand scale of Fig. 3 indicates the reddening by blanketing of the V— B (in log int. scale) and (B — V)j colour indices. The size of the blanketing effect and its dependence on temperature is similar to that found by Babu and Shylaja (1982, their Fig. 2). They compared their temperatures and (B— V)j indices with the temperature scale for main sequence stars of Code et al. (1976). Because of the blanketing effect, the photometric gravities are also underestimated by a few tenths in logg. Blanketing effects must be also present in 5 Del stars since they also are metallic line stars. Indeed the 7^ffofthe one specimen we have in common with Babu and Shylaja (HD 207098 = 5 Cap) is estimated too cool by us by 600 K and Davis Philip et al. even found it to be 900 K cooler. © European Southern Observatory • Provided by the NASA Astrophysics Data System 40 I I • Am 1 1 1 1 1 -1200 o 6 Del + 8 Set • • _ -800 0 • • • • -400 • • • 0 • • • + I I 1 1 1 1 1 7000 8000 9000 10000 AIV-B) [AlB-Vljl ' 0.06 I [0.151 0.03 [0.075] Fig. 3. Diagram showing the difference between photometric and energy distribution temperature (A Teff) as a function of the latter (Teff(BS)). The scale on the right roughly indicates the reddening by line blanketing on V— B (in log int. scale) and on (B -mag) V)j (in The one 5 Set star we have in common with Babu and Shylaja (HD 197461 = 5 Del) is estimated 400 K and 100 K hotter by us and Davis Philip et al., respectively, while one should expect an opposite inconsistency. Perhaps this inconsistency is not so surprising since the star is complicated by the presence of a spectroscopically detected companion by Preston (quoted by Kuhi and Danziger, 1967). Both stars are plotted in Fig. 3 (for ATe[f we also took the average between the two photometric methods). 6. A comparison of the colour indices V—B, b—y and B2 — V1 We investigated the relations between the temperature sensitive colour indices V— B of the Walraven system with b — y (Stromgren system) and with B2 — Vl (Geneva system) for normal A/F type stars and for Am, 5 Del, and <5 Set stars. For this purpose we searched the uvbyfl catalogues of Hauck and Mermilliod (1980) and Breger (1979) and the Geneva system catalogues of Rufener (1976) and Hauck and Curchod (1980) for stars in common. The V— B colour indices of the normal main sequence stars were taken from Lub and Pel (1977). Figure 4 shows the b — y/V—B diagram. The curve represents the relation for main sequence stars within the range V—B = 0 — 0.17. The maximum dispersion around this curve (of which the details are not shown) is ±0.01 in V— B. It is obvious that the Am and 5 Del stars at constant b — y are too red by 0.005-0.010 in log intensity scale with respect to normal stars. Apparently the V—B index is more sensitive for line blanketing than the b — y index. This is comprehensible since Aeff(B) < 2eff(b) (4325 and 4670 A, respectively). The Am star HD 170920 (61 Ser) already discussed before, is also in Fig. 4 peculiar (triangle). Giants and supergiants taken from the list of Pel (1976) are indeed situated above this line, thus supporting our suspicion that this star is rather far evolved. (In fact this diagram is nothing more than an ordinary two-colour diagram.) Although most of the Am, <5 Del, and 5 Set stars are supposed to be in a slightly evolved phase, they lie in general below the main sequence. Apparently the spectral anomalies are still stronger than the gravity effects in most of these stars. However many <5Sct stars tend to lie also on or above it, V-B 0.05 - Q10 - 0.15 Fig. 4. The b — y/V— B diagram for Am, S Del, and S Set stars. The full curve is the relation for normal main sequence stars of type A/F V-B 0 0.10 I I • i Am \M.S. o 6 Del \ + 6 Set V i i • \ i -0.1 0.1 B2"V1 Fig. 5. The B2- VJV-B diagram for Am, 5Del, and 5Set stars. The full line is the relation for normal main sequence stars of type A/F indicating less influence by metallicity. The two 8 Set stars with high reddenings (g Pav and V479Tau) and known b — y indices are indicated by large crosses. Their deviating position far below the main sequence relation, may be caused by the fact that high reddenings act differently on the two colour indices. Figure 5 shows the B2— VJV— B diagram. We have fewer stars in common with the Geneva system observers. The range in © European Southern Observatory • Provided by the NASA Astrophysics Data System M. J. J. Wiertz and A. M. van Genderen: VBL UW photometry of Am, 5 Del, and (5 Set stars 41 V— B is then also smaller than in Fig. 4. The relation for main sequence stars has a similar spread as in Fig. 4. At constant B2-VVV-B is redder by 0.005-0.010 in log int. scale, thus also here V—B is apparently more sensitive for line blanketing than B2 — V1 with the same amount as in Fig. 4. Yet the difference in kef! of B and B2 is not so large as between B and b (^(^2) =4480 A). The reason is likely that Aeff(F)>/leff(71) (5467 and 5405 A, respectively). Again the Am star HD 170920 (21 Ser) is situated above the main sequence (triangle). It appears that a linear eye-fitted relation exists between V— B and B2 — V1 for main sequence stars within the range — 0.1 < V— B <0.32 or — 0.3