micus superficialis or buccalis, 4 cases with nervus mentalis, cut and labelled) show projections into two long ventral and one short dorsal fascicle (Fig. 1 b, e). The relative position of the long ventral fascicles varies for different nerves: projections from the lower jaw (nervus mentalis) consist of more dor- somedial and ventrolateral fascicles; the projections of the upper jaw (ner- vus ophthalmicus superficialis and buc- calis) are in more lateral and separated fascicles (Fig. 1 b, e). In contrast to the head nerves, in the trunk lateral-line nerve (7 cases) labelling is found only in two long separate fascicles (Fig. I f). Analysis of the sectioned material shows that the short dorsal projection courses along a distinct group of peri- ventricular cells. The long ventral fasci- cles of the head and the trunk run alongside a group of cells which are in a subpial position posterior to the VIIIth nerve root. Comparison with the rhombencephalic alar plate of other anamniotic verte- brates indicates that the dorsal periven- tricular cell group corresponds with re- spect to position, length, and afferent supply to the nucleus dorsalis. The group of subpially located cells is simi- lar to the nucleus intermedius [5, 12] with respect to position, extension, and afferent supply. Consequently, gymno- phionans are the only amphibian order with nucleus intermedius cells located subpially. Earlier reports [7] seemingly have taken the nucleus intermedius, perhaps because of its unusual posi- tion, for a part of the vestibular nucleus ventralis, and have consequently mis- taken the nucleus dorsalis for the nu- cleus intermedius. In summary, our data show that larval Ichthyophis kohtaoensis possess two types of organs neuromasts and am- pullary organs - which have distinctive anatomical features. They also possess a characteristic pattern of lateral-line afferents and cell masses in the rhomb- encephalic alar plate like those of uro- deles [8], sturgeons [13], lungfish [14], cartilaginous fish [15] and lampreys [16, 17]. We suggest that in Ichthyophis single afferent fibers of ampullary or- gans enter the brain via the dorsal root and terminate in the dorsal nucleus as has been demonstrated unequivocally in urodeles [8]. The lack of a dorsal nucleus projection from the trunk lat- eral-line nerve implies that there are no ampullary organs on the trunk of Ich- thyophis. This in fact is also the case in urodeles [8] and many other electrore- ceptive vertebrates [1]. Thus, 3 out of 5 criteria favour the existence of electro- reception in larval Ichthyophis, namely a distinct receptor, a distinct afferent supply, and a brain area specialized in receiving these afferents. In addition, electrophysiological data on adults of other gymnophionan species (Thy- phloneetes) show a sensitivity within the range of other electroreceptors (Bra- ford and McCormick, pers. comm.). Unlike almost all anuran larvae, but like urodeles, larval Ichthyophis koh- taoensis are predators which feed on daphnia, insect larvae and tadpoles [18]. Like urodeles [19], they may use electroreception to detect and localize larger prey. We wish to thank Prof. P. Grrner and Dr. H. Mfinz for their helpful sugges- tions on an earlier draft of this paper and Dr. D. Forsythe for correcting the English. This work was supported by the Deutsche Forschungsgemeinschaft (DFG, Fr 572/1-4 and SFB 45 D 1). Received October 16, 1984 1. Bullock, T.H., Bodznick, D.A., North- cutt, R.G. : Brain Res. Rev. 6, 25 (1983) 2. Bodznick, D., Preston, D.G.: J. Comp. Physiol. 152, 109 (1983) 3. Miinz, H., Claas, B., Fritzsch, B. : ibid. 154, 33 (1984) 4. Bodznick, D.A., Northcutt, R.G.: Science 212, 465 (1981) 5. Fritzsch, B., Nikundiwe, A.M., Will, U. : J. Comp. Neurol. 229, 45 (1984) 6. Hetherington, T.E., Wake, M.H.: Zoo- morphology 93, 209 (1979) 7. Norris, H.W., Hughes, S.P.: J. Mor- phol. 31,489 (1919) 8. Fritzsch, B.: Cell Tiss. Res. 218, 581 (1981) 9. Fritzsch, B., Wahnschaffe, U. : ibid. 229, 483 (1983) 10. Flock, A., J6rgensen, J.M.: ibid. 152, 283 (1974) 11. J6rgensen, J.M.: Acta Zool. Stockholm 63, 211 (1982) 12. McCormick, C.A. : J. Morphol. 171,159 (1982) 13. New, J.G., Northcutt, R.G.: J. Comp. Neurol. 225, 129 (1984) 14. Northcutt, R.G.: Soc. Neurosci. Abstr. 9, 1167 (1983) 15. Koester, D.M.: J. Comp. Neurol. 221, 199 (1983) 16. Ronan, M.C., Northcutt, R.G.: Soc. Neurosci. Abstr. 8, 764 (1982) 17. Fritzsch, B., et al. : Z. Naturforsch. 39, 854 (1984) 18. Crapon de Caprona, M.-D., Himstedt, W. : Salamandra (submitted) 19. Himstedt, W., Kopp, J., Schmidt, W.: Naturwissenschaften 69, 552 (1982) Genealogical Evidence for Random Mating in a Natural Population of the Great Tit (Parusmajor L.) A.J. van Noordwijk*, P.H. van Tienderen, and G. de Jong Department of Population and Evolutionary Biology, University of Utrecht, NL-3508 TB Utrecht J.H. van Balen Institute for Ecological Research, NL-6666 GA Heteren In every sexually reproducing species the mating structure is very important in deciding the fate of new genes. Yet, our knowledge about mating structures in natural populations is limited to two different levels that are widely apart. On the one hand, there often are obser- )' Present address: Zoologisches Institut der Universit/it Basel, Rheinsprung 9, CH- 4051 Basel vations on the behaviour of individuals and on the other hand global estimates can be derived indirectly from the levels of homozygosity. We report data on an intermediary level, showing that in a great tit population of 50-150 breed- ing individuals mating is at random and that dispersal within the studied area is also random. This follows from the fact that partners, neighbours and randomly chosen birds of opposite 104 Naturwissenschaften 72 (1985) Springer-Verlag 1985