Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://star.arm.ac.uk/~csj/papers/newsletter/ccp7n21_jeffery1.ps Äàòà èçìåíåíèÿ: Tue Oct 30 11:41:40 2001 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 11:22:44 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: binary star

Hydrogen­deficient stars and stellar systems
C. Simon Jeffery
Dept of Physics and Astronomy, University of St Andrews, Fife KY16 9SS, Scotland (email: csj @ st­and.ac.uk).
Abstract. Hydrogen­deficiency characterises many stars in
the late stages of evolution, with hydrogen abundances in the
range n H ~10 --5 --10 --1 . The history of the study of these objects
is introduced, and the properties of most classes of low­mass
star with hydrogen abundances in this range are summarised.
For orientation, the locations of all analysed objects are
compared in the g--T eff diagram. The classes may be divided
into three main groups, each with several subdivisions:
i. carbon­normal: HdB, HdN, HesdB, HesdO -- , O(He), DO,
DB, AM Cvn
ii. carbon­rich: HdC, RCrB, EHe, hot RCrB, HesdO +
iii. carbon­strong: [WC­L], [WC­E}, O(C), PG1159
Some conclusions regarding evolutionary connections are
drawn, and an attempt made to identify key questions still to
be solved.
1. Introduction
The first international conference on hydrogen­deficient (H­
def) stars was held in Mysore, India, in November 1985
(Hunger et al. 1986). The absence of hydrogen, normally a
dominant source of astrophysical opacity, has severe
implications for the physics and structure of certain stars,
and challenges the theory of stellar evolution. The necessity
of developing innovative models for the interpretation of H­
def star observations has led to the close involvement of
`Collaborative Computational Project No. 7' (CCP7). To
review this association, CCP7 held a workshop entitled
`Hydrogen­deficient Stellar Systems' at the University of St
Andrews in March 1994.
Since the discovery of the first H­def stars, the number
and variety of known astronomical objects which have lost
nearly all of their original hydrogen has increased
considerably, not least since the Mysore conference. In order
to establish a framework for the CCP7 workshop and its
proceedings, the author was asked to review this `zoo' of H­
def stars. In doing so, he has drawn heavily on the work of
others, particularly Bidelman's (1986) exemplary history
and, in order to maintain contemporaneity, the contributions
of delegates. For brevity, many important results have been
neglected; it is hoped that this personal view will be extended
in due course.
One consequence of the rapid growth and comparitive
infancy of the subject is that nomenclature is often imprecise,
and sometimes misleading. In navigating the `zoo', the
author has drawn on several sources to refine current usage.
Until such time as a definitive classification of H­def stars is
established, it is intended that terms introduced herein will
help to resolve some confusions.
2. History
On May 28, 1795, a Mr Pigott of Bath, England, was unable
to detect a star in the Northern Crown which he had
previously measured to be of the 6th or 7th magnitude (Pigott
1797). Subsequent observations during that summer
demonstrated its slow reappearance until, in September, it
disappeared again from view, becoming fainter than 11th
magnitude. This object, now known as the variable star R
CrB, had become the first H­def star to attract attention. The
interpretation prevalent in that day for stellar variability was
rotation, by analogy with the Sun. Pigott deduced a rotation
period of 10.5 months for RCrB but added ``Future
observations will determine how far [it] is rightly settled''.
110 years later, Ludendorff (1908) compiled a remarkable
light curve for R CrB, spanning 62 years and 35 observers.
This clearly demonstrates the rapid decline and slow rise to
maximum light at irregular intervals, now well recognised
as the `RCrB phenomenon'. At the same time, evidence for
the unusual composition of some stars was beginning to
emerge (Fleming 1891, Ludendorff 1906, Cannon 1912,
Rufus 1915). However, the reluctance of astronomers to
admit exceptions to their ``hydrogenic'' view of the universe
became a dominant influence (Payne 1925, Joy & Humason
1923, Plaskett 1927). It was only when irrefutable evidence
had been established (Berman 1935, Struve & Sherman
1940, Greenstein 1940) that astronomers would accept ``that,
somehow, a very substantial amount of hydrogen had been
lost in a few exceptional stars''.
