The karyotype of Actitis macularius (AMA)
The karyotype of AMA, which was studied herein for the first time, has 2n = 92, and thus exhibits a 2n greater than the proposed 2n=80 for the PAK . This difference reflects the occurrence of fission rearrangements in all macrochromosomes. As this is a common feature among Scolopaci (Table 10, this is not a distinctive feature specific of Actitis macularius.
Chromosomal rearrangements among Actitis macularius (AMA), Burhinus oedicnemus (BOE) and Gallus gallus (GGA)
Using BOE probes to paint the karyotype of a species of Scolopacidae allowed us to detect the rearrangements that occurred in the phylogenetic branch leading to the AMA karyotype. Unlike the conserved state observed for the first pairs of many avian species [16, 19, 22, 23], including Burhinus oedicnemus , pairs 1, 2, 3, 4 and 6 of PAK are fissioned in AMA. Possibly other Scolopacidae with high 2n and similar chromosomes may have undergone the same rearrangements.
Many BOE probes hybridized on the long arm of the W in AMA, as observed in Larus argentatus . This suggests that the W carries numerous variable copies (repetitive regions) homologous to the autosomes of species in order Charadriiformes. A similar arrangement was found in the Passeriform, Glyphorynchus spirurus . An experiment to test the possibility of repetitive DNA sequences spread in autosomes and W would be the isolation of this sequence and its use as a DNA probe for FISH in AMA karyotype.
Despite BOE belonging to the same order as AMA, we observed no conservation of the macrochromosome pairs, except for the Z and W. The fission of submetacentric BOE1 is clear in the first two AMA acrocentric pairs. Pericentric inversion may have occurred after this fission, leading to the formation of two pairs with small short arms (Figure 4A). Alternatively, the short arm may be the result of telomeric amplification  or centromeric repositioning . The submetacentric BOE2 is divided into four telocentric pairs in AMA (pairs 3 and 11-13) due to multiple fissions; we were not able to define which segment of BOE2 was found in each AMA pair (Figure 4B). BOE3 experienced fission, giving rise to pairs AMA4, 14 and 15 (Figure 4C). Fission of BOE4 gave rise to AMA6 and AMA16 (Figure 4D). BOE5 was divided into two pairs, AMA7 and AMA8, but this was not by fission. Cytogenetic studies demonstrated that the fusion of PAK7 and PAK8 is a characteristic of suborder Charadrii ; the presence of the separate chromosomal pairs in Actitis is the ancestral form, and BOE5 is the derived form (Figure 4E). Fission of BOE6 gave rise to AMA9 and 10 (Figure 4F). BOE8 did not hybridize in the AMA genome, possibly for technical reasons. So, this is first confirmation of the fissions in pairs BOE2, 3, 4 and 6 and also the first demonstration of fission in pair BOE1 in Scolopaci.
Fissions in BOE1, 2 and 5, were observed in Glyphorynchus spirurus , Strigiformes, Passeriformes, Columbiformes and Falconiformes also have the fission of GGA1 (BOE1) [20, 27-29]. For Scolopaci, in contrast, a fission of PAK1 (= GGA1, BOE1) seems to be a character shared only among members of the Scolopacidae family. Its presence in other orders would therefore be an example of homoplasy.
The correspondences among the AMA, BOE, LAR, GGA and PAK chromosomes are shown in Table 2.
Chromosome evolution in suborder Scolopaci
It is accepted that the ancestral putative karyotype (PAK) with 2n = 80, which is commonly found in several orders of birds, remained conserved for about 100 million years, with few variations for Neoaves . However, order Charadriiformes presents a high level of karyotypic diversity . An interesting point is that the three suborders originated in the late Cretaceous between 79 and 102 Mya , which indicates that little time has passed from the origin of PAK to the ancestral Charadriiformes karyotype.
Suborder Scolopaci has a high diploid number, ranging from 78 to 98 chromosomes . In addition to Actitis macularius (described here), chromosome painting was previously used to examine the karyotype of Jacana jacana . Our comparative analysis between these two species (Table 2) shows that both share the following fissions: PAK2 (GGA2, BOE2) in four segments; PAK3 (GGA3, BOE3) in three segments; and PAK4 (GGA4q, BOE4) and PAK6 (GGA6, BOE9) in two segments. After these fissions, a series of fusions occurred between several chromosome pairs in the evolutionary branch that gave rise to JJA . Although the literature lacks any additional chromosome painting study of the Jacanidae, the karyotype described from the other genus of this family, Hydrophasianus, has the same diploid number and appears similar to JJA on conventional staining . This suggests that the chromosomal characteristics found in JJA are not restricted to this species and may be chromosome signatures for the Jacanidae family (Figure 5).
An interesting feature is seen for chromosome PAK1 (= BOE1, GGA1): It is split into two pairs in AMA but remains whole in JJA. This suggests that PAK1 underwent fission in the branch that led to AMA but remained in its ancestral form in JJA. Since the morphology of the full-length PAK1 chromosome is quite different from that of its split version, information about the timing of this fission can be obtained by analyzing other karyotypes along the branch that leads to AMA (family Scolopacidae), even using conventional staining data. The karyotypes of genera Tringa, Calidris, Arenaria and Limosa [7, 9-11, 13, 14] show the first chromosomal pair as an acrocentric that is similar in size to the long arm of PAK1, according to the measurements performed by Hammar . This suggests that fission occurred in the branches that lead to these genera (Figure 5). Two branches cast some doubt on this proposition, however. The Gallinago and Scolopax genera have similar karyotypes, in which the first pair is a submetacentric chromosome [7, 10, 13]. Hammar  measured the chromosomes of several species of Charadriiformes and demonstrated that the first whole-chromosome pair (PAK1) is equivalent to 14% of the karyotype. In contrast, that of Tringa (long arm of PAK1) corresponds to 10%. The first pair of the Gallinago karyotype corresponds to 9.3% of the karyotype, suggesting that it is similar to Tringa, but with an inversion. Chromosomal painting studies are needed in these species to confirm if the first pair of Tringa and Gallinago are homeologues. The second branch that generates doubt is the one that leads to Numenius and Bartramia (Figure 5). The second chromosomal pair of the karyotype of Numenius arquata is an acrocentric corresponding to 9.8% of the karyotype ; it may be equivalent to the first pair of the other species of Scolopaci (the first pair of the karyotype of Numenius is a metacentric of similar size, possibly the result of a fusion). Studies with chromosome painting in Numenius and/or Bartramia are needed to test this possible correspondence. Thus, it is not clear whether the fission break in PAK1 occurred at the base of the branch that gave rise to the Scolopacidae family or after the separation of the branch that gave rise to Numenius and Bartramia (Figure 5). If additional studies confirm that Numenius arquata pair 2 is equivalent to the first chromosome of the other species of Scolopacidae, the fission of PAK1 would be a chromosomal signature for this family.
The rearrangements described here are restricted to suborder Scolopaci, since chromosomal painting in Larus argentatus , a species of suborder Lari (a sister group of Scolopaci) [1, 4], revealed a karyotype similar to the ancestral birds in pairs PAK1-4, with fusions of microchromosomes with macrochromosomes (LAR4 and 7-9) and none of the fissions observed in Scolopaci (Table 2).
The data analyzed here allow us to propose an ancestral karyotype for suborder Scolopaci using PAK as an outgroup, in which chromosome PAK1 is preserved, PAK2 is broken into four pairs, PAK3 is fragmented into three segments and PAK4 and PAK6 are divided into two segments each (Scolopaci Putative Ancestral Karyotype, SPAK, Figure 5).