Zuckerman, UCLA. We will begin with George Wallerstein. Wallerstein: The commonly accepted doctrine (fortunately not yet dogma) is that cool stars evolve along the AGB through a sequence of types M-MS-S-SC-C. Along this sequence the abundance of C increases by the mixing of helium-burning products to the stellar surface. At type SC the C/0 ratio passes through unity. While this sort of sequence may well be followed by some stars I would like to point out some apparent inconsistencies as

more »

... guidance for both discussion here and future research. 1. Among the carbon stars there are two groups that do not fit the sequence -these are the early R stars (Dominy 1984) and the C-rich carbon stars (often called type J) which constitute 13% of the sample of Lambert §_t al. (1986). The former do not fit because their luminosities are too low for the AGB, while the latter have converted C to C via hydrogen burning, presumably in a shell. Furthermore, the J stars often do not show s-process enhancements, thus indicating that they have not available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0252921100063454 Downloaded from https://www.cambridge.org/core. IP address: 207.241.231.81, on 01 May 2019 at 14:05:15, subject to the Cambridge Core terms of use, 408 Outstanding Problems in Research on PRGs passed through the S star phase but rather evolved directly from M to C. 2. The oxygen isotope ratios of the SC stars do not fit the sequence. Their 16 0/ i7 0 ratios range from 75 to 1200 while the range for MS and S stars is 550 to 3000, and the range for the low-13 C carbon stars is 550 to 4000. Surprisingly, the 16 0/ i7 0 range for Barium stars is 100 to 500, much like the SC stars. Furthermore the Ba stars have enhanced carbon, though with carbon still less than oxygen, and the coolest Ba star (HD 121447) has also been classified as the hottest SC star. Are the SC stars binaries that are about to dump on their companions? If so, we are observing them at a remarkable time because several SC stars show technetium, which indicates that they are still in the shell flashing stage. In any case, a search for binarism in SC stars would be worthwhile (Wallerstein 1988 and references therein). A simple evolutionary sequence of M-MS-S-SC-C along the AGB should be reflected in the periods of the long period variables (LPV's) of those spectral types. However, if you eliminate the M-type LPV's of short period, high velocity, and small mass, there is no such correlation except for a small tendency for the C-type LPV's to have periods near 450 days. In fact, the discovery of the OH/IR stars shows that oxygen-rich LPV's are found with periods as great as 1000 days. They were not previously known because they are hidden in their own dust. Thus many AGB stars evolve to be stars of extreme radius without becoming carbon stars. Perhaps they are of higher mass than those which become carbon stars. With these anomalies in mind, as well as others that have been discussed (or presented in poster papers) at this meeting, we should conclude that evolutionary sequences certainly depend on initial mass and are likely to be affected by both mass loss and mixing. While mass loss available at https://www.cambridge.org/core/terms. https://doi.