Ho. Phytogeny and biogeography of Merluccius 



101 



Figure 8 



Dispersal routes of hakes infeiTed from the area summary cladogram 

 in Figure 7. 



population living on both sides of the southern part of 

 South America south of 40° S. Interestingly, both 

 populations, in spite of being distantly separated by the 

 Pacific, carry the same species of copepod parasite— 

 Neobrachiella insidiosa lageniformes— on their gills 

 (Kabata and Ho 1981), a clear indication that the New 

 Zealand population must have originated from the 

 hakes in South American waters. 



Range expansion and vicariance of species C 



According to the proposed history of the relative 

 motion of the plates in the Gulf of Mexico/Caribbean 

 region, Pindell and Dewey (1982) concluded that the 

 large-scale eastward migration of the Caribbean plate 

 started in the Oligocene (about 36 MA) and eventually 

 placed the Greater Antilles in their present positions. 

 It is assumed that species C, as in species B, expanded 

 southward prior to the initiation of this Caribbean plate 

 movement. However, being a northern population, it 

 did not reach as far south into the South Atlantic as 

 Species B. Instead, it expanded into the Pacific when 

 the Panamanian seaway was opened in the Oligocene. 

 As with the hake in the eastern Atlantic, species C 

 was prevented from inhabiting the waters of low lati- 

 tudes when the climate warmed in the Early Miocene. 

 By this restriction, species C was divided into an Atlan- 

 tic population, which eventually gave rise to silver hake 

 M. bilinearis and a Pacific population (species G, Fig. 7) 

 which became the ancestor of the three eastern Pacific 

 species. The fossil remains of the Pacific hake M. pro- 

 dudus are common in the Pliocene deposit of Califor- 

 nia (Fitch 1969, Fitch and Reimer 1967, Zinsmeister 

 1970), indicating that the divergence of species G into 

 species H and Pacific hake (M. prodiietus or its imme- 

 diate ancestor) took place either in Miocene or Pliocene. 



Species H was very likely confined to the North 

 Pacific off the coast of Mexico until the Pliocene when 

 the Panamanian isthmus was reestablished at about 3 

 MA (Haq 1984). Panamanian hake M. angustimanus 

 is found from Baja California to Colombia, but not in 

 the Caribbean; therefore, species H must have moved 

 southward after closing of the Panamanian seaways. 

 The vicariance event responsible for the separation of 

 species H into Panamanian hake and Chilean hake is 

 unclear, but the influence of the Ice Ages during the 

 late Pliocene and Pleistocene may have played a role. 



Conclusion 



The phylogenetic hypothesis of hake (Fig. 3) presented 

 here is the most parsimonious scheme derived from the 

 cladistic analysis of the osteological data of Inada 

 (1981). It is congiiient with the scheme of evolution pro- 

 posed by Ho (1974) for the hake-specific copepod 

 parasites. This is a testable model; it can be corrobor- 

 ated by including more anatomical data (e.g., muscu- 

 lature), ontogeny, karyology, DNA sequences, and 

 allozymes. Parasitological information from proto- 

 zoans, helminths, and crustaceans can also provide fur- 

 ther corroboration, particularly when hake-specific 

 parasites are found among them. 



The vicariance model proposed for the hake biogeog- 

 raphy is based on the adopted most-parsimonious tree 

 that shows congruence with the available parasito- 

 logical information. The inferred pattern of biogeog- 

 i-aphy, particularly the track across the Atlantic in the 

 low latitudes (Fig. 8), is yet to be reported for demer- 

 sal fishes. However, it should be mentioned that Van 

 der Spoel and Heyman (1983) considered that the 

 planktonic faunas of the Panama Passage and South 

 American inland sea as ancestral to all eastern Atlan- 

 tic distant neritic taxa, and proposed a similar eastward 

 dispersal. 



The ranges of ancestral hake waxed and waned with 

 the fall and rise of paleoceanographic temperatures in 

 the Tertiary, dispersed during the cooling periods and 

 fragmented by the development of warming trends. 

 Development of major lowstands in sea level is also 

 viewed as an effective vicariant event. This model can 

 be tested with additional details of the plate-tectonic 

 models and paleoceanographic-climatic history. Works 

 on other marine life with similar distributions would 

 test the validity of the three inferred general tracks. 



Both the phylogenetic hypothesis and biogeographic 

 model differ from current views, but they are viewed 

 as the better explanation of available data. They are 

 presented here as a working model subject to modifica- 

 tion as more exact infoi-mation becomes available. 



