Differentiated sex chromosome pairs in diverse species display certain common characteristics, normally comprising one largely heterochromatic genetically inactive chromosome and one euchromatic genetically active chromosome (e.g. the mammalian Y and X respectively). It is widely accepted that dimorphic sex chromosomes evolved from homologous pairs of autosomes. Although the exact mechanisms through which the pair diverged are not fully understood, an initial suppression of recombination in the sex-determining region is required by all of the major theories. Here we address the question of the mechanism by which this initial suppression of recombination occurs. Our model postulates that the stochastic, de novo accumulation of heterochromatin in the sex determining region can delay pairing of the sex chromosomes in meiosis, resulting in a decrease in recombination. Data to support this model is presented from the cichlid fish, Oreochromis niloticus. Although such a decrease would in most circumstances be evolutionarily disadvantageous, if the region concerned included the major sex determining gene and other gene(s) with sex-specific functions, then this would be selectively advantageous and could trigger the process(es) which, ultimately, lead to the differentiation of the sex chromosomes.

Environmental effects on sex determination in Litopenaeus vannamei were studied by rearing day 1 postlarvae at three temperatures, under three photoperiods, at high density and by starving. None of the environmental conditions affected sex determination or differential development of gender in this species. From day 50, the development of the endopodite of the first pair of pleopods revealed the first external differentiation, showing a triangular structure with three setae in females, whereas a tubular structure remained in males. Juvenile shrimp sex differentiation took place from days 50–90, independent of size, only if postlarvae reached a development threshold of 150 mg of body weight and 20 mm of body length previously. Histology and scanning electron microscopy of the vas deferens revealed that the androgenic gland (AG) is a single 2‐mm cord attached in the subterminal ejaculatory region, just before the distal vas deferens narrows. The AG is composed of large oval cells containing vacuolated cytoplasm, and each cell has a prominent rounded nucleus, similar to all descriptions of the AG in Malacostracans, so we assume that it should have the same function in sex differentiation.

In the Nile tilapia, Oreochromis niloticus, sex determination is primarily genetic, with XX females and XY males. While the X and Y chromosomes (the largest pair) cannot be distinguished in mitotic chromosome spreads, analysis of comparative hybridization of X and Y chromosome derived probes (produced, by microdissection and DOP-PCR, from XX and YY genotypes, respectively) to different genotypes (XX, XY and YY) has demonstrated that sequence differences exist between the sex chromosomes. Here we report the characterization of these probes, showing that a significant proportion of the amplified sequences represent various transposable elements. We further demonstrate that concentrations of a number of these individual elements are found on the sex chromosomes and that the distribution of two such elements differs between the X and Y chromosomes. These findings are discussed in relation to sex chromosome differentiation in O. niloticus and to the changes expected during the early stages of sex chromosome evolution.

Bivalent 1 of the synaptonemal complex (SC) in XY male Oreochromis niloticusshows an unpaired terminal region in early pachytene. This appears to be related to recombination suppression around a sex determination locus. To allow more detailed analysis of this, and unpaired regions in the karyotype of other Oreochromis species, we developed techniques for FISH on SC preparations, combined with DAPI staining. DAPI staining identified presumptive centromeres in SC bivalents, which appeared to correspond to the positions observed in the mitotic karyotype (the kinetochores could only be identified sporadically in silver stained EM SC images). Furthermore, two BAC clones containing Dmo (dmrt4) and OniY227 markers that hybridize to known positions in chromosome pair 1 in mitotic spreads (near the centromere, FLpter 0.25, and the putative sex determination locus, FLpter 0.57, respectively) were used as FISH probes on SCs to verify that the presumptive centromere identified by DAPI staining was located in the expected position. Visualization of both the centromere and FISH signals on bivalent 1 allowed the unpaired region to be positioned at Flpter 0.80 to 1.00, demonstrating that the unpaired region is located in the distal part of the long arm(s). Finally, differences between mitotic and meiotic measurements are discussed. (Foresti et al., 1993; Carrasco et al., 1999; Campos-Ramos et al., 2001;Griffin et al., 2002;Campos-Ramos et al., 2003). The exact relationship between these unpaired regions and sex determining loci in this genus is not clear: for example, a terminal region of the largest bivalent shows delayed pairing in XY O. niloticus but not in XX or YY genotypes (Carrasco et al., 1999), however sex-linked LG1 markers in this species have been mapped by FISH onto a small pair of chromosomes (Lee et al., 2003; Mota-Velasco, unpublished observations; Cnaani et al., 2008). Unpairing in both the largest bivalent and a small bivalent have been observed in WZ O. aureus at pachytene (Campos-Ramos et al., 2001), and linkage studies suggest two unlinked genes affect sex determination in this species, with the dominant one (WZ/ZZ, in LG3) mapping to the largest pair of chromosomes (Lee et al., 2004; Cnaani et al., 2008).The kinetochore (centromere) has been visualised in some TEM synaptonemal complex preparations in O. niloticus (Carrasco et al., 1999), allowing orientation of the unpaired region with respect to the chromosome, but this has not been achieved consistently. In this study, we set out to develop a technique that would allow both identification of the centromere and FISH on pachytene stage chromosomes, using male O. niloticus, with the objective of being able to simultaneously visualise unpaired 4 regions, centromeres and FISH markers at pachytene, to further the study of the role of delayed meiotic pairing in the evolution of sex determination in this genus. 5 Materials and Methods Experimental fishPhenotypic male and female O. niloticus (originating from Lake Manzallah, Egypt) used in this experi...

22Total synaptonemal complex (SC) lengths were estimated from 23Oreochromis aureus Steindachner (which has a WZ/ZZ sex determination 24 system), O. mossambicus Peters and O. niloticus L. (both of which have 25 XX/XY sex determination systems). The total SC length in oocytes was 26 greater than that in spermatocytes in all three species (194±30 µm and 27 134±13 µm, 187±22 µm and 127±17 µm, 193±37 µm and 144±19 µm, 28 respectively). These sex-specific differences did not appear to be influenced 29 by the type of sex determination system (the female/male total SC length 30