H1

" 22.5 oc

parnassica >22 „c jjvyvvyyyyywL

menalon

>28 °C parnon iimnum Tiffin Hfft

>240C

>240C

FIGURE 9.2 Distribution map and oscillograms of song of the species of the subgenus Parnassiana (genus Platycleis) and of stridulatory movements (in eight species only; synchronous registration of left tegmen movement and sound). Localities of recorded specimen on CD, distribution data from Willemse, F., Catalogue of the Orthoptera of Greece, Fauna Graeciae, Hellenic Zoological Society, Athens, 1984, p. 1, i-xii, 1-275; Willemse, L. and Willemse, F., Entomol. Ber. Amsterdam, 47, 105-107, 1987; Heller, K.-G. and Willemse, F., Entomol. Ber., 49, 144-156, 1989.

10 species (Figure 9.3; Heller, 1988; Heller and Willemse, 1989). In all species the song contains two groups of elements: sequences of very short syllables (microsyllables) which can be positioned at the beginning (Figure 9.2: P. parnassica) or end (Figure 9.2: P. tymphrestos) of a syllable group, and sequences where short and long syllables (macrosyllables) alternate. This pattern is unknown in any other European tettigoniid. Series of microsyllables are produced by many other species related to Platycleis (examples in Heller, 1988; see also Figure 9.1: Metrioptera saussuriana) and they are possibly directed towards other males (Samways, 1976). From a comparison with related species it can be assumed that the alternate sequence with macrosyllables carries species-specific information. However, in all Parnassiana this sequence is quite similar. All syllable types are repeated at about the same rate in all species, so there is no evidence for distinct song differences, although the amount of song variation is still insufficiently known for all species. It is known for example that the duration of the alternate syllable series and the frequency of microsyllable series can be very variable within one species (Heller, 1988), indicating that these parameters are not useful for species recognition, but they seem to differ reliably between other species. The differences in male genitalia, however, are so pronounced that the forms have been described as different species. These data suggest that in this group song does not form the starting point for speciation. Some observations on Mt. Tsoumerka may even provide evidence that two species (P. tenuis and an unidentified one) with quite similar songs but different genitalia, occur on the same location (Heller and Willemse, 1989). The same situation as

Central Greece Peloponnisos

FIGURE 9.3 Male genitalia of the species of the subgenus Parnassiana (genus Platycleis). First row: left cercus, dorsal view. Second row: titillator, anterior view. Third row: titillator, lateral view. (Figures from Willemse, F., Entomol. Ber. Amsterdam, 40, 103-112, 1980; Willemse, L. and Willemse, F., Entomol. Ber. Amsterdam, 47, 105-107, 1987; Heller, K.-G. and Willemse, F., Entomol. Ber., 49, 144-156, 1989.) Bottom row: distribution.

Central Greece Peloponnisos

FIGURE 9.3 Male genitalia of the species of the subgenus Parnassiana (genus Platycleis). First row: left cercus, dorsal view. Second row: titillator, anterior view. Third row: titillator, lateral view. (Figures from Willemse, F., Entomol. Ber. Amsterdam, 40, 103-112, 1980; Willemse, L. and Willemse, F., Entomol. Ber. Amsterdam, 47, 105-107, 1987; Heller, K.-G. and Willemse, F., Entomol. Ber., 49, 144-156, 1989.) Bottom row: distribution.

in Parnassiana occurs in some other groups where songs are similar but genitalia are different in allopatric populations.

Allopatric Forms Differing in Genitalic Morphology, but not in Song

Many published examples describe allopatric forms which differ in genitalic morphology, but not in song. Probably the most striking example is found among the Greek species of Eupholidoptera. Among the about 21 completely allo- or parapatric species, the song pattern of all those studied (14) is the same, with the possible exception of the east Aegean E. prasina, which may have a faster syllable rate (Heller, 1988; Willemse and Heller, 2001). In male genitalia, however, very large differences between most species are observed (Willemse, 1985; Willemse and Heller, 2001).

Psorodonotus and the subgenus Tessellana of Platycleis contain distinctly less subspecies or species, respectively, than Eupholidoptera in Europe, but present otherwise the same picture. Within each genus the three subspecies of Psorodonotus and four species of Tessellana have nearly identical songs (Heller, 1988), but differ in morphology of genitalia (Harz, 1969). Other examples are Acrometopa macropoda, Platycleis albopunctata, Zeuneriana, Pholidoptera, Pterolepis spoliata, Callimenus and Uromenus (Heller, 1988).

Ragge and Reynolds (1998) discussed two species complexes within the genus Metrioptera — the M. saussuriana and M. roeselii complexes. The widespread M. saussuriana and the local M. buysonii, which do not differ in song, show small but stable differences in genitalia. They occur close together in the Pyrenees (maps in Voisin, 2003). The Italian M. caprai shows only small differences in genitalia compared with M. saussuriana. The song of some populations of this species is very similar to that of M. saussuriana, but others have a higher syllable number per syllable group. In the M. roeselii complex several local allied forms have been described in southern Europe. Some show relatively large differences in genitalia, but the song patterns of all of them are the same (Ragge and Reynolds, 1998; Fontana, 2001).

In the ampliatus group of Poecilimon, songs of the two parapatric species, P. ampliatus and P. ebneri, are quite similar, while the male cerci show distinct differences. P. ampliatus differs further from P. ebneri in the possession of a large dorsal gland which the female feeds on during mating (Heller and Lehmann, 2004).