Over the subsequent 40 years, a steady stream of H­def
stars were discovered during ground­based low­dispersion
spectroscopic surveys and photographic patrols; these are
referred to here as the `classical' H­def stars (e.g. RCrB stars,
extreme helium stars). With access to UV and X­ray
telescopes, and with the advent of modern model atmosphere
techniques, there has been an explosion in the number of
objects known to be extremely H­deficient; the recent
discoveries constitute the `new' H­deficient stars (e.g.
PG1159 stars, DO white dwarfs). In the main, these objects
are enriched in carbon, but a few are not; these are
considered here as `carbon­normal' stars, although possible
connections with the carbon­rich stars should not be
overlooked. Apart from these objects, which are generally of
low­mass, are a number of young massive stars in which
magnetic fields (intermediate helium stars), strong stellar

C.S.Jeffery: Hydrogen­deficient stars
winds (Wolf­Rayets) or mass transfer (Algols) have modified
the surface composition. These were not considered further
in the workshop.
3. The classical H­def stars
In a survey of carbon stars, Bidelman (1953) found that the
majority were `normal', but that a small number were H­
deficient. Amongst these H­def carbon stars, there is a
spectroscopic distinction between those characterised by
weak Balmer and strong C I absorption (cf. R CrB) and those
with strong CN and C 2 molecular bands (cf. S Aps). There is
also a photometric distinction between those which show
large amplitude (>5 mags) variations and a number which do
not. The former are well known as RCrB stars, the latter
have become generally known as `hydrogen­deficient carbon
stars' (HdC), although Warner's (1967) original use of the
term was intended to apply to all H­def objects with excess
carbon. Spectroscopic analyses for T eff , g, and abundances
have been carried out for a number of these stars (Warner
1967, SchÆnberner 1975, Cottrell & Lambert 1982 -- CL82,
Lambert & Rao 1994, preprint), their locations in the g--T eff
diagram are shown in Fig. 1 (Warner's measurements have
been normalised so that R CrB matches CL82) and their
abundances are summarized in Table 1. Pressing
requirements include reliable measurements of the surface
abundances (especially He, C) and isotope ratios (C 12 /C 13 ).
Line­blanketed model atmospheres for RCrB mixtures are
urgently required.
Table 1. Average surface properties of hydrogen­deficient stars, Abundances are given as log n x /n tot , all at least ±0.3 dex; the spread is
generally larger. Exceptional cases are noted in parenthesis.
Class T eff /kK log g n H n He n C n N n O n Fe Refs
Carbon­rich
HdC 6 --1...0 --5 0 --2 --4 15
RCrB 6...8 0 --4 0 --1.5 --2.5 --2.0 --4 2
EHe 10...30 1...3 --3 0 --2.1 --3.1 --3.1 --4.6 7,8
hot RCrB 15...20 2...3 --1 0 --2.0 (--4 1 ) --3.5 --2 10,12
HesdO + 40...80 4...5 --1 0 --1.7 (--4 2 ) --2.3 6
Carbon­strong
[WC--L] 20...46 2 --0.15 --0.59 --1.5 13
[WC--E] 100...120 2...6 --0.15 --0.67 --1.9 --1.3 13
PG1159 65...170 5...8
(
--0.19
<--1.5 3
--0.52
--0.24 3
--1.3
--0.4 3 )
17
Carbon­normal?
