DE SYLVA: MUGILOIDEI 



533 



Fig. 282. Postflexion larvae (A, B) and juvenile (C) of silver mullet. 

 Mugil curema. (A) 7.0 mm; (B) 14.5 mm; (C) 25.5 mm. From Anderson 

 (1957). 



"querimana stage." An analysis of the genera and species pos- 

 sessing this trait has not been undertaken. Biochemical studies 

 on mugilid systematics have been undertaken by Callegarini 

 and Basaglia (1978) and by Autem and Bonhomme (1980) in 

 the Mediterranean, but no studies have been carried out on a 

 worldwide basis. 



As stated in the discussions on Sphyraenoidei and Polyne- 

 moidei (this volume), they have been closely linked with the 

 Mugiloidei phyletically. Previously, the athennids had been 

 placed within this assemblage, but Rosen ( 1 964) has clearly 

 shown that the atherinids belong in a separate superorder con- 

 taining the flyingfishes and livebearers. The Mugiloidei appear 

 more closely related osteologically to the Sphyraenoidei than 

 they are to the Polynemoidei. 



A brief history of the higher classification of these groups is 

 reviewed here. The suborder Percesoces had included the Ath- 

 erinidae. Mugilidae, and Sphyraenidae (Jordan and Evermann, 

 1868-1900), but Starks (1900) questioned their similanty, though 

 he believed them to be quite close based upon the decided 

 branching of the epiotic crests. Superficially, the mugiloid-sphy- 

 raenoid skeleton resembles that of atherinoids, but Hollister 

 (1937) pointed out an important developmental difference be- 

 tween them. In Athehna. the lowermost hypural plate develops 

 as a single entity. In Mugil and Sphyraena this plate forms from 

 two distinct elements. Berg ( 1 940) separated the Mugilidae, with 

 the Sphyraenidae and the Atherinidae, from the Perciformes as 

 the order Mugiliformes because they have abdominal pelvic fins, 

 a relatively primitive character. Rosen (1964) also pointed out 

 the similarities among mugiloids, sphyraenoids, and polyne- 

 moids in ossification of the skull, especially the common pres- 

 ence of a subocular shelf, the jaw suspension and feeding mech- 

 anism, jaw musculature, and the pharyngobranchial and 

 opercular apparatuses. Further, Rosen stated that "the embryos 

 of mullet (Anderson, 1957) and barracuda (Orton, 1955b) are 

 small and contain a large oil globule .... A forward-displaced 

 heart is also characteristic of Oryzias . . . but not of Sphyraena 

 (Orton, 1955b;Shojimaetal., 1957), and probably not of A/i/.g//." 



Removal of the Mugilidae from the suborder Percoidei is 

 supported by studies of blood plasma and plasma proteins (Sul- 

 ya et al., 1960). Plasma proteins of mugilids are less complex 

 than those of any other family considered to be Perciformes, 

 and show relationships to some species of Cypriniformes and 

 Clupeiformes (Gunter et al.. 1961). In contrast, plasma proteins 

 of some species of 3 perciform families, the Carangidae, Sciaeni- 

 dae, and Scombridae, do not differ greatly from those of the 

 Mugilidae. Based on this, the Mugilidae could be regarded either 

 as belonging to the most primitive perciform group or as branch- 

 ing from some early perciform. 



The early life history stages do not appear to offer useful hints 

 as to phyletic relations with other taxa, except that the Mugi- 

 loidei have 23 myotomes dunng larval development, a feature 

 shared with the Polynemoidei and Sphyraenoidei. 



RosENSTiEL School of Marine and Atmospheric Science, 

 University of Miami, 4600 Rickenbacker Causeway, 

 Miami, Florida 33149. 