A more complicated situation is found in Ephippiger. There has been a long discussion about the taxonomic status of species related to Ephippiger ephippiger (Kidd and Ritchie, 2000). Obviously local forms do exist, but their status is unclear. They differ mainly in male genitalia (Fontana, Kidd in preparation; Kidd and Ritchie, 2000) and song with high variability concerning syllable numbers per syllable group. In the western part of the range of E. ephippiger three mainly allopatric archetypes (Kidd and Ritchie, 2000) may be found, but they hybridise where they meet. Obviously neither differences in song nor in genitalia are large enough to prevent mixing of the gene pools. Over the whole range of the species, the pattern of song variation and its relationship to morphological variation is not sufficiently known.

Another well-known example from North America is the genus Orchelimum. Two parapatric species, O. nigripes and O. pulchellum, differ slightly in genitalia, but not in song (Walker, 1974). Hybridisation between these forms has been documented in two contact zones (Shapiro, 1998). It should be noted that Orchelimum species are long-winged animals capable of relatively fast dispersal while all other examples mentioned above concern short-winged animals.

Belocephalus contains five species falling into two groups (see Walker and Moore, 2004). Members of both groups may occur sympatrically. The three are morphologically quite similar; allopatric species of the B. sabalis group all have the same song, but one species differs clearly in life-history traits, while the other two are geographically separated. The other two allopatric species of the B. subapterus group are morphologically similar, but differ mainly in song and belong to the species treated in the following subsection. However, within B. subapterus there are two geographically separated forms (with hybrid zone) differing in male genitalia only.

Allopatric Forms Differing in Song, but not in Genitalic Morphology

In a few examples among allopatric forms, only the song seems to have changed. Species will be mentioned here where only the song and the stridulatory organs together have changed, or at least the song much more than genitalia.

The most distinctive examples are found in Decticus verrucivorus. This well-known and widespread species shows some variability in wing length and body size (Samways and Harz, 1982). In the song, which has been recorded at many localities in Europe (Ragge and Reynolds, 1998), no significant differences have been found, except in Spain. Specimens from the Iberian Peninsula produced a quite distinct song, which was used to establish a subspecies, Decticus verrucivorus assiduus (Ingrisch et al., 1992). No differences were found in genitalia. Recently, in another form of this group from south Italy, a song pattern was detected which also differs slightly from that of the nominate subspecies (Fontana et al., 1999).

Two other examples of song change with no, or only minor, divergence in genitalic morphology are found in Poecilimon. Heller (1984) described the subspecies P. obesus artedentatus, now usually considered as a species (Willemse and Heller, 1992), differentiated clearly in song and morphology of the stridulatory file. This is a possible example of reproductive character displacement because P. artedentatus occurs in an area where another species of Poecilimon, P. nobilis, is found with a song very similar to that of P. obesus. The second example concerns Poecilimon paros. This island species was discovered by its song, which differs distinctly from that of its allopatric sister, P. hamatus (Heller and Reinhold, 1992). A small difference in the shape of the male cerci has now been found, which can be used for morphological separation of both species. A similar discovery was made recently in Turkey, where the new species P. martinae differs from its sister species, P. inflatus, mainly in song, but also in morphology of the male cerci (Heller, 2004).

In east Asia, two species of Hexacentrus, H. japonicus and H. unicolour, are very similar in morphology, but differ in song (Inagaki et al., 1986). They are largely allopatric, but overlap in a small area in Japan where they prefer separate habitats (Inagaki et al., 1986). Morphologically they can be differentiated by small differences in male genitalia (Inagaki and Sugimoto, 1994) and the structure of the stridulatory file (Inagaki et al., 1990). In the laboratory interspecific hybrids can be obtained (F1 and F2) (Inagaki and Sugimoto, 1994).

At this point those rare examples should be mentioned where not only allo- but also sympatric species differ only by song. Among bushcrickets only some species of North American Amblycorypha belong to this category (Walker et al., 2003). Walker (1964) expected that such cases of cryptic species would be relatively common, but he also predicted that, after discovery of a cryptic species by its song, morphological differences would later be found (see Poecilimon example above). This may also apply, for example, to the different song forms of Phyllomimus inversus in Malaysia, for which the colour of the "knees" of the hind legs may be diagnostic (Heller, 1995).

Unusually Variable Species

Besides species in which allopatric forms differ mainly in genitalia or song, there are some examples where allopatric forms differ in morphology and song but are still considered members of one species.

Allo- and parapatric populations of several species of Pterophylla have been analysed in North America. In P. camellifolia, Shaw and Carlson (1969) made one of the first studies of this type, demonstrating song differences together with differences in male genitalia. Because of presumed hybridisation, they considered both parapatric forms as members of one species. Den Hollander and Barrientos (1994) found acoustic and morphometric differences between allopatric populations of P. beltrani in Mexico. Morphometric differences (most importantly in male cerci) seem to have diverged more rapidly than song. In hybridisation experiments these authors demonstrated that there are neither strong premating nor postmating reproductive barriers. They compared P. beltrani also to the allopatric P. robertsi (Barrientos and Den Hollander, 1994). As expected for different species, they found differences in morphology and song. They found some overlap in the measured song characters between both species, but not in morphological characters. Hybridisation experiments indicated strong postmating barriers (successful matings and egg deposition, but no hatching of offspring), although in a second series of tests some offspring were obtained (Barrientos-Lozano, 1998).

Was this article helpful?

0 0

Post a comment