HdB 10...14 1 --4 0 --4 --3 --3 --4 4
HdN 20...25 3...4 --2...--1 0 --1.5 --3 9,11
HesdB 28...31 5 --2.5 0 --2 5
HesdO -- 45 5.5 0 --2.3 --3.1 --3.0 3
O(He) 100 6 --0.7 0 <--2.3 --2.3 14
DO 45...120 7...8 --1 0 --2...--5 ­4 <--4 18,19,20
DB 11...28 8 <--4.5 0 <--5 4 16
Sun 0 --1 --3.4 --4.0 --3.1 --4.5 1
The large­amplitude variations of RCrB stars have
intrigued astronomers for nearly 200 years; Loreta (1934)
and O'Keefe (1939) were the first to propose that they were
due to an `eclipse' of the photosphere by dense clouds of
carbon­rich dust ­ or soot. Subsequent observations have
demonstrated the association of these dust `ejections' with
photometric variations of amplitude ~0.1 mag and period
~40 days due to radial pulsations (Alexander et al. 1972). A
strong IR­excess (Kilkenny & Whittet 1984 -- KW84, Walker
1986) due to a cool dust shell (500 -- 1000K) some distance
from the star reflects the photospheric pulsations even when
the line­of­sight is obscured by dust (Feast 1986). The RCrB
stars are also characterised by both broad and narrow
emission­line regions (e.g. Querci & Querci 1978, Rao &
Lambert 1993 RL93), and by a considerable range of H
abundances (cf. CL82, RL93). Evidence for a very low
surface­brightness reflection nebula around some stars is
emerging (Pollacco et al. 1991). Present questions concern
the composition of the dust (molecular or graphitic), the
ejection mechanism and location of formation, and the
geometry of the whole star/dust/nebula complex.
The second classical group of H­def stars includes the B­
and A­type `helium' stars which, at low dispersion, show no
Balmer lines and strong He I lines (cf. HD124448, Popper
1942). Here again there is spectral diversity, with the main
group showing strong C II (especially l426.7nm) absorption.
Individual `extreme helium stars' (EHe) show a range of
properties, including small amplitude variations attributed to
radial (Saio & Jeffery 1988) and non­radial (Jeffery et al.
1985) pulsations, RCrB­type variations and IR­excess dust
shells (hot RCrB, Hoffleit 1930, KW 1984). The most
luminous objects exhibit a permitted emission­line `shell'

C.S.Jeffery: Hydrogen­deficient stars
spectrum (Jeffery & Heber 1992,1993) or a strong emission­
line `wind' spectrum (late­type WC stars ­ [WC­L]; Webster
& Glass 1974). Their surface properties are reviewed by
Jeffery (1992) and Leuenhagen (1994), their g--T eff locations
are shown in Fig. 1. Of interest is the discrepancy in carbon
abundance between the EHes/hot RCrBs (n c ~0.01) and the
[WC­L]s (n c ~0.25) (Table 1).
4. The new H­def stars
The advent of large­scale surveys of UV­excess stars and
access to ultraviolet and X­ray wavelengths from space
platforms during the last 30 years has led to the discovery of
new classes of hot star. Because of their high T eff , hydrogen
is not readily detected in the optical spectrum. Therefore
detailed studies using model atmosphere techniques have
been crucial in demonstrating that many of these are
hydrogen­deficient.
The subdwarf O stars had been known for some years
(Greenstein & Sargent 1974) but the spectroscopic
classification was too coarse to discriminate the chemical
deviations present in many objects. Most of the helium­rich
subdwarf O stars (HesdO) were discovered during Drilling's
(1983) low­dispersion surveys of OB + stars and are
characterised by strong He II absorption, the He II Pickering
series being so strong as to prevent detection of the Balmer
lines. One group, analysed by Husfeld et al. (1989), indicate
L/M ratios and abundances similar to the EHes, and are
referred to here as the luminous HesdOs, or HesdO + . A
second group shows, in addition, He I absorption and has
lower T eff and L/M (Dreizler 1993). These stars are probably
evolved helium main­sequence objects (cf. sdOB stars, Heber
1986). They are referred to here as compact HesdOs, or
HesdO -- , but should properly be considered amongst the
carbon­normal stars discussed below. The properties of both
groups are indicated in Fig. 1 and Table 1.
The discovery of multiperiodic low­amplitude light
variations in the very hottest white dwarfs (eg PG1159--035;
McGraw et al. 1979); and their potential use for
asteroseismolgy (Kawaler et al. 1985) provoked considerable
effort by both photometrists (WET, Winget et al. 1991) and
spectroscopists (Wesemael et al. 1985, Werner et al. 1991).
Whilst some PG1159 stars are surrounded by planetary
nebulae, and some show multiperiodic pulsations, not all
properties are shared -- even by spectroscopic twins (cf.
