CHROMOSOMAL EVOLUTION AND SPECIATION
135
certain dragonflies, in which there appears to be an "unstable" fusion
between the X and an autosome, which become dissociated in some
cells (Oksala, 1943), is peculiar and needs reinvestigation.
Changes of type (2) are less common. They occur in species where
the X is metacentric and where there is at least one pair of metacentric
autosomes. One seems to have taken place at an early stage in the
phylogeny of the mantids and has been transmitted to a large group of
related genera included by taxonomists in several subfamilies (HughesSchrader 1950, 1953). Following White (1941), it has generally been
assumed that all the XiX 2 Y mantids represent a strictly monophyletic
group, the descendants of a single species in which the mechanism first
established itself. However, as more and more genera and subfamilies
(some of them highly divergent in structure) are shown to possess the
XiX 2 Y system, the original hypothesis needs to be reexamined. Since
in the Australian morabine grasshoppers ΧχΧ 2 Υ mechanisms seem to
have evolved on no less than four separate occasions, it may be that
there are groups in which an inherent tendency to such a development
exists, so that ΧχΧ 2 Υ systems are likely to arise in a number of different
evolutionary lineages. However, in view of the fact that almost all of
them have 2n S = 27 (see Table 1) it seems likely that the closely related
genera of the Mantinae sensu stricto are a monophyletic group in the
sense employed above, although some of the aberrant XiX 2 Y genera
(e.g., Choeradodis,
Orihodera, Antemna, and Melliera) may have acquired their sex chromosome mechanism independently. Type (2) transformations may also have occurred independently in two species of fleas
studied by Bayreuther (1954).
In the North American grasshopper genus Paratylotropidia
a change
of type (1) probably occurred first (in the ancestry of P.
beutenmulleri)
and was then followed by a change of type (3), in the phylogeny of P.
morset and P. brunneri (King and Beams, 1938; White, 1953). The
ΧχΧοΥ mechanisms of the nine species of Australian Morabinae mentioned above seem to have arisen by a similar sequence of events (White,
unpublished data). A transformation of type (3) has also taken place
in Orosophila miranda, while one of type (4) has occurred in D. americana americana but not in D. americana texana (Patterson and Stone,
1952). In the rat-kangaroo Fötorous tridactylus,
Sharman and Barber
(1952) have shown that an XYiY 2 mechanism exists in the male, having
presumably arisen through a type (4) transformation. But these authors
believe that the fusion between the X and an autosome was of the
tandem rather than the centric type. Multiple sex chromosome mechanisms are also known in the shrew Sorex araneus (Bovey, 1948) and
in two species of gerbils (Matthey, 1952, 1954b). Although perhaps
135
certain dragonflies, in which there appears to be an "unstable" fusion
between the X and an autosome, which become dissociated in some
cells (Oksala, 1943), is peculiar and needs reinvestigation.
Changes of type (2) are less common. They occur in species where
the X is metacentric and where there is at least one pair of metacentric
autosomes. One seems to have taken place at an early stage in the
phylogeny of the mantids and has been transmitted to a large group of
related genera included by taxonomists in several subfamilies (HughesSchrader 1950, 1953). Following White (1941), it has generally been
assumed that all the XiX 2 Y mantids represent a strictly monophyletic
group, the descendants of a single species in which the mechanism first
established itself. However, as more and more genera and subfamilies
(some of them highly divergent in structure) are shown to possess the
XiX 2 Y system, the original hypothesis needs to be reexamined. Since
in the Australian morabine grasshoppers ΧχΧ 2 Υ mechanisms seem to
have evolved on no less than four separate occasions, it may be that
there are groups in which an inherent tendency to such a development
exists, so that ΧχΧ 2 Υ systems are likely to arise in a number of different
evolutionary lineages. However, in view of the fact that almost all of
them have 2n S = 27 (see Table 1) it seems likely that the closely related
genera of the Mantinae sensu stricto are a monophyletic group in the
sense employed above, although some of the aberrant XiX 2 Y genera
(e.g., Choeradodis,
Orihodera, Antemna, and Melliera) may have acquired their sex chromosome mechanism independently. Type (2) transformations may also have occurred independently in two species of fleas
studied by Bayreuther (1954).
In the North American grasshopper genus Paratylotropidia
a change
of type (1) probably occurred first (in the ancestry of P.
beutenmulleri)
and was then followed by a change of type (3), in the phylogeny of P.
morset and P. brunneri (King and Beams, 1938; White, 1953). The
ΧχΧοΥ mechanisms of the nine species of Australian Morabinae mentioned above seem to have arisen by a similar sequence of events (White,
unpublished data). A transformation of type (3) has also taken place
in Orosophila miranda, while one of type (4) has occurred in D. americana americana but not in D. americana texana (Patterson and Stone,
1952). In the rat-kangaroo Fötorous tridactylus,
Sharman and Barber
(1952) have shown that an XYiY 2 mechanism exists in the male, having
presumably arisen through a type (4) transformation. But these authors
believe that the fusion between the X and an autosome was of the
tandem rather than the centric type. Multiple sex chromosome mechanisms are also known in the shrew Sorex araneus (Bovey, 1948) and
in two species of gerbils (Matthey, 1952, 1954b). Although perhaps