Experientia 36 (1980), Birkh~iuserVerlag, Basel (Schweiz) Cy Sb F2 flies from different sibships were intercrossed leading to four classes of offspring with phenotypes +, Cy, Sb and Cy Sb. In each class brothers and sisters were then mated. This scheme allowed crossing between Is-/Is- sibs in the Cy class only if the Is gene was located on the third chromosome, or in the Sb class only if the gene was on the 2nd chromosome. However, in the + class all sibs were Is-/ls-. As the hatchability of inbred eggs of the respective marker classes indicates (0.36___0.20 for + ; 0.54 + 0.11 for Cy; 0.92+0.07 for Sb sibs), inbreeding depression was associated with the 3rd chromosome. I believe this observation reflects the action of a mutator 6'7 or controlling element s'9. Such elements which can change the expression of structural genes l~ perhaps by interacting with a regulator~ system 11, are transposal~le to different loci in the genome 9'1~ TranspositionS0 or chromosomal contam- ination12-14 might account for normal hatchability of Is-/ls + eggs bearing an induced chromosome as is the case when 1 parent possesses an Is- allele 2. According to this mutator hypothesis 6, disruption of genetic suppression of mutator activity through hybridisation between populations or shifts from inbreeding to outbreeding can lead to an increase in mutation frequency. These mutations or chromosomal ab- normalities are then revealed by inbreeding. Since inbreeding depression appears to be directly associat- ed with mutator activity and related phenomena x5'16,inbred matings may play a major role in the course of evolutionary processes. My hypothesis differs from the classical view of inbreeding depression, which postulates concealed dele- 171 terious genes 17. Of course both mechanisms could be opera- five. 1 I thank R. Grantham and J.M. Legay (Lyon, France), C. Krimbas (Athens, Greece), D. Mohler (USA) and especially R. Milkman (USA) for helpful criticism and suggestions. The Laboratoire de Biomrtrie is associated with CNRS Laboratory No. 243. 2 C. Birmont, Mech. Ageing Develop. 8, 21 (1978). 3 C. Birmont and J. Bouletreau-Merle, Experientia 34, 273 (1978). 4 C. Birmont and C. Lemaitre, C. r. Acad. Sci., D 286, 1715 (1978). 5 J. David, Bull. Biol. Fr. Belg. 93, 472 (1959). 6 J.N. Thompson and R. C. Woodruff, Nature 274, 317 (1978). 7 M.M. Green, Mutable and mutator loci, in: The Genetics and Biology of Drosophila, p. 929. Academic Press, London 1976. 8 B. McClintock, Cold Spring Harb. Syrup. quant. Biol. 16, 13 (1951). 9 M.G. Kidwell, J.F. Kidwell and M. Nei, Genetics 75, 133 (1973). 10 P. Nevers and H. Saedler, Nature 268, 109(1977). 11 R.C. Ullrich, Genetics 88, 709 (1978). 12 E. Slatko, Genetics 90, 105 (1978). 13 G. Picard, Molec. gen. Genet. 164, 235 (1978). 14 A. Pelisson, Genet. Res., Camb. 32, 113 (1978). 15 G. Picard, Genetics 83, 107 (1976). 16 A. Bucheton and G. Picard, Heredity 40, 207 (1978). 17 R,C. Lewontin, The Genetic Basis of Evolutionary Change. Columbia University Press, New York and London 1974. The karyotype of Typhloneetes compressicauda (Amphibia: Gymnophiona) with comments on chromosome evolution in caecilians 1 M.H. Wake, J.C. Hafner, M. S. Hafner, L.L. Klosterman and J. L. Patton Museum of Vertebrate Zoology and Department of Zoology, University of California, Berkeley (CA 94720, USA), 23 January 1979 Summary. Typhlonectes compressicauda has a diploid number of 28. Its karyotype, when compared to that of other caecilians, suggests some discordance in the hypothesized model of chromosome reduction in the evolution of amphibian lineages. Karyotypic information is now available for 15 species representin8 3 of the 5 currently recognized families of caecilians2-L The only member of the aquatic New World family Typhlonectidae karyotyped thus far is Chthonerpe- ton indistinctum 5. This species, while having a number of derived morphological and physiological features that are associated with its aquatic habitus, also has the lowest diploid number reported for caecilians (2n= 20) and it lacks microchromosomes. Based on this and similar lines of evidence, it has been suggested that karyotypes provide evidence in support of the hypothesis that the general pattern of amphibian chromosome evolution is one of reduction in chromosome number (with loss of microchro- mosomes) 6,7. Moreover, this reductional trend in chromo- some evolution may be correlated with derived states in other features of amphibian biology8-11. In this report we describe the karyotype of another member of the family Typhlonectidae, Typhlonectes compressicauda, and recon- sider the 'reduction' hypothesis of chromosomal evolution in light of new evidence presented herein. Material and methods. 3 individuals (2 females and 1 male) of Typhlonectes compressicauda from Cienga Santo Tom~ts, Departamento Atlantico, Colombia, were karyotyped. Spe- cimens and karyotypic preparations will be deposited in the Museum of Vertebrate Zoology, University of California, Berkeley. Animals were injected i.p. with a 0.05% colchi- cine solution 6 h prior to sacrifice. The best preparations were obtained from an animal that had been injected with 0.5 ml of warm yeast suspension both 48 and 24 h prior to colchicine injection 12. Air-dried slides were prepared using gut epithelium and spleen according to the method of Patton 13 except incubation in hypotonic solution was for 1 h and centrifugation was at 700 rpm. Metaphase spreads in which most or all of the chromosomes were not overlap- ping were used to determine the diploid number. 17 chromosomal spreads were analyzed. Results. Typhlonectes compressicauda has a chromosomal complement consisting of 28 biarmed elements (figure): The karyotype contains 3 groups of chromosomes: metacentrics (2 large pairs, 3 medium to small pairs); submetacentrics (5 medinm-sized pairs); and subtelocen- trics (4 small pairs). Discussion. A comparison of the nonpreferentially stained karyotypes of the 2 species of typhlonectid caecilians,