PG1159--035 & PG1520+525). Very high effective
temperatures lead to extreme non­LTE conditions in the
atmospheres, where opacity is provided by very highly­
ionized species (He II, C IV, N V, O VI, ...). Their T eff --g
locations (Werner 1993, Fig. 1) place the PG1159s around
the `knee' of the evolution tracks of low­mass stellar
remnants contracting onto the white­dwarf sequence, whilst
their surface abundances (n He ~0.7, n C ~0.3, Table 1) suggest
substantial mixing between a remnant He envelope and C
core. One object (H1504+65) appears to be a remnant CO
core, having lost its entire H+He envelope (Werner 1991).
Extreme hydrogen­deficiency is also observed amongst
the white dwarfs, notably the DO (He II) and DB (He I)
sequences. The DOs may be related to the PG1159 stars
having lower L/M and T eff (Fig. 1), but similar H/He ratios
(cf. Werner et al. 1994). The discrepancy in carbon
abundance may be due to gravitational settling/levitation
(Werner et al. 1994). Whilst DA (hydrogen­rich) white
dwarfs are found at all T eff (100,000--15,000 K), there is a
remarkable gap in T eff between the DOs (100,000--45,000 K)
and DBs (25,000--12,000 K). The higher frequency of DBs
than DOs relative to DAs in comparable T eff intervals
(Liebert 1986) suggests that the H­deficiency observed in
DBs is not entirely due to the effects of stellar evolution, but
to other processes in the stellar atmosphere. Considerable
uncertainty persists regarding the evolution (DO®DA?,
DA®DB?) and the origin of the T eff gap.
The vast majority of planetary nebulae (PN) and their
central stars have normal hydrogen abundances, the
canonical view being that nebulae are ejected from the
surfaces of very luminous stars as they leave the asymptotic
giant branch and contract to become DA white dwarfs.
Evidence that some central stars are hydrogen­deficient is
provided by their carbon­rich Wolf­Rayet spectra (e.g. NGC
40, Smith & Aller 1969). A considerable number are now
known, mostly with WC emission­line spectra (cf. Mendez et
al. 1986), and may be divided into late (WC­types 8--10,
[WC­L], see above) and early (WC­types 2--4, [WC­E])
types. In addition are two groups with absorption­line spectra
(O(He) and O(C)) and one with mixed absorption and
emission lines, the Of­WR(C) stars Abell 30 and Abell 78.
The O(C) stars appear to be spectroscopically identical to the
PG1159 stars; the results of recent analyses (Leuenhagen
1994) are shown in Fig. 1, assuming M H =0.6M O . In contrast,
the O(He) stars appear to be carbon­normal (n C <0.005,
Rauch et al. 1994), prompting speculation about their
evolutionary status and relation to the otherwise similar
PG1159 stars.
Spectral variability seems to characterise the PN central
stars; one (Longmore 4) was observed to change spectral type
from PG1159 to [WC­E] and back within a few weeks
(Werner et al. 1992). The WR(C) central star of Abell 58
(=V605 Aql) underwent a nova­like event in 1919 (Bidelman
1971), having previously resembled the HdC star HD182040
(Lundmark 1921). The intriguing central star Henize 1--5
(=FG Sge) brightened by 4 magnitudes between 1894 and
1960 and changed spectral type from B4 I to F7 I/K2 Ib
between 1955 and 1968 (Herbig & Boyarchuk 1968, Kipper
& Kipper 1989). Latterly, it declined and bluened rapidly
(Jurcsik 1992). Although clearly carbon­rich, it is not yet
clear whether FG Sge is hydrogen­deficient and whether its
behaviour is associated with an RCrB­like event (Wallerstein
1990) or the helium­shell flash of a normal PN central star
(Herbig & Boyarchuk 1968).
The nebulae around the Of­WR(C) stars present a
dramatic picture of the evolution of a H­def central star.
Whilst the outermost (oldest) regions of the nebulae have
normal hydrogen abundances; knots close to the central star
are substantially hydrogen­defiicient (Hazard et al. 1980,
Manchado et al. 1988, Seitter 1987). High spatial resolution
spectroscopy and images demonstrate jet­like structures in

C.S.Jeffery: Hydrogen­deficient stars
the central regions (Pollacco et al. 1992, Clegg, 1994 ­ this
workshop) indicative of recent mass­loss events. Are the
central regions of other PN with WC, O(He) or O(C) central
stars also hydrogen­deficient? What is the mechanism
responsible for the knotty or jet structure observed, and is the
star able to lose sufficient mass by this process to account for
the hydrogen­deficiency? How was the carbon enriched?
