This twin register was organized as a resource for studies on the eye. It has been possible to share this resource with other disciplines through effective collaboration. Conversely, only through the assistance of other investigators who are working with twins, will it be practical to undertake certain studies which are important in ophthalmology. The purpose of this presentation is to convey a description of our register and some of its limitations, and to specify the need for collaboration with others.

The original purpose in assembling this register was to identify a group of twins readily available for multiple clinical examinations. Since the first examining facility was to be located near the National Institutes of Health, practical considerations required that twin residence be limited to the metropolitan Washington, D. C. area. Twins of all ages were included in the register, since proposed investigations pertained to different age categories. No effort was made to select twins through randomization procedures; all twins who were identified were invited to membership in the register. A variety of methods were used to locate twins, including contact with schools and clubs, but the largest proportion of registrants were recommended by other twins.

Twins were offered a thorough eye examination, along with the solicitation of their interest in participating in future research activities. The clinical examination was offered for the following reasons:

1) To encourage informed interest in proposed future investigations by acquainting the twins with our facility and purpose;

2) To perform base-line observations, so as to identify and enumerate ocular characteristics as an aid in selecting subsets of twins for future studies;

3) To obtain data for possible estimates of heritability of selected ocular characteristics.

The Mendel Institute in Rome houses a Twin Register that was initiated by the senior author even before the opening of the Institute (September 1953).

The purpose of the Twin Register, originally limited to research on twin concordances as to blood glutathione levels (a subject that is still under investigation, as shown by the report presented at this Symposium by Gedda and Tatarelli), was gradually widened to cover the full horizon of twin studies.

The realization that the twin method, in its various formulations, is irreplaceable in most areas of biomedical research, led to the decision to include in the Register each and every contactable twin pair; wide use of mass media was made for recruitment purposes.

Recruitment generally occurred at very early ages, and all pairs are constantly followed. Formerly, definite zygosity determinations were obtained only for those pairs that were selected for specific research purposes. Zygosity determinations are currently being extended to all those pairs who come to the Institute's clinics, within the limits of the available budget (it should be noted that the Institute is a private foundation which provides free medical care to any twin. Lack of capital funds forces the Institute to limit its research activities to those for which grants are obtained).

The Swedish Twin Registry of the Karolinska Institute comprises about 10 000 sets of twins, and was built up in 1961-1962. It covers about 95% of all Swedish same-sexed twins that were born in the country in 1886-192 5 and were still living as unbroken pairs at the time of compilation.

The compilation procedure has been presented in detail by Cederlöf (1966a), and will be summarized as follows. Since the middle of the last century, a rather unique system for population registration has existed in Sweden. It is based not only on birth records, but also on migration records. As soon as an individual moves away from the place he lives, he has to notify the authorities as to where he moves. This system made it possible for us to follow-up any individual from his place of birth to his next place of living, and so on, until we found his current address or had to discard him as deceased. We started, in that way, with about 41 000 multiple birth records, handscreened from about 3½ million birth records from all Sweden during the period in question, and ended with a series of about 12 000 located sets of twins. Of the remainder, about 26 000 had to be discarded, due to death of one or both partners; 1 000, due to emigration to foreign countries; and 100, due to hospitalization in mental asylums, or home for aged people, etc. In the compilation procedure we lost track of about 1 100 individuals, or 1.5%.

The zygosity of the twins has been determined with the aid of some similarity questions contained in a questionnaire, mailed to all the twins in the registry. Out of several questions, one proved to be of high reliability, namely the very simple question whether the twins as children were as alike as two peas in a pod, or of family likeness only. We assumed that if both twins in a set agreed in saying that they had been as alike as two peas in a pod, the pair was MZ. If they, on the other hand, agreed in saying that they had been of family likeness only, we regarded the pair as DZ. We validated this questionnaire diagnosis by analyzing the outcome from a subsample of 200 pairs who had also been examined serologically in regard to five markers, A1A2BO, MN, Rh, Hp, and Gm. The results were reported here in Rome in 1961 and showed an agreement in diagnoses of 95% for MZ and 90% for DZ (Cederlöf et al, 1061).

A systematic register of twin births, covering the southernmost province of Sweden from 1870 onwards, was erected at the Psychiatric Department of the University of Lund in the late forties; it is thus the oldest of the Scandinavian twin registers. The register contains 20 000-25 000 twin pairs, unselected for survival. By confrontation with some 15 000-20 000 consecutive Cases of acute concussion of the brain, from surgical wards, and a similar number of alcoholics, and of patients with peptic ulcer, a number of MZ pairs discordant for each of these diagnoses were extracted and thoroughly examined in a psychiatric and psychological study of nervous sequels, organic deterioration, and the influence of stress.