5. The carbon­normal H­def stars
Carbon­normal H­def stars include the well known binary
star u Sgr (Campbell 1899, Greenstein 1940), prototype of
the small class of H­def binaries (HdB) with periods of
between 25 and 360 days. They show substantial mass
transfer effects (Nariai 1967,1970) and radial pulsations
(Malcolm & Bell 1986) and are probably population I stars
undergoing a second­stage of mass transfer (case BB;
SchÆnberner & Drilling 1983). Measurement of the u Sgr
secondary (Dudley & Jeffery 1990) indicates that the primary
is ~3M O and likely to become a Type Ib supernova (cf.
Uomoto 1986).
The (single) stars HD144941 and V652 Her may
constitute a class of H­def nitrogen­rich star (HdN), although
the status of the former is uncertain. V652 Her shows well­
defined radial pulsations with a 2.5hr period; analysis of
these has yielded a distance­independent mass of 0.7 M O
(Lynas­Gray et al. 1984).
A group of stars classified sdO(D) in the Palomar­Green
survey (Green et al. 1986), showing strong HeI absorption
but weak or absent Balmer lines, has received little attention.
Also called helium­rich subdwarf B stars (HesdB, Moehler
et al. 1990), they appear to lie above the extended horizontal
branch (Heber et al. 1988).
Finally, a small group of stars with DB­like spectra show
short­period variations. These AM Cvn systems show
phenomena asoociated with accreting binaries (cataclysmic
variables), but despite 40 years of effort, firm orbital periods
and other basic data remain elusive (Marsh 1994).
6. Evolution
Recent analyses indicate that the H­def stars may be divided
into three main groups dependent on carbon abundance. The
carbon­normal stars belong to several different classes with
widely different characteristics, and include HdBs, HdNs,
HesdBs and HesdO -- s. There is at present little evidence to
connect any of these groups. The HdCs, RCrBs, hot RCrBs,
EHes and HesdO + s share a common carbon­rich abundance;
typically (with variations) n He ~0.98, n C ~0.02, n N ~0.002,
n O ~0.002 (Table 1), indicative of a common evolutionary
track
HdC/RCrB®EHe/hot RCrB®HesdO + .
Similarly, WC stars and PG1159s share a common
abundance pattern; typically n He ~0.7, n C ~0.3, n O ~0.05 (Table
1), indicative of a common evolutionary track. If
gravitational levitation is resposible for the submersion of
surface carbon in DOs, the sequence may be
[WC­L]®[WC­E]®PG1159®DO.
That there are substantial abundance variations within all
classes indicates that the mechanism for injecting stars onto
either sequence must mix the stellar material heavily, and
contain at least one free parameter in addition to mass and
original composition. The question really arises as to
whether the two sequences are really distinct (where are the
EHe descendants at the `knee' ?). The variable star V348 Sgr
has been variously classified as a hot RCrB star, a WC11
central star, and an extreme helium star. Leuenhagen &
Hamman (1994) measured n C ~0.25, but for the hot RCrB
star DY Cen, n C ~0.01 (Jeffery & Heber 1993). What is the
origin of this discrepancy? The measurement of additional
surface abundances, e.g. neon, and the refinement of all
existing model atmosphere methods will have a significant
impact on the interpretation of H­def star evolution.
Acknowledgements. I am grateful to Klaus Werner and Phil Hill for valuable
advice during the preparation of this review.
References
Alexander J.B. et al. 1972: MNRAS 158, 305
Anders E., Grevesse N. 1989: Geochim.Cosmochim.Acta 53, 197
Berman L. 1935: ApJ 81, 369
Bidelman W.P. 1953: ApJ 117, 25
Bidelman W.P. 1971: ApJ 165, L7
Bidelman W.P. 1986: IAU Coll. 87, p. 3
Campbell W.W. 1899; ApJ 10, 241
Cannon A.J. 1912: Harvard Ann. 56, 107
Cottrell P.L., Lambert D.J. 1982: ApJ 261, 595: CL82
Dreizler S. 1993: AA 273, 212
Dreizler S., Werner K., Jordan S., Hagen H. 1994: AA 286, 463
Drilling J.S. 1983: ApJ 270, L13
Dudley R.E. 1992: Ph.D. Thesis, St Andrews University
Dudley R.E., Jeffery C.S. 1990: MNRAS 247, 400
Feast M.W. 1986: IAU Coll. 87,151
Fleming W.P. 1891: Astr.Nach. 126, 165
Green R.F., Schmidt M., Liebert J. 1986: ApJS 61, 305
Greenstein J.L. 1940: ApJ 91, 438
Greenstein J.L. & Sargent A. 1974: ApJS 28, 157
Hazard C., Terlevich R., Norton D.C., Sargent W.L.W., Ferland G.
1980: Nature 285, 463
Heber U. 1986: AA 155, 33
Heber U., Dreizler A., de Boer K.S., Moehler S., Richtler T. 1988:
Astr. Ges. Abs. Ser 1, 16
Herbig G.H., Boyarchuk A.A. 1968: ApJ 153, 397
Hoffleit D. 1930: Harvard Bull 874, 1
Hunger K., Rao N.K., SchÆnberner D. (editors) 1986. Hydrogen­
deficient stars and related objects (IAU Coll. 87), Reidel,
Dordrecht, Holland.
Husfeld D., Butler K., Heber U., Drilling J.S. 1989: AA 222, 150
Jeffery C.S. 1992: `The Atmospheres of Early­Type Stars', p. 293,
eds. U.Heber & C.S.Jeffery, Springer­Verlag, Heidelberg,
Germany
Jeffery C.S., 1993: AA 279, 188.
Jeffery C.S., Heber,U., 1992: AA 260, 133
Jeffery C.S., Heber U., 1993: AA 270, 167
Jeffery C.S., Heber U., Hill P.W., 1986: IAU Coll. 87, p. 101
Jeffery C.S., Heber U., Hill P.W., Pollacco D. 1988: MNRAS 231,
175

C.S.Jeffery: Hydrogen­deficient stars
Jeffery C.S., Skillen I., Hill P.W., Kilkenny D., Malaney R.A.,
Morrison K., 1985: MNRAS 217, 701
Joy A.H., Humason M.L. 1923: PASP 35, 327
Jurcsik J. 1992: IBVS 3775
Kawaler S.D., Hansen C.J., Winget D.E. 1985: ApJ 295, 547
Kilkenny D., Whittet D.C.B. 1984: MNRAS 208, 25: KW84
Kipper T., Kipper M., 1989: IAU Coll 106, 146
Leuenhagen U. 1994: CCP7 Newsletter No. 21
Leuenhagen U., Hamman 1994: AA 283, 567
Liebert J. 1986: IAU Coll 87, 367
Loreta E. 1934: Astr.Nachr. 254, 151
Ludendorff H. 1906. Astr.Nachr. 173, 3
Ludendorff H. 1908. Publ.Astr.Obs.Potsdam 57, 1
Lundmark K. 1921: PASP 33, 314
Lynas­Gray A.E., SchÆnberner D., Hill P.W., Heber U. 1984:
MNRAS 209, 387.
McGraw J.T., Starrfield S.G., Liebert J., Green R.F. 1979: IAU
Coll. 53, 377
Malcolm G., Bell S.A. 1986: MNRAS 222, 543
Manchado A., Pottasch S.R., Mampaso A. 1988: AA 191, 128
Marsh T.R. 1994: CCP7 Newsletter No. 21
MÈndez R.H., Miguel C.H., Heber U., Kudritzki R.P. 1986: IAU
Coll. 87, 323
Moehler S. Richtler T., de Boer K.S., Dettmar R.J., Heber U. 1990:
AAS, 53
Nariai K. 1967: PASJ 19, 564
Nariai K. 1970: PASJ 22, 559
O'Keefe J.A. 1939: ApJ 90, 294
Payne C.H. 1925: Stellar Atmospheres p. 189 (Cambridge, Mass.)
Pigott E. 1797: Phil.Trans.Roy.Soc.Lon. 1, 133
Plaskett J.S. 1927: Pub.Dom.Astrophys.Obs. 4, 1
Pollacco D., Hill P.W. Houziaux L. Manfroid J. 1991: MNRAS
248, 1P
Pollacco D., Lawson W.A., Clegg R.E.S., Hill P.W. 1992: MNRAS
257, 33P
Popper D.M. 1942: PASP 54, 160
Querci M. & Querci F. 1978: AA 70, L45
Rauch T., KÆppen J., Werner K., 1994: AA 286, 543
Rao N.K., Lambert D.L. 1993: AJ 105, 1915: RL93
Rufus W.C. 1915: Pub.Michigan Obs 2, 103
Saio H., Jeffery C.S., 1988: ApJ 328, 714
SchÆnberner D. 1975: AA 44, 381
SchÆnberner D., Drilling J.S. 1983: ApJ 268, 225
Seitter W.C. 1987: ESO Messenger 50, 14
Smith L.F. & Aller L.H. 1969: ApJ 157, 1245
Struve O., Sherman F., 1940: ApJ 91, 428
Uomoto A. 1986: ApJ 310, L35
Walker H., 1986: IAU Coll. 87, p.407
Wallerstein G. 1990: ApJS 74, 755
Warner B. 1967: MNRAS 137, 119
Webster B.L., Glasse I.S. 1974: MNRAS 166, 491
Wegner G., Nelan E.P. 1987: ApJ 319, 916
Wesemael F, Green R.F., Liebert J. 1985: ApJS 58, 379
Werner K. 1991: AA 251, 147
Werner K. 1993: `White Dwarfs: Advances in Observation and
Theory', p.67, ed. M.A.Barstow, Kluwer, Dordrecht, Holland.
Werner K., Heber U., Hunger K. 1991: AA, 244 437
Werner K., Hamann W.­R., Heber U., Napiwotski R., Rauch T.,
Wesselowski U., 1992: AA 259, L69
Werner K., Heber U., Fleming T.A. 1994a: AA 284, 907
Werner K., Dreizler S., Wolff B. 1994b: AA submitted
Winget D.E. et al. 1991: ApJ 378, 326

C.S.Jeffery: Hydrogen­deficient stars
Figure 1. The loci of hydrogen­deficient stars in effective temperature and surface gravity. The loci of hydrogen­burning main­sequence
stars (H­MS), the main­sequence for pure helium stars (He­MS), and the Eddington luminosity limit for radiative stability are also shown.
The stellar classes, with prototypes, are as follows:
Low­mass hydrogen­deficient stars Prototype
C­rich HdC u H­def carbon stars; late­type giants HD 182040
RCrB ª R Coronae Borealis variables; F supergiants often obscured R CrB
EHe n Extreme helium stars; H­def B,A supergiants, strong carbon BD+10 2179
hot RCrB H R Coronae Borealis variables with HeI, CII absorption MV Sgr
HesdO +
l Luminous helium­rich subdwarf O stars LSE 153
C­strong [WC­L] ¬ late­type C­rich Wolf­Rayet PN central stars (WC8­10) CPD­56 8032
[WC­E] X early­type C­rich Wolf­Rayet PN central stars (WC2­4) NGC6751
Of­WC X H­def PN central stars with emission and absorption lines Abell 30
O(C) H­def PN central stars with CIV absorption lines NGC246
PG1159 s Extremely hot sdO stars, T eff >100,000K; ºO(C) without PN PG1159­035
C­normal HdB ¡ Hydrogen­deficient binary stars; N­rich, C­poor u Sgr
HdN o Hydrogen­deficient nitrogen­rich subdwarfs BD+13 3224
HesdB ² Helium­rich subdwarf B stars PG1544+487
HesdO -- I Compact helium­rich subdwarf O stars LS IV+10 9
O(He) S H­def PN central stars with HeII absorption lines K1­27
DO D Extremely hot white dwarfs; mainly HeII absorption PG1034+001
DB ÿ White dwarfs showing only HeI absorption GD358
AM CVn Interacting He­rich white dwarf binaries (?) AM